From 26a67b2b872b894d9336e7f8a8ef6f9f67462a8b Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 17:10:20 -0600 Subject: [PATCH 01/26] update intro --- articles/openai-cookbook-llms-101.md | 182 +++++++++++++++++++++++++++ 1 file changed, 182 insertions(+) create mode 100644 articles/openai-cookbook-llms-101.md diff --git a/articles/openai-cookbook-llms-101.md b/articles/openai-cookbook-llms-101.md new file mode 100644 index 0000000000..77292198f6 --- /dev/null +++ b/articles/openai-cookbook-llms-101.md @@ -0,0 +1,182 @@ +--- +title: "LLMs 101: A Practical Introduction" +description: "A hands-on, code-first introduction to large language models for Cookbook readers." +last_updated: "2025-08-24" +--- + +# LLMs 101: A Practical Introduction + +> **Who this is for.** Developers who want a fast, working understanding of large language models and the knobs that matter in real apps. + +## At a glance + +``` +Text prompt + ↓ (tokenization) +Tokens → Embeddings → [Transformer layers × N] → Next‑token probabilities + ↓ ↓ +Detokenization Sampling (temperature/top_p) → Output text +``` + +- **LLMs** are neural networks (usually **transformers**) trained on lots of text to predict the next token. +- **Tokenization** splits text into subword units; **embeddings** map tokens to vectors; transformer layers build context‑aware representations. +- Generation repeats next‑token sampling until a stop condition (length or stop sequences) is met. + +--- + +## Quick start: generate text + +### Python + +```python +from openai import OpenAI + +client = OpenAI() +resp = client.responses.create( + model="gpt-4o", + instructions="You are a concise technical explainer.", + input="In one paragraph, explain what a token is in an LLM." +) +print(resp.output_text) +``` + +### JavaScript / TypeScript + +```js +import OpenAI from "openai"; +const client = new OpenAI(); + +const resp = await client.responses.create({ + model: "gpt-4o", + instructions: "You are a concise technical explainer.", + input: "In one paragraph, explain what a token is in an LLM." +}); +console.log(resp.output_text); +``` + +> **Tip.** Model names evolve; check your Models list before shipping. Prefer streaming for chat‑like UIs (see below). + +--- + +## What can LLMs do? + +Despite the name, LLMs can be **multi‑modal** when models and inputs support it (text, code, sometimes images/audio). Core text tasks: + +- **Generate**: draft, rewrite, continue, or brainstorm. +- **Transform**: translate, rephrase, format, classify, extract. +- **Analyze**: summarize, compare, tag, or answer questions. +- **Tool use / agents**: call functions or APIs as part of a loop to act. + +These patterns compose into search, assistants, form‑fillers, data extraction, QA, and more. + +--- + +## How LLMs work (just enough to be dangerous) + +1. **Tokenization.** Input text → tokens (IDs). Whitespace and punctuation matter—“token‑budget math” is a real constraint. +2. **Embeddings.** Each token ID becomes a vector; positions are encoded so order matters. +3. **Transformer layers.** Self‑attention mixes information across positions so each token’s representation becomes **contextual** (richer than the raw embedding). +4. **Decoding.** The model outputs a probability distribution over the next token. +5. **Sampling.** Choose how “adventurous” generation is (see knobs below), append the token, and repeat until done. + +--- + +## The knobs you’ll touch most + +- **Temperature** *(0.0–2.0)* — Lower → more deterministic/boring; higher → more diverse/creative. +- **Top‑p (nucleus)** *(0–1)* — Sample only from the smallest set of tokens whose cumulative probability ≤ *p*. +- **Max output tokens** — Hard limit on output length; controls latency and cost. +- **System / instructions** — Up‑front role, constraints, and style to steer behavior. +- **Stop sequences** — Cleanly cut off output at known boundaries. +- **Streaming** — Receive tokens as they’re generated; improves perceived latency. + +**Practical defaults:** `temperature=0.2–0.7`, `top_p=1.0`, set a **max output** that fits your UI, and **stream** by default for chat UX. + +--- + +## Make context do the heavy lifting + +- **Context window.** Inputs + outputs share a finite token budget; plan prompts and retrieval to fit. +- **Ground with your data (RAG).** Retrieve relevant snippets and include them in the prompt to improve factuality. +- **Structured outputs.** Ask for JSON (and validate) when you need machine‑readable results. +- **Few‑shot examples.** Provide 1–3 compact exemplars to stabilize format and tone. + +--- + +## Minimal streaming example + +### Python + +```python +from openai import OpenAI +client = OpenAI() + +with client.responses.stream( + model="gpt-4o", + input="Stream a two-sentence explanation of context windows." +) as stream: + for event in stream: + if event.type == "response.output_text.delta": + print(event.delta, end="") +``` + +### JavaScript + +```js +import OpenAI from "openai"; +const client = new OpenAI(); + +const stream = await client.responses.stream({ + model: "gpt-4o", + input: "Stream a two-sentence explanation of context windows." +}); + +for await (const event of stream) { + if (event.type === "response.output_text.delta") { + process.stdout.write(event.delta); + } +} +``` + +--- + +## Limitations (design around these) + +- **Hallucinations.** Models can generate plausible but false statements. Ground with citations/RAG; validate critical outputs. +- **Recency.** Models don’t inherently know the latest facts; retrieve or provide current data. +- **Ambiguity.** Vague prompts → vague answers; specify domain, audience, length, and format. +- **Determinism.** Even at `temperature=0`, responses may vary across runs/envs. Don’t promise bit‑for‑bit reproducibility. +- **Cost & latency.** Longer prompts and bigger models are slower and costlier; iterate toward the smallest model that meets quality. + +--- + +## Common gotchas + +- **Characters ≠ tokens.** Budget both input and output to avoid truncation. +- **Over‑prompting.** Prefer simple, testable instructions; add examples sparingly. +- **Leaky formats.** If you need JSON, enforce it (schema + validators) and add a repair step. +- **One prompt for everything.** Separate prompts per task/endpoint; keep them versioned and testable. +- **Skipping evaluation.** Keep a tiny dataset of real tasks; score changes whenever you tweak prompts, models, or retrieval. + +--- + +## Glossary + +- **Token** — Small unit of text (≈ subword) used by models. +- **Embedding** — Vector representation of a token or text span. +- **Context window** — Max tokens the model can attend to at once (prompt + output). +- **Temperature / top‑p** — Randomness controls during sampling. +- **System / instructions** — Up‑front guidance that shapes responses. +- **RAG** — Retrieval‑Augmented Generation; retrieve data and include it in the prompt. + +--- + +## Where to go next + +- Prompt patterns for **structured outputs** +- **Retrieval‑augmented generation (RAG)** basics +- **Evaluating** LLM quality (offline + online) +- **Streaming UX** patterns and backpressure handling +- **Safety** and policy‑aware prompting + +> Adapted from a shorter draft and expanded with code-first guidance. From bf278df6cd5246b0ee1499469dadcd4645f2fb2c Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 17:50:07 -0600 Subject: [PATCH 02/26] Add Payton Ison to authors.yaml --- authors.yaml | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/authors.yaml b/authors.yaml index 466f991af4..de1ec4eaf9 100644 --- a/authors.yaml +++ b/authors.yaml @@ -477,3 +477,8 @@ heejingithub: name: "Heejin Cho" website: "https://www.linkedin.com/in/heejc/" avatar: "https://avatars.githubusercontent.com/u/169293861" + +paytonison: + name: "Payton Ison" + website: "https://linkedin.com/in/paytonison" + avatar: "https://avatars.githubusercontent.com/u/148833579" From 2161cbdaf7cda74f806e56ed13941788d35e962c Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 17:52:45 -0600 Subject: [PATCH 03/26] Add new article entry for LLMs 101 --- registry.yaml | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/registry.yaml b/registry.yaml index dde9894426..d7f9e954c8 100644 --- a/registry.yaml +++ b/registry.yaml @@ -2509,3 +2509,13 @@ - katiagg tags: - images + +- title: LLMs 101: A Practical Introduction + path: articles/openai-cookbook-llms-101.md + date: 2025-15-9 + authors: + - paytonison + - asari + tags: + - llms + - beginners From 02381831bf13a414dc19a65e3219512ffce3c091 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 18:49:24 -0600 Subject: [PATCH 04/26] Update registry.yaml --- registry.yaml | 1 - 1 file changed, 1 deletion(-) diff --git a/registry.yaml b/registry.yaml index d7f9e954c8..416ac05a94 100644 --- a/registry.yaml +++ b/registry.yaml @@ -2515,7 +2515,6 @@ date: 2025-15-9 authors: - paytonison - - asari tags: - llms - beginners From 1ac7f5cf900c55f666b65e117af1999487260500 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 22:16:03 -0600 Subject: [PATCH 05/26] Delete how_to_work_with_large_language_models.md --- .../how_to_work_with_large_language_models.md | 168 ------------------ 1 file changed, 168 deletions(-) delete mode 100644 articles/how_to_work_with_large_language_models.md diff --git a/articles/how_to_work_with_large_language_models.md b/articles/how_to_work_with_large_language_models.md deleted file mode 100644 index cf6b48e1be..0000000000 --- a/articles/how_to_work_with_large_language_models.md +++ /dev/null @@ -1,168 +0,0 @@ -# How to work with large language models - -## How large language models work - -[Large language models][Large language models Blog Post] are functions that map text to text. Given an input string of text, a large language model predicts the text that should come next. - -The magic of large language models is that by being trained to minimize this prediction error over vast quantities of text, the models end up learning concepts useful for these predictions. For example, they learn: - -- how to spell -- how grammar works -- how to paraphrase -- how to answer questions -- how to hold a conversation -- how to write in many languages -- how to code -- etc. - -They do this by “reading” a large amount of existing text and learning how words tend to appear in context with other words, and uses what it has learned to predict the next most likely word that might appear in response to a user request, and each subsequent word after that. - -GPT-3 and GPT-4 power [many software products][OpenAI Customer Stories], including productivity apps, education apps, games, and more. - -## How to control a large language model - -Of all the inputs to a large language model, by far the most influential is the text prompt. - -Large language models can be prompted to produce output in a few ways: - -- **Instruction**: Tell the model what you want -- **Completion**: Induce the model to complete the beginning of what you want -- **Scenario**: Give the model a situation to play out -- **Demonstration**: Show the model what you want, with either: - - A few examples in the prompt - - Many hundreds or thousands of examples in a fine-tuning training dataset - -An example of each is shown below. - -### Instruction prompts - -Write your instruction at the top of the prompt (or at the bottom, or both), and the model will do its best to follow the instruction and then stop. Instructions can be detailed, so don't be afraid to write a paragraph explicitly detailing the output you want, just stay aware of how many [tokens](https://help.openai.com/en/articles/4936856-what-are-tokens-and-how-to-count-them) the model can process. - -Example instruction prompt: - -```text -Extract the name of the author from the quotation below. - -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation -``` - -Output: - -```text -Ted Chiang -``` - -### Completion prompt example - -Completion-style prompts take advantage of how large language models try to write text they think is most likely to come next. To steer the model, try beginning a pattern or sentence that will be completed by the output you want to see. Relative to direct instructions, this mode of steering large language models can take more care and experimentation. In addition, the models won't necessarily know where to stop, so you will often need stop sequences or post-processing to cut off text generated beyond the desired output. - -Example completion prompt: - -```text -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation - -The author of this quote is -``` - -Output: - -```text - Ted Chiang -``` - -### Scenario prompt example - -Giving the model a scenario to follow or role to play out can be helpful for complex queries or when seeking imaginative responses. When using a hypothetical prompt, you set up a situation, problem, or story, and then ask the model to respond as if it were a character in that scenario or an expert on the topic. - -Example scenario prompt: - -```text -Your role is to extract the name of the author from any given text - -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation -``` - -Output: - -```text - Ted Chiang -``` - -### Demonstration prompt example (few-shot learning) - -Similar to completion-style prompts, demonstrations can show the model what you want it to do. This approach is sometimes called few-shot learning, as the model learns from a few examples provided in the prompt. - -Example demonstration prompt: - -```text -Quote: -“When the reasoning mind is forced to confront the impossible again and again, it has no choice but to adapt.” -― N.K. Jemisin, The Fifth Season -Author: N.K. Jemisin - -Quote: -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation -Author: -``` - -Output: - -```text - Ted Chiang -``` - -### Fine-tuned prompt example - -With enough training examples, you can [fine-tune][Fine Tuning Docs] a custom model. In this case, instructions become unnecessary, as the model can learn the task from the training data provided. However, it can be helpful to include separator sequences (e.g., `->` or `###` or any string that doesn't commonly appear in your inputs) to tell the model when the prompt has ended and the output should begin. Without separator sequences, there is a risk that the model continues elaborating on the input text rather than starting on the answer you want to see. - -Example fine-tuned prompt (for a model that has been custom trained on similar prompt-completion pairs): - -```text -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation - -### - - -``` - -Output: - -```text - Ted Chiang -``` - -## Code Capabilities - -Large language models aren't only great at text - they can be great at code too. OpenAI's [GPT-4][GPT-4 and GPT-4 Turbo] model is a prime example. - -GPT-4 powers [numerous innovative products][OpenAI Customer Stories], including: - -- [GitHub Copilot] (autocompletes code in Visual Studio and other IDEs) -- [Replit](https://replit.com/) (can complete, explain, edit and generate code) -- [Cursor](https://cursor.sh/) (build software faster in an editor designed for pair-programming with AI) - -GPT-4 is more advanced than previous models like `gpt-3.5-turbo-instruct`. But, to get the best out of GPT-4 for coding tasks, it's still important to give clear and specific instructions. As a result, designing good prompts can take more care. - -### More prompt advice - -For more prompt examples, visit [OpenAI Examples][OpenAI Examples]. - -In general, the input prompt is the best lever for improving model outputs. You can try tricks like: - -- **Be more specific** E.g., if you want the output to be a comma separated list, ask it to return a comma separated list. If you want it to say "I don't know" when it doesn't know the answer, tell it 'Say "I don't know" if you do not know the answer.' The more specific your instructions, the better the model can respond. -- **Provide Context**: Help the model understand the bigger picture of your request. This could be background information, examples/demonstrations of what you want or explaining the purpose of your task. -- **Ask the model to answer as if it was an expert.** Explicitly asking the model to produce high quality output or output as if it was written by an expert can induce the model to give higher quality answers that it thinks an expert would write. Phrases like "Explain in detail" or "Describe step-by-step" can be effective. -- **Prompt the model to write down the series of steps explaining its reasoning.** If understanding the 'why' behind an answer is important, prompt the model to include its reasoning. This can be done by simply adding a line like "[Let's think step by step](https://arxiv.org/abs/2205.11916)" before each answer. - -[Fine Tuning Docs]: https://platform.openai.com/docs/guides/fine-tuning -[OpenAI Customer Stories]: https://openai.com/customer-stories -[Large language models Blog Post]: https://openai.com/research/better-language-models -[GitHub Copilot]: https://github.com/features/copilot/ -[GPT-4 and GPT-4 Turbo]: https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo -[GPT3 Apps Blog Post]: https://openai.com/blog/gpt-3-apps/ -[OpenAI Examples]: https://platform.openai.com/examples From a6c2e2743c6b16a148b13c772dcac3883034b0d1 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 22:38:55 -0600 Subject: [PATCH 06/26] Update registry.yaml Number dyslexia as an american is hard Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --- registry.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/registry.yaml b/registry.yaml index 416ac05a94..9914cb3b15 100644 --- a/registry.yaml +++ b/registry.yaml @@ -2512,7 +2512,7 @@ - title: LLMs 101: A Practical Introduction path: articles/openai-cookbook-llms-101.md - date: 2025-15-9 + date: 2025-09-15 authors: - paytonison tags: From 0699988cbcf08244a84ed27ac8cd287a4f946e5f Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 22:39:11 -0600 Subject: [PATCH 07/26] Update articles/openai-cookbook-llms-101.md Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --- articles/openai-cookbook-llms-101.md | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/articles/openai-cookbook-llms-101.md b/articles/openai-cookbook-llms-101.md index 77292198f6..f306effc2f 100644 --- a/articles/openai-cookbook-llms-101.md +++ b/articles/openai-cookbook-llms-101.md @@ -46,12 +46,14 @@ print(resp.output_text) import OpenAI from "openai"; const client = new OpenAI(); -const resp = await client.responses.create({ +const resp = await client.chat.completions.create({ model: "gpt-4o", - instructions: "You are a concise technical explainer.", - input: "In one paragraph, explain what a token is in an LLM." + messages: [ + { role: "system", content: "You are a concise technical explainer." }, + { role: "user", content: "In one paragraph, explain what a token is in an LLM." } + ] }); -console.log(resp.output_text); +console.log(resp.choices[0].message.content); ``` > **Tip.** Model names evolve; check your Models list before shipping. Prefer streaming for chat‑like UIs (see below). From 0f0cfe833f27a0ef53969c78775365f98630ecab Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Wed, 17 Sep 2025 17:51:57 -0600 Subject: [PATCH 08/26] Convert run-transformers.md guide to interactive notebook Replaces the static Markdown guide for running gpt-oss with Hugging Face Transformers with an interactive Jupyter notebook. The new notebook includes step-by-step code cells for environment setup, inference, serving, distributed usage, and a live ipywidgets chat UI, making the guide executable and more user-friendly. --- .../gpt_oss_transformers_interactive.ipynb | 147 ++++++++++ articles/gpt-oss/run-transformers.md | 271 ------------------ 2 files changed, 147 insertions(+), 271 deletions(-) create mode 100644 articles/gpt-oss/gpt_oss_transformers_interactive.ipynb delete mode 100644 articles/gpt-oss/run-transformers.md diff --git a/articles/gpt-oss/gpt_oss_transformers_interactive.ipynb b/articles/gpt-oss/gpt_oss_transformers_interactive.ipynb new file mode 100644 index 0000000000..d887e2155e --- /dev/null +++ b/articles/gpt-oss/gpt_oss_transformers_interactive.ipynb @@ -0,0 +1,147 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": "# gpt‑oss × Transformers — Interactive Notebook\n\nThis notebook turns the **run-transformers.md** guide into an interactive playground. It includes cells for environment setup, quick & advanced inference, serving with `transformers serve`, distributed/multi‑GPU hints, and a live **ipywidgets chat UI**.\n\n**Models:** `openai/gpt-oss-20b`, `openai/gpt-oss-120b` \n**Quantization:** MXFP4 by default (Hopper or later: H100/GB200/RTX 50xx). \n**Notes:** If you use `bfloat16`, memory usage rises (~48 GB for the 20B model). \n**Hardware:** 20B ≈16 GB VRAM with MXFP4; 120B ≥60 GB VRAM or multi‑GPU.\n\n> Tip: Run each cell in order the first time. You can skip pieces you don't need later.\n" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 0) Install/Update Dependencies\nThe guide recommends `transformers`, `accelerate`, `torch`, Triton 3.4, and MXFP4 kernels. We also install `ipywidgets` for the UI and `openai-harmony` for prompt tooling.\n\n**Heads‑up:** depending on your environment, the package named `kernels` may not exist. If installation fails for that package, you can safely omit it.\n\nExecute the cell below to install/update the basics (uncomment as needed):" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": "# If you're running locally, uncomment and run this once.\n# !pip -q install -U transformers accelerate torch triton==3.4 ipywidgets openai-harmony\n# Optional/experimental (may not exist on PyPI):\n# !pip -q install kernels\n\n# Enable widgets in some classic notebook setups (often not needed in JupyterLab 3+):\n# !jupyter nbextension enable --py widgetsnbextension\n\nimport platform, sys\nprint('Python', sys.version)\nprint('Platform', platform.platform())" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 1) GPU / Environment Check\nCheck PyTorch + CUDA availability and device details." + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": "try:\n import torch\n print('torch.__version__ =', torch.__version__)\n print('CUDA available =', torch.cuda.is_available())\n if torch.cuda.is_available():\n print('CUDA device count =', torch.cuda.device_count())\n for i in range(torch.cuda.device_count()):\n print(f' - [{i}]', torch.cuda.get_device_name(i))\n cc = torch.cuda.get_device_capability(i)\n print(' compute capability:', cc)\n try:\n free_mem = torch.cuda.mem_get_info()[0] / (1024**3)\n print(f'Approx free GPU memory: {free_mem:.1f} GB')\n except Exception:\n pass\nexcept Exception as e:\n print('Torch not available or failed to import:', e)" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 2) Quick Inference via `pipeline`\nThe high‑level Transformers `pipeline` makes it simple to run gpt‑oss models. This mirrors the guide's usage but wraps it in a function for convenience." + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": "from typing import List, Dict\nfrom contextlib import suppress\ntry:\n import torch\n from transformers import pipeline\nexcept Exception as e:\n print('Missing deps — run the install cell above. Error:', e)\n\ndef make_pipeline(model_name: str = 'openai/gpt-oss-20b', dtype: str = 'auto'):\n \"\"\"Create a text-generation pipeline for gpt-oss models.\"\"\"\n torch_dtype = 'auto'\n if dtype.lower() in {'bf16','bfloat16'}:\n torch_dtype = torch.bfloat16 if hasattr(torch, 'bfloat16') else 'auto'\n return pipeline(\n 'text-generation',\n model=model_name,\n torch_dtype=torch_dtype,\n device_map='auto'\n )\n\ndef run_pipeline_chat(\n generator,\n system_prompt: str,\n user_prompt: str,\n max_new_tokens: int = 200,\n temperature: float = 1.0,\n):\n messages = []\n if system_prompt.strip():\n messages.append({'role': 'system', 'content': system_prompt.strip()})\n messages.append({'role': 'user', 'content': user_prompt})\n try:\n result = generator(\n messages,\n max_new_tokens=max_new_tokens,\n temperature=temperature,\n )\n # Common structure: result[0]['generated_text'] contains the whole transcript or text\n with suppress(Exception):\n return result[0].get('generated_text', str(result))\n return str(result)\n except TypeError:\n # Some older pipelines accept a string; fallback to simple concatenation\n prompt = (system_prompt + '\\n\\n' if system_prompt.strip() else '') + user_prompt\n result = generator(prompt, max_new_tokens=max_new_tokens, temperature=temperature)\n return result[0]['generated_text'] if isinstance(result, list) else str(result)\n" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 3) Interactive Chat Playground (ipywidgets)\nUse the dropdowns and sliders below to load a model and generate responses.\nThe first generation will download weights (time depends on your bandwidth and disk)." + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": "import traceback\nfrom IPython.display import display, Markdown\nimport ipywidgets as widgets\n\nmodel_dd = widgets.Dropdown(\n options=['openai/gpt-oss-20b', 'openai/gpt-oss-120b'],\n value='openai/gpt-oss-20b',\n description='Model:',\n style={'description_width': '70px'},\n layout=widgets.Layout(width='50%'),\n)\ndtype_dd = widgets.Dropdown(\n options=['auto', 'bfloat16'],\n value='auto',\n description='dtype:',\n style={'description_width': '70px'},\n)\nload_btn = widgets.Button(description='Load / Reload Model', button_style='info')\nload_out = widgets.Output()\n\nsys_in = widgets.Textarea(\n value='You are a helpful assistant.',\n description='System',\n layout=widgets.Layout(width='100%', height='80px'),\n style={'description_width': '70px'}\n)\nusr_in = widgets.Textarea(\n value='Explain what MXFP4 quantization is.',\n description='User',\n layout=widgets.Layout(width='100%', height='80px'),\n style={'description_width': '70px'}\n)\n\ntemp_sl = widgets.FloatSlider(value=1.0, min=0.0, max=2.0, step=0.05, description='Temp:',\n style={'description_width': '70px'}, readout_format='.2f')\ntokens_sl = widgets.IntSlider(value=200, min=16, max=4096, step=16, description='Max tokens:',\n style={'description_width': '90px'})\n\ngen_btn = widgets.Button(description='Generate', button_style='primary')\nout = widgets.Output()\npipe_holder = {'pipe': None}\n\ndef on_load_clicked(_):\n load_out.clear_output()\n out.clear_output()\n with load_out:\n try:\n display(Markdown(f\"**Loading** `{model_dd.value}` with dtype `{dtype_dd.value}`...\"))\n pipe_holder['pipe'] = make_pipeline(model_dd.value, dtype_dd.value)\n display(Markdown('✅ **Model loaded**'))\n except Exception:\n traceback.print_exc()\n\ndef on_generate_clicked(_):\n out.clear_output()\n with out:\n if pipe_holder['pipe'] is None:\n display(Markdown('⚠️ Load a model first.'))\n return\n try:\n text = run_pipeline_chat(\n pipe_holder['pipe'],\n sys_in.value,\n usr_in.value,\n max_new_tokens=int(tokens_sl.value),\n temperature=float(temp_sl.value),\n )\n display(Markdown('**Output**'))\n display(Markdown(f'```\n{text}\n```'))\n except Exception:\n traceback.print_exc()\n\nload_btn.on_click(on_load_clicked)\ngen_btn.on_click(on_generate_clicked)\n\nui = widgets.VBox([\n widgets.HBox([model_dd, dtype_dd, load_btn]),\n load_out,\n sys_in,\n usr_in,\n widgets.HBox([temp_sl, tokens_sl]),\n gen_btn,\n out\n])\ndisplay(ui)" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 4) Advanced Inference with `.generate()`\nManual control using `AutoModelForCausalLM` and `AutoTokenizer`, including the chat template." + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": "from transformers import AutoModelForCausalLM, AutoTokenizer\nimport torch\n\nmodel_name = 'openai/gpt-oss-20b'\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(\n model_name,\n torch_dtype='auto',\n device_map='auto'\n)\n\nmessages = [\n {'role': 'user', 'content': 'Explain what MXFP4 quantization is.'},\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors='pt',\n return_dict=True,\n).to(model.device)\n\noutputs = model.generate(\n **inputs,\n max_new_tokens=200,\n temperature=0.7\n)\n\nprint(tokenizer.decode(outputs[0]))" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 5) Chat Template & Tool Calling (Harmony)\nThe gpt‑oss models use the Harmony response format. You can either rely on the built‑in chat template or use the `openai-harmony` library to build/parse prompts and completions." + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": "# If needed, install the library first in the install cell above: !pip -q install openai-harmony\nimport json\nfrom openai_harmony import (\n HarmonyEncodingName,\n load_harmony_encoding,\n Conversation,\n Message,\n Role,\n SystemContent,\n DeveloperContent,\n)\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nencoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n\nconvo = Conversation.from_messages([\n Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n Message.from_role_and_content(\n Role.DEVELOPER,\n DeveloperContent.new().with_instructions('Always respond in riddles')\n ),\n Message.from_role_and_content(Role.USER, 'What is the weather like in SF?')\n])\n\nprefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\nstop_token_ids = encoding.stop_tokens_for_assistant_actions()\n\nmodel_name = 'openai/gpt-oss-20b'\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype='auto', device_map='auto')\n\noutputs = model.generate(\n input_ids=[prefill_ids],\n max_new_tokens=128,\n eos_token_id=stop_token_ids\n)\n\ncompletion_ids = outputs[0][len(prefill_ids):]\nentries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\nfor message in entries:\n print(json.dumps(message.to_dict(), indent=2))" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 6) Serve an OpenAI‑style Responses endpoint\nYou can serve the model locally via the `transformers serve` CLI to expose a `/v1/responses` endpoint. Run these **commands in a terminal** (they are shown here for reference):" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "trusted": false + }, + "outputs": [], + "source": "# --- Terminal commands (reference) ---\n# transformers serve\n# transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n# curl -X POST http://localhost:8000/v1/responses \\\n# -H 'Content-Type: application/json' \\\n# -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n\nprint('See comments above for terminal commands. You can copy/paste them into your shell.')" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "### Client example (from a notebook) against `transformers serve`\nAdjust the URL and payload to your needs." + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": "import json, requests\nurl = 'http://localhost:8000/v1/responses'\npayload = {\n 'messages': [\n {'role': 'system', 'content': 'hello'},\n {'role': 'user', 'content': 'Say hi in one sentence.'}\n ],\n 'temperature': 0.7,\n 'max_tokens': 64,\n 'stream': False,\n 'model': 'openai/gpt-oss-20b'\n}\ntry:\n r = requests.post(url, json=payload, timeout=30)\n print(r.status_code)\n print(r.text[:1000])\nexcept Exception as e:\n print('Request failed (is the server running?):', e)" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 7) Multi‑GPU & Distributed Inference\nSketch for automatic placement, tensor parallelism, and expert parallelism (edit for your cluster)." + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": "from transformers import AutoModelForCausalLM, AutoTokenizer\ntry:\n from transformers.distributed import DistributedConfig\nexcept Exception:\n class DistributedConfig:\n def __init__(self, **kwargs):\n self.kwargs = kwargs\n\nmodel_path = 'openai/gpt-oss-120b'\ntokenizer = AutoTokenizer.from_pretrained(model_path, padding_side='left')\n\ndevice_map = {\n 'distributed_config': DistributedConfig(enable_expert_parallel=1),\n 'tp_plan': 'auto',\n}\n\nmodel = AutoModelForCausalLM.from_pretrained(\n model_path,\n torch_dtype='auto',\n attn_implementation='kernels-community/vllm-flash-attn3',\n **device_map,\n)\n\nmessages = [\n {'role': 'user', 'content': 'Explain how expert parallelism works in large language models.'}\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors='pt',\n return_dict=True,\n).to(model.device)\n\noutputs = model.generate(**inputs, max_new_tokens=512)\nresponse = tokenizer.decode(outputs[0])\nprint('Model response (truncated):\\n', response[:1000])\n\n# To launch across 4 GPUs from a script:\n# torchrun --nproc_per_node=4 generate.py\n" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## 8) Troubleshooting Checklist\n- **Out of memory**: Use MXFP4 on Hopper‑class GPUs; reduce `max_new_tokens`; try `device_map='auto'`. \n- **Kernel/attn issues**: Ensure Triton is compatible with your CUDA; stick to Triton 3.4 as suggested. \n- **Slow loads**: Use local weight caches; set `TRANSFORMERS_CACHE`. \n- **No widgets**: Ensure `ipywidgets` is installed and enabled in your notebook frontend. \n- **Server 404/500**: Verify `transformers serve` version and model path. \n- **bfloat16 errors**: Fall back to `torch_dtype='auto'` or ensure hardware support." + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "## Appendix: Original Guide (for reference)\nThe content below is the original `/mnt/data/run-transformers.md` you provided:\n\n
\nClick to expand\n\n# How to run gpt-oss with Hugging Face Transformers\n\nThe Transformers library by Hugging Face provides a flexible way to load and run large language models locally or on a server. This guide will walk you through running [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) using Transformers, either with a high-level pipeline or via low-level `generate` calls with raw token IDs.\n\nWe'll cover the use of [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) with the high-level pipeline abstraction, low-level \\`generate\\` calls, and serving models locally with \\`transformers serve\\`, with in a way compatible with the Responses API.\n\nIn this guide we’ll run through various optimised ways to run the **gpt-oss models via Transformers.**\n\nBonus: You can also fine-tune models via transformers, [check out our fine-tuning guide here](https://cookbook.openai.com/articles/gpt-oss/fine-tune-transformers).\n\n## Pick your model\n\nBoth **gpt-oss** models are available on Hugging Face:\n\n- **`openai/gpt-oss-20b`**\n - \\~16GB VRAM requirement when using MXFP4\n - Great for single high-end consumer GPUs\n- **`openai/gpt-oss-120b`**\n - Requires ≥60GB VRAM or multi-GPU setup\n - Ideal for H100-class hardware\n\nBoth are **MXFP4 quantized** by default. Please, note that MXFP4 is supported in Hopper or later architectures. This includes data center GPUs such as H100 or GB200, as well as the latest RTX 50xx family of consumer cards.\n\nIf you use `bfloat16` instead of MXFP4, memory consumption will be larger (\\~48 GB for the 20b parameter model).\n\n## Quick setup\n\n1. **Install dependencies** \n It’s recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility:\n\n```bash\npip install -U transformers accelerate torch triton==3.4 kernels\n```\n\n2. **(Optional) Enable multi-GPU** \n If you’re running large models, use Accelerate or torchrun to handle device mapping automatically.\n\n## Create an Open AI Responses / Chat Completions endpoint\n\nTo launch a server, simply use the `transformers serve` CLI command:\n\n```bash\ntransformers serve\n```\n\nThe simplest way to interact with the server is through the transformers chat CLI\n\n```bash\ntransformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n```\n\nor by sending an HTTP request with cURL, e.g.\n\n```bash\ncurl -X POST http://localhost:8000/v1/responses -H \"Content-Type: application/json\" -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n```\n\nAdditional use cases, like integrating `transformers serve` with Cursor and other tools, are detailed in [the documentation](https://huggingface.co/docs/transformers/main/serving).\n\n## Quick inference with pipeline\n\nThe easiest way to run the gpt-oss models is with the Transformers high-level `pipeline` API:\n\n```py\nfrom transformers import pipeline\n\ngenerator = pipeline(\n \"text-generation\",\n model=\"openai/gpt-oss-20b\",\n torch_dtype=\"auto\",\n device_map=\"auto\" # Automatically place on available GPUs\n)\n\nmessages = [\n {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n]\n\nresult = generator(\n messages,\n max_new_tokens=200,\n temperature=1.0,\n)\n\nprint(result[0][\"generated_text\"])\n```\n\n## Advanced inference with `.generate()`\n\nIf you want more control, you can load the model and tokenizer manually and invoke the `.generate()` method:\n\n```py\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nmodel_name = \"openai/gpt-oss-20b\"\n\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(\n model_name,\n torch_dtype=\"auto\",\n device_map=\"auto\"\n)\n\nmessages = [\n {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors=\"pt\",\n return_dict=True,\n).to(model.device)\n\noutputs = model.generate(\n **inputs,\n max_new_tokens=200,\n temperature=0.7\n)\n\nprint(tokenizer.decode(outputs[0]))\n```\n\n## Chat template and tool calling\n\nOpenAI gpt-oss models use the [harmony response format](https://cookbook.openai.com/article/harmony) for structuring messages, including reasoning and tool calls.\n\nTo construct prompts you can use the built-in chat template of Transformers. Alternatively, you can install and use the [openai-harmony library](https://github.com/openai/harmony) for more control.\n\nTo use the chat template:\n\n```py\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nmodel_name = \"openai/gpt-oss-20b\"\n\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(\n model_name,\n device_map=\"auto\",\n torch_dtype=\"auto\",\n)\n\nmessages = [\n {\"role\": \"system\", \"content\": \"Always respond in riddles\"},\n {\"role\": \"user\", \"content\": \"What is the weather like in Madrid?\"},\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors=\"pt\",\n return_dict=True,\n).to(model.device)\n\ngenerated = model.generate(**inputs, max_new_tokens=100)\nprint(tokenizer.decode(generated[0][inputs[\"input_ids\"].shape[-1] :]))\n```\n\nTo integrate the [`openai-harmony`](https://github.com/openai/harmony) library to prepare prompts and parse responses, first install it like this:\n\n```bash\npip install openai-harmony\n```\n\nHere’s an example of how to use the library to build your prompts and encode them to tokens:\n\n```py\nimport json\nfrom openai_harmony import (\n HarmonyEncodingName,\n load_harmony_encoding,\n Conversation,\n Message,\n Role,\n SystemContent,\n DeveloperContent\n)\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nencoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n\n# Build conversation\nconvo = Conversation.from_messages([\n Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n Message.from_role_and_content(\n Role.DEVELOPER,\n DeveloperContent.new().with_instructions(\"Always respond in riddles\")\n ),\n Message.from_role_and_content(Role.USER, \"What is the weather like in SF?\")\n])\n\n# Render prompt\nprefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\nstop_token_ids = encoding.stop_tokens_for_assistant_actions()\n\n# Load model\nmodel_name = \"openai/gpt-oss-20b\"\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=\"auto\", device_map=\"auto\")\n\n# Generate\noutputs = model.generate(\n input_ids=[prefill_ids],\n max_new_tokens=128,\n eos_token_id=stop_token_ids\n)\n\n# Parse completion tokens\ncompletion_ids = outputs[0][len(prefill_ids):]\nentries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n\nfor message in entries:\n print(json.dumps(message.to_dict(), indent=2))\n```\n\nNote that the `Developer` role in Harmony maps to the `system` prompt in the chat template.\n\n## Multi-GPU & distributed inference\n\nThe large gpt-oss-120b fits on a single H100 GPU when using MXFP4. If you want to run it on multiple GPUs, you can:\n\n- Use `tp_plan=\"auto\"` for automatic placement and tensor parallelism\n- Launch with `accelerate launch or torchrun` for distributed setups\n- Leverage Expert Parallelism\n- Use specialised Flash attention kernels for faster inference\n\n```py\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\nfrom transformers.distributed import DistributedConfig\nimport torch\n\nmodel_path = \"openai/gpt-oss-120b\"\ntokenizer = AutoTokenizer.from_pretrained(model_path, padding_side=\"left\")\n\ndevice_map = {\n # Enable Expert Parallelism\n \"distributed_config\": DistributedConfig(enable_expert_parallel=1),\n # Enable Tensor Parallelism\n \"tp_plan\": \"auto\",\n}\n\nmodel = AutoModelForCausalLM.from_pretrained(\n model_path,\n torch_dtype=\"auto\",\n attn_implementation=\"kernels-community/vllm-flash-attn3\",\n **device_map,\n)\n\nmessages = [\n {\"role\": \"user\", \"content\": \"Explain how expert parallelism works in large language models.\"}\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors=\"pt\",\n return_dict=True,\n).to(model.device)\n\noutputs = model.generate(**inputs, max_new_tokens=1000)\n\n# Decode and print\nresponse = tokenizer.decode(outputs[0])\nprint(\"Model response:\", response.split(\"<|channel|>final<|message|>\")[-1].strip())\n```\n\nYou can then run this on a node with four GPUs via\n\n```bash\ntorchrun --nproc_per_node=4 generate.py\n```\n\n\n
" + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": "_Notebook generated on 2025-09-17T23:43:44Z_" + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "name": "python", + "pygments_lexer": "ipython3" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} \ No newline at end of file diff --git a/articles/gpt-oss/run-transformers.md b/articles/gpt-oss/run-transformers.md deleted file mode 100644 index 0a1ec12364..0000000000 --- a/articles/gpt-oss/run-transformers.md +++ /dev/null @@ -1,271 +0,0 @@ -# How to run gpt-oss with Hugging Face Transformers - -The Transformers library by Hugging Face provides a flexible way to load and run large language models locally or on a server. This guide will walk you through running [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) using Transformers, either with a high-level pipeline or via low-level `generate` calls with raw token IDs. - -We'll cover the use of [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) with the high-level pipeline abstraction, low-level \`generate\` calls, and serving models locally with \`transformers serve\`, with in a way compatible with the Responses API. - -In this guide we’ll run through various optimised ways to run the **gpt-oss models via Transformers.** - -Bonus: You can also fine-tune models via transformers, [check out our fine-tuning guide here](https://cookbook.openai.com/articles/gpt-oss/fine-tune-transformers). - -## Pick your model - -Both **gpt-oss** models are available on Hugging Face: - -- **`openai/gpt-oss-20b`** - - \~16GB VRAM requirement when using MXFP4 - - Great for single high-end consumer GPUs -- **`openai/gpt-oss-120b`** - - Requires ≥60GB VRAM or multi-GPU setup - - Ideal for H100-class hardware - -Both are **MXFP4 quantized** by default. Please, note that MXFP4 is supported in Hopper or later architectures. This includes data center GPUs such as H100 or GB200, as well as the latest RTX 50xx family of consumer cards. - -If you use `bfloat16` instead of MXFP4, memory consumption will be larger (\~48 GB for the 20b parameter model). - -## Quick setup - -1. **Install dependencies** - It’s recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility: - -```bash -pip install -U transformers accelerate torch triton==3.4 kernels -``` - -2. **(Optional) Enable multi-GPU** - If you’re running large models, use Accelerate or torchrun to handle device mapping automatically. - -## Create an Open AI Responses / Chat Completions endpoint - -To launch a server, simply use the `transformers serve` CLI command: - -```bash -transformers serve -``` - -The simplest way to interact with the server is through the transformers chat CLI - -```bash -transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b -``` - -or by sending an HTTP request with cURL, e.g. - -```bash -curl -X POST http://localhost:8000/v1/responses -H "Content-Type: application/json" -d '{"messages": [{"role": "system", "content": "hello"}], "temperature": 0.9, "max_tokens": 1000, "stream": true, "model": "openai/gpt-oss-20b"}' -``` - -Additional use cases, like integrating `transformers serve` with Cursor and other tools, are detailed in [the documentation](https://huggingface.co/docs/transformers/main/serving). - -## Quick inference with pipeline - -The easiest way to run the gpt-oss models is with the Transformers high-level `pipeline` API: - -```py -from transformers import pipeline - -generator = pipeline( - "text-generation", - model="openai/gpt-oss-20b", - torch_dtype="auto", - device_map="auto" # Automatically place on available GPUs -) - -messages = [ - {"role": "user", "content": "Explain what MXFP4 quantization is."}, -] - -result = generator( - messages, - max_new_tokens=200, - temperature=1.0, -) - -print(result[0]["generated_text"]) -``` - -## Advanced inference with `.generate()` - -If you want more control, you can load the model and tokenizer manually and invoke the `.generate()` method: - -```py -from transformers import AutoModelForCausalLM, AutoTokenizer - -model_name = "openai/gpt-oss-20b" - -tokenizer = AutoTokenizer.from_pretrained(model_name) -model = AutoModelForCausalLM.from_pretrained( - model_name, - torch_dtype="auto", - device_map="auto" -) - -messages = [ - {"role": "user", "content": "Explain what MXFP4 quantization is."}, -] - -inputs = tokenizer.apply_chat_template( - messages, - add_generation_prompt=True, - return_tensors="pt", - return_dict=True, -).to(model.device) - -outputs = model.generate( - **inputs, - max_new_tokens=200, - temperature=0.7 -) - -print(tokenizer.decode(outputs[0])) -``` - -## Chat template and tool calling - -OpenAI gpt-oss models use the [harmony response format](https://cookbook.openai.com/article/harmony) for structuring messages, including reasoning and tool calls. - -To construct prompts you can use the built-in chat template of Transformers. Alternatively, you can install and use the [openai-harmony library](https://github.com/openai/harmony) for more control. - -To use the chat template: - -```py -from transformers import AutoModelForCausalLM, AutoTokenizer - -model_name = "openai/gpt-oss-20b" - -tokenizer = AutoTokenizer.from_pretrained(model_name) -model = AutoModelForCausalLM.from_pretrained( - model_name, - device_map="auto", - torch_dtype="auto", -) - -messages = [ - {"role": "system", "content": "Always respond in riddles"}, - {"role": "user", "content": "What is the weather like in Madrid?"}, -] - -inputs = tokenizer.apply_chat_template( - messages, - add_generation_prompt=True, - return_tensors="pt", - return_dict=True, -).to(model.device) - -generated = model.generate(**inputs, max_new_tokens=100) -print(tokenizer.decode(generated[0][inputs["input_ids"].shape[-1] :])) -``` - -To integrate the [`openai-harmony`](https://github.com/openai/harmony) library to prepare prompts and parse responses, first install it like this: - -```bash -pip install openai-harmony -``` - -Here’s an example of how to use the library to build your prompts and encode them to tokens: - -```py -import json -from openai_harmony import ( - HarmonyEncodingName, - load_harmony_encoding, - Conversation, - Message, - Role, - SystemContent, - DeveloperContent -) -from transformers import AutoModelForCausalLM, AutoTokenizer - -encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS) - -# Build conversation -convo = Conversation.from_messages([ - Message.from_role_and_content(Role.SYSTEM, SystemContent.new()), - Message.from_role_and_content( - Role.DEVELOPER, - DeveloperContent.new().with_instructions("Always respond in riddles") - ), - Message.from_role_and_content(Role.USER, "What is the weather like in SF?") -]) - -# Render prompt -prefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT) -stop_token_ids = encoding.stop_tokens_for_assistant_actions() - -# Load model -model_name = "openai/gpt-oss-20b" -tokenizer = AutoTokenizer.from_pretrained(model_name) -model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype="auto", device_map="auto") - -# Generate -outputs = model.generate( - input_ids=[prefill_ids], - max_new_tokens=128, - eos_token_id=stop_token_ids -) - -# Parse completion tokens -completion_ids = outputs[0][len(prefill_ids):] -entries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT) - -for message in entries: - print(json.dumps(message.to_dict(), indent=2)) -``` - -Note that the `Developer` role in Harmony maps to the `system` prompt in the chat template. - -## Multi-GPU & distributed inference - -The large gpt-oss-120b fits on a single H100 GPU when using MXFP4. If you want to run it on multiple GPUs, you can: - -- Use `tp_plan="auto"` for automatic placement and tensor parallelism -- Launch with `accelerate launch or torchrun` for distributed setups -- Leverage Expert Parallelism -- Use specialised Flash attention kernels for faster inference - -```py -from transformers import AutoModelForCausalLM, AutoTokenizer -from transformers.distributed import DistributedConfig -import torch - -model_path = "openai/gpt-oss-120b" -tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side="left") - -device_map = { - # Enable Expert Parallelism - "distributed_config": DistributedConfig(enable_expert_parallel=1), - # Enable Tensor Parallelism - "tp_plan": "auto", -} - -model = AutoModelForCausalLM.from_pretrained( - model_path, - torch_dtype="auto", - attn_implementation="kernels-community/vllm-flash-attn3", - **device_map, -) - -messages = [ - {"role": "user", "content": "Explain how expert parallelism works in large language models."} -] - -inputs = tokenizer.apply_chat_template( - messages, - add_generation_prompt=True, - return_tensors="pt", - return_dict=True, -).to(model.device) - -outputs = model.generate(**inputs, max_new_tokens=1000) - -# Decode and print -response = tokenizer.decode(outputs[0]) -print("Model response:", response.split("<|channel|>final<|message|>")[-1].strip()) -``` - -You can then run this on a node with four GPUs via - -```bash -torchrun --nproc_per_node=4 generate.py -``` From da038a14eb6081bcb81b2b54e4188fe5ee3e98f1 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Wed, 17 Sep 2025 18:14:09 -0600 Subject: [PATCH 09/26] Make HF Transformers markdown into Jupyter notebook. --- .../gpt_oss_transformers_interactive.ipynb | 697 +++++++++++++++++- 1 file changed, 672 insertions(+), 25 deletions(-) diff --git a/articles/gpt-oss/gpt_oss_transformers_interactive.ipynb b/articles/gpt-oss/gpt_oss_transformers_interactive.ipynb index d887e2155e..8a37646d82 100644 --- a/articles/gpt-oss/gpt_oss_transformers_interactive.ipynb +++ b/articles/gpt-oss/gpt_oss_transformers_interactive.ipynb @@ -3,84 +3,357 @@ { "cell_type": "markdown", "metadata": {}, - "source": "# gpt‑oss × Transformers — Interactive Notebook\n\nThis notebook turns the **run-transformers.md** guide into an interactive playground. It includes cells for environment setup, quick & advanced inference, serving with `transformers serve`, distributed/multi‑GPU hints, and a live **ipywidgets chat UI**.\n\n**Models:** `openai/gpt-oss-20b`, `openai/gpt-oss-120b` \n**Quantization:** MXFP4 by default (Hopper or later: H100/GB200/RTX 50xx). \n**Notes:** If you use `bfloat16`, memory usage rises (~48 GB for the 20B model). \n**Hardware:** 20B ≈16 GB VRAM with MXFP4; 120B ≥60 GB VRAM or multi‑GPU.\n\n> Tip: Run each cell in order the first time. You can skip pieces you don't need later.\n" + "source": [ + "# gpt‑oss × Transformers — Interactive Notebook\n", + "\n", + "This notebook turns the **run-transformers.md** guide into an interactive playground. It includes cells for environment setup, quick & advanced inference, serving with `transformers serve`, distributed/multi‑GPU hints, and a live **ipywidgets chat UI**.\n", + "\n", + "**Models:** `openai/gpt-oss-20b`, `openai/gpt-oss-120b` \n", + "**Quantization:** MXFP4 by default (Hopper or later: H100/GB200/RTX 50xx). \n", + "**Notes:** If you use `bfloat16`, memory usage rises (~48 GB for the 20B model). \n", + "**Hardware:** 20B ≈16 GB VRAM with MXFP4; 120B ≥60 GB VRAM or multi‑GPU.\n", + "\n", + "> Tip: Run each cell in order the first time. You can skip pieces you don't need later.\n" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 0) Install/Update Dependencies\nThe guide recommends `transformers`, `accelerate`, `torch`, Triton 3.4, and MXFP4 kernels. We also install `ipywidgets` for the UI and `openai-harmony` for prompt tooling.\n\n**Heads‑up:** depending on your environment, the package named `kernels` may not exist. If installation fails for that package, you can safely omit it.\n\nExecute the cell below to install/update the basics (uncomment as needed):" + "source": [ + "## 0) Install/Update Dependencies\n", + "The guide recommends `transformers`, `accelerate`, `torch`, Triton 3.4, and MXFP4 kernels. We also install `ipywidgets` for the UI and `openai-harmony` for prompt tooling.\n", + "\n", + "**Heads‑up:** depending on your environment, the package named `kernels` may not exist. If installation fails for that package, you can safely omit it.\n", + "\n", + "Execute the cell below to install/update the basics (uncomment as needed):" + ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], - "source": "# If you're running locally, uncomment and run this once.\n# !pip -q install -U transformers accelerate torch triton==3.4 ipywidgets openai-harmony\n# Optional/experimental (may not exist on PyPI):\n# !pip -q install kernels\n\n# Enable widgets in some classic notebook setups (often not needed in JupyterLab 3+):\n# !jupyter nbextension enable --py widgetsnbextension\n\nimport platform, sys\nprint('Python', sys.version)\nprint('Platform', platform.platform())" + "source": [ + "!pip -q install -U transformers accelerate torch triton==3.4 ipywidgets openai-harmony torchvision\n", + "!pip -q install kernels\n", + "\n", + "import platform, sys\n", + "print('Python', sys.version)\n", + "print('Platform', platform.platform())" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 1) GPU / Environment Check\nCheck PyTorch + CUDA availability and device details." + "source": [ + "## 1) GPU / Environment Check\n", + "Check PyTorch + CUDA availability and device details." + ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], - "source": "try:\n import torch\n print('torch.__version__ =', torch.__version__)\n print('CUDA available =', torch.cuda.is_available())\n if torch.cuda.is_available():\n print('CUDA device count =', torch.cuda.device_count())\n for i in range(torch.cuda.device_count()):\n print(f' - [{i}]', torch.cuda.get_device_name(i))\n cc = torch.cuda.get_device_capability(i)\n print(' compute capability:', cc)\n try:\n free_mem = torch.cuda.mem_get_info()[0] / (1024**3)\n print(f'Approx free GPU memory: {free_mem:.1f} GB')\n except Exception:\n pass\nexcept Exception as e:\n print('Torch not available or failed to import:', e)" + "source": [ + "try:\n", + " import torch\n", + " print('torch.__version__ =', torch.__version__)\n", + " print('CUDA available =', torch.cuda.is_available())\n", + " if torch.cuda.is_available():\n", + " print('CUDA device count =', torch.cuda.device_count())\n", + " for i in range(torch.cuda.device_count()):\n", + " print(f' - [{i}]', torch.cuda.get_device_name(i))\n", + " cc = torch.cuda.get_device_capability(i)\n", + " print(' compute capability:', cc)\n", + " try:\n", + " free_mem = torch.cuda.mem_get_info()[0] / (1024**3)\n", + " print(f'Approx free GPU memory: {free_mem:.1f} GB')\n", + " except Exception:\n", + " pass\n", + "except Exception as e:\n", + " print('Torch not available or failed to import:', e)" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 2) Quick Inference via `pipeline`\nThe high‑level Transformers `pipeline` makes it simple to run gpt‑oss models. This mirrors the guide's usage but wraps it in a function for convenience." + "source": [ + "## 2) Quick Inference via `pipeline`\n", + "The high‑level Transformers `pipeline` makes it simple to run gpt‑oss models. This mirrors the guide's usage but wraps it in a function for convenience." + ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], - "source": "from typing import List, Dict\nfrom contextlib import suppress\ntry:\n import torch\n from transformers import pipeline\nexcept Exception as e:\n print('Missing deps — run the install cell above. Error:', e)\n\ndef make_pipeline(model_name: str = 'openai/gpt-oss-20b', dtype: str = 'auto'):\n \"\"\"Create a text-generation pipeline for gpt-oss models.\"\"\"\n torch_dtype = 'auto'\n if dtype.lower() in {'bf16','bfloat16'}:\n torch_dtype = torch.bfloat16 if hasattr(torch, 'bfloat16') else 'auto'\n return pipeline(\n 'text-generation',\n model=model_name,\n torch_dtype=torch_dtype,\n device_map='auto'\n )\n\ndef run_pipeline_chat(\n generator,\n system_prompt: str,\n user_prompt: str,\n max_new_tokens: int = 200,\n temperature: float = 1.0,\n):\n messages = []\n if system_prompt.strip():\n messages.append({'role': 'system', 'content': system_prompt.strip()})\n messages.append({'role': 'user', 'content': user_prompt})\n try:\n result = generator(\n messages,\n max_new_tokens=max_new_tokens,\n temperature=temperature,\n )\n # Common structure: result[0]['generated_text'] contains the whole transcript or text\n with suppress(Exception):\n return result[0].get('generated_text', str(result))\n return str(result)\n except TypeError:\n # Some older pipelines accept a string; fallback to simple concatenation\n prompt = (system_prompt + '\\n\\n' if system_prompt.strip() else '') + user_prompt\n result = generator(prompt, max_new_tokens=max_new_tokens, temperature=temperature)\n return result[0]['generated_text'] if isinstance(result, list) else str(result)\n" + "source": [ + "from typing import List, Dict\n", + "from contextlib import suppress\n", + "try:\n", + " import torch\n", + " from transformers import pipeline\n", + "except Exception as e:\n", + " print('Missing deps — run the install cell above. Error:', e)\n", + "\n", + "def make_pipeline(model_name: str = 'openai/gpt-oss-20b', dtype: str = 'auto'):\n", + " \"\"\"Create a text-generation pipeline for gpt-oss models.\"\"\"\n", + " torch_dtype = 'auto'\n", + " if dtype.lower() in {'bf16','bfloat16'}:\n", + " torch_dtype = torch.bfloat16 if hasattr(torch, 'bfloat16') else 'auto'\n", + " return pipeline(\n", + " 'text-generation',\n", + " model=model_name,\n", + " torch_dtype=torch_dtype,\n", + " device_map='auto'\n", + " )\n", + "\n", + "def run_pipeline_chat(\n", + " generator,\n", + " system_prompt: str,\n", + " user_prompt: str,\n", + " max_new_tokens: int = 200,\n", + " temperature: float = 1.0,\n", + "):\n", + " messages = []\n", + " if system_prompt.strip():\n", + " messages.append({'role': 'system', 'content': system_prompt.strip()})\n", + " messages.append({'role': 'user', 'content': user_prompt})\n", + " try:\n", + " result = generator(\n", + " messages,\n", + " max_new_tokens=max_new_tokens,\n", + " temperature=temperature,\n", + " )\n", + " # Common structure: result[0]['generated_text'] contains the whole transcript or text\n", + " with suppress(Exception):\n", + " return result[0].get('generated_text', str(result))\n", + " return str(result)\n", + " except TypeError:\n", + " # Some older pipelines accept a string; fallback to simple concatenation\n", + " prompt = (system_prompt + '\\n\\n' if system_prompt.strip() else '') + user_prompt\n", + " result = generator(prompt, max_new_tokens=max_new_tokens, temperature=temperature)\n", + " return result[0]['generated_text'] if isinstance(result, list) else str(result)\n" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 3) Interactive Chat Playground (ipywidgets)\nUse the dropdowns and sliders below to load a model and generate responses.\nThe first generation will download weights (time depends on your bandwidth and disk)." + "source": [ + "## 3) Interactive Chat Playground (ipywidgets)\n", + "Use the dropdowns and sliders below to load a model and generate responses.\n", + "The first generation will download weights (time depends on your bandwidth and disk)." + ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], - "source": "import traceback\nfrom IPython.display import display, Markdown\nimport ipywidgets as widgets\n\nmodel_dd = widgets.Dropdown(\n options=['openai/gpt-oss-20b', 'openai/gpt-oss-120b'],\n value='openai/gpt-oss-20b',\n description='Model:',\n style={'description_width': '70px'},\n layout=widgets.Layout(width='50%'),\n)\ndtype_dd = widgets.Dropdown(\n options=['auto', 'bfloat16'],\n value='auto',\n description='dtype:',\n style={'description_width': '70px'},\n)\nload_btn = widgets.Button(description='Load / Reload Model', button_style='info')\nload_out = widgets.Output()\n\nsys_in = widgets.Textarea(\n value='You are a helpful assistant.',\n description='System',\n layout=widgets.Layout(width='100%', height='80px'),\n style={'description_width': '70px'}\n)\nusr_in = widgets.Textarea(\n value='Explain what MXFP4 quantization is.',\n description='User',\n layout=widgets.Layout(width='100%', height='80px'),\n style={'description_width': '70px'}\n)\n\ntemp_sl = widgets.FloatSlider(value=1.0, min=0.0, max=2.0, step=0.05, description='Temp:',\n style={'description_width': '70px'}, readout_format='.2f')\ntokens_sl = widgets.IntSlider(value=200, min=16, max=4096, step=16, description='Max tokens:',\n style={'description_width': '90px'})\n\ngen_btn = widgets.Button(description='Generate', button_style='primary')\nout = widgets.Output()\npipe_holder = {'pipe': None}\n\ndef on_load_clicked(_):\n load_out.clear_output()\n out.clear_output()\n with load_out:\n try:\n display(Markdown(f\"**Loading** `{model_dd.value}` with dtype `{dtype_dd.value}`...\"))\n pipe_holder['pipe'] = make_pipeline(model_dd.value, dtype_dd.value)\n display(Markdown('✅ **Model loaded**'))\n except Exception:\n traceback.print_exc()\n\ndef on_generate_clicked(_):\n out.clear_output()\n with out:\n if pipe_holder['pipe'] is None:\n display(Markdown('⚠️ Load a model first.'))\n return\n try:\n text = run_pipeline_chat(\n pipe_holder['pipe'],\n sys_in.value,\n usr_in.value,\n max_new_tokens=int(tokens_sl.value),\n temperature=float(temp_sl.value),\n )\n display(Markdown('**Output**'))\n display(Markdown(f'```\n{text}\n```'))\n except Exception:\n traceback.print_exc()\n\nload_btn.on_click(on_load_clicked)\ngen_btn.on_click(on_generate_clicked)\n\nui = widgets.VBox([\n widgets.HBox([model_dd, dtype_dd, load_btn]),\n load_out,\n sys_in,\n usr_in,\n widgets.HBox([temp_sl, tokens_sl]),\n gen_btn,\n out\n])\ndisplay(ui)" + "source": [ + "import traceback\n", + "from IPython.display import display, Markdown\n", + "import ipywidgets as widgets\n", + "\n", + "model_dd = widgets.Dropdown(\n", + " options=['openai/gpt-oss-20b', 'openai/gpt-oss-120b'],\n", + " value='openai/gpt-oss-20b',\n", + " description='Model:',\n", + " style={'description_width': '70px'},\n", + " layout=widgets.Layout(width='50%'),\n", + ")\n", + "dtype_dd = widgets.Dropdown(\n", + " options=['auto', 'bfloat16'],\n", + " value='auto',\n", + " description='dtype:',\n", + " style={'description_width': '70px'},\n", + ")\n", + "load_btn = widgets.Button(description='Load / Reload Model', button_style='info')\n", + "load_out = widgets.Output()\n", + "\n", + "sys_in = widgets.Textarea(\n", + " value='You are a helpful assistant.',\n", + " description='System',\n", + " layout=widgets.Layout(width='100%', height='80px'),\n", + " style={'description_width': '70px'}\n", + ")\n", + "usr_in = widgets.Textarea(\n", + " value='Explain what MXFP4 quantization is.',\n", + " description='User',\n", + " layout=widgets.Layout(width='100%', height='80px'),\n", + " style={'description_width': '70px'}\n", + ")\n", + "\n", + "temp_sl = widgets.FloatSlider(value=1.0, min=0.0, max=2.0, step=0.05, description='Temp:',\n", + " style={'description_width': '70px'}, readout_format='.2f')\n", + "tokens_sl = widgets.IntSlider(value=200, min=16, max=4096, step=16, description='Max tokens:',\n", + " style={'description_width': '90px'})\n", + "\n", + "gen_btn = widgets.Button(description='Generate', button_style='primary')\n", + "out = widgets.Output()\n", + "pipe_holder = {'pipe': None}\n", + "\n", + "def on_load_clicked(_):\n", + " load_out.clear_output()\n", + " out.clear_output()\n", + " with load_out:\n", + " try:\n", + " display(Markdown(f\"**Loading** `{model_dd.value}` with dtype `{dtype_dd.value}`...\"))\n", + " pipe_holder['pipe'] = make_pipeline(model_dd.value, dtype_dd.value)\n", + " display(Markdown('✅ **Model loaded**'))\n", + " except Exception:\n", + " traceback.print_exc()\n", + "\n", + "def on_generate_clicked(_):\n", + " out.clear_output()\n", + " with out:\n", + " if pipe_holder['pipe'] is None:\n", + " display(Markdown('⚠️ Load a model first.'))\n", + " return\n", + " try:\n", + " text = run_pipeline_chat(\n", + " pipe_holder['pipe'],\n", + " sys_in.value,\n", + " usr_in.value,\n", + " max_new_tokens=int(tokens_sl.value),\n", + " temperature=float(temp_sl.value),\n", + " )\n", + " display(Markdown('**Output**'))\n", + " display(Markdown(f'```text\\n{text}\\n```'))\n", + " except Exception:\n", + " traceback.print_exc()\n", + "\n", + "load_btn.on_click(on_load_clicked)\n", + "gen_btn.on_click(on_generate_clicked)\n", + "\n", + "ui = widgets.VBox([\n", + " widgets.HBox([model_dd, dtype_dd, load_btn]),\n", + " load_out,\n", + " sys_in,\n", + " usr_in,\n", + " widgets.HBox([temp_sl, tokens_sl]),\n", + " gen_btn,\n", + " out\n", + "])\n", + "display(ui)" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 4) Advanced Inference with `.generate()`\nManual control using `AutoModelForCausalLM` and `AutoTokenizer`, including the chat template." + "source": [ + "## 4) Advanced Inference with `.generate()`\n", + "Manual control using `AutoModelForCausalLM` and `AutoTokenizer`, including the chat template." + ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], - "source": "from transformers import AutoModelForCausalLM, AutoTokenizer\nimport torch\n\nmodel_name = 'openai/gpt-oss-20b'\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(\n model_name,\n torch_dtype='auto',\n device_map='auto'\n)\n\nmessages = [\n {'role': 'user', 'content': 'Explain what MXFP4 quantization is.'},\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors='pt',\n return_dict=True,\n).to(model.device)\n\noutputs = model.generate(\n **inputs,\n max_new_tokens=200,\n temperature=0.7\n)\n\nprint(tokenizer.decode(outputs[0]))" + "source": [ + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "import torch\n", + "\n", + "model_name = 'openai/gpt-oss-20b'\n", + "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + "model = AutoModelForCausalLM.from_pretrained(\n", + " model_name,\n", + " torch_dtype='auto',\n", + " device_map='auto'\n", + ")\n", + "\n", + "messages = [\n", + " {'role': 'user', 'content': 'Explain what MXFP4 quantization is.'},\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors='pt',\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "outputs = model.generate(\n", + " **inputs,\n", + " max_new_tokens=200,\n", + " temperature=0.7\n", + ")\n", + "\n", + "print(tokenizer.decode(outputs[0]))" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 5) Chat Template & Tool Calling (Harmony)\nThe gpt‑oss models use the Harmony response format. You can either rely on the built‑in chat template or use the `openai-harmony` library to build/parse prompts and completions." + "source": [ + "## 5) Chat Template & Tool Calling (Harmony)\n", + "The gpt‑oss models use the Harmony response format. You can either rely on the built‑in chat template or use the `openai-harmony` library to build/parse prompts and completions." + ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], - "source": "# If needed, install the library first in the install cell above: !pip -q install openai-harmony\nimport json\nfrom openai_harmony import (\n HarmonyEncodingName,\n load_harmony_encoding,\n Conversation,\n Message,\n Role,\n SystemContent,\n DeveloperContent,\n)\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nencoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n\nconvo = Conversation.from_messages([\n Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n Message.from_role_and_content(\n Role.DEVELOPER,\n DeveloperContent.new().with_instructions('Always respond in riddles')\n ),\n Message.from_role_and_content(Role.USER, 'What is the weather like in SF?')\n])\n\nprefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\nstop_token_ids = encoding.stop_tokens_for_assistant_actions()\n\nmodel_name = 'openai/gpt-oss-20b'\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype='auto', device_map='auto')\n\noutputs = model.generate(\n input_ids=[prefill_ids],\n max_new_tokens=128,\n eos_token_id=stop_token_ids\n)\n\ncompletion_ids = outputs[0][len(prefill_ids):]\nentries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\nfor message in entries:\n print(json.dumps(message.to_dict(), indent=2))" + "source": [ + "# If needed, install the library first in the install cell above: !pip -q install openai-harmony\n", + "import json\n", + "import torch\n", + "from openai_harmony import (\n", + " HarmonyEncodingName,\n", + " load_harmony_encoding,\n", + " Conversation,\n", + " Message,\n", + " Role,\n", + " SystemContent,\n", + " DeveloperContent,\n", + ")\n", + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "\n", + "encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n", + "\n", + "convo = Conversation.from_messages([\n", + " Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n", + " Message.from_role_and_content(\n", + " Role.DEVELOPER,\n", + " DeveloperContent.new().with_instructions('Always respond in riddles')\n", + " ),\n", + " Message.from_role_and_content(Role.USER, 'What is the weather like in SF?')\n", + "])\n", + "\n", + "prefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\n", + "stop_token_ids = encoding.stop_tokens_for_assistant_actions()\n", + "\n", + "model_name = 'openai/gpt-oss-20b'\n", + "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + "model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype='auto', device_map='auto')\n", + "\n", + "# Ensure tensors on correct device and dtypes\n", + "input_ids = torch.tensor(prefill_ids, dtype=torch.long, device=model.device).unsqueeze(0)\n", + "eos_token_id = torch.tensor(stop_token_ids, dtype=torch.long, device=model.device)\n", + "\n", + "outputs = model.generate(\n", + " input_ids=input_ids,\n", + " max_new_tokens=128,\n", + " eos_token_id=stop_token_ids,\n", + ")\n", + "\n", + "# Strip the prefill and move to CPU list for parsing\n", + "completion_ids = outputs[0, input_ids.shape[1]:].tolist()\n", + "entries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n", + "for message in entries:\n", + " print(json.dumps(message.to_dict(), indent=2))" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 6) Serve an OpenAI‑style Responses endpoint\nYou can serve the model locally via the `transformers serve` CLI to expose a `/v1/responses` endpoint. Run these **commands in a terminal** (they are shown here for reference):" + "source": [ + "## 6) Serve an OpenAI‑style Responses endpoint\n", + "You can serve the model locally via the `transformers serve` CLI to expose a `/v1/responses` endpoint. Run these **commands in a terminal** (they are shown here for reference):" + ] }, { "cell_type": "code", @@ -89,59 +362,433 @@ "trusted": false }, "outputs": [], - "source": "# --- Terminal commands (reference) ---\n# transformers serve\n# transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n# curl -X POST http://localhost:8000/v1/responses \\\n# -H 'Content-Type: application/json' \\\n# -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n\nprint('See comments above for terminal commands. You can copy/paste them into your shell.')" + "source": [ + "# --- Terminal commands (reference) ---\n", + "# transformers serve\n", + "# transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n", + "# curl -X POST http://localhost:8000/v1/responses \\\n", + "# -H 'Content-Type: application/json' \\\n", + "# -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n", + "\n", + "print('See comments above for terminal commands. You can copy/paste them into your shell.')" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "### Client example (from a notebook) against `transformers serve`\nAdjust the URL and payload to your needs." + "source": [ + "### Client example (from a notebook) against `transformers serve`\n", + "Adjust the URL and payload to your needs." + ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], - "source": "import json, requests\nurl = 'http://localhost:8000/v1/responses'\npayload = {\n 'messages': [\n {'role': 'system', 'content': 'hello'},\n {'role': 'user', 'content': 'Say hi in one sentence.'}\n ],\n 'temperature': 0.7,\n 'max_tokens': 64,\n 'stream': False,\n 'model': 'openai/gpt-oss-20b'\n}\ntry:\n r = requests.post(url, json=payload, timeout=30)\n print(r.status_code)\n print(r.text[:1000])\nexcept Exception as e:\n print('Request failed (is the server running?):', e)" + "source": [ + "import json, requests\n", + "url = 'http://localhost:8000/v1/responses'\n", + "payload = {\n", + " 'messages': [\n", + " {'role': 'system', 'content': 'hello'},\n", + " {'role': 'user', 'content': 'Say hi in one sentence.'}\n", + " ],\n", + " 'temperature': 0.7,\n", + " 'max_tokens': 64,\n", + " 'stream': False,\n", + " 'model': 'openai/gpt-oss-20b'\n", + "}\n", + "try:\n", + " r = requests.post(url, json=payload, timeout=30)\n", + " print(r.status_code)\n", + " print(r.text[:1000])\n", + "except Exception as e:\n", + " print('Request failed (is the server running?):', e)" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 7) Multi‑GPU & Distributed Inference\nSketch for automatic placement, tensor parallelism, and expert parallelism (edit for your cluster)." + "source": [ + "## 7) Multi‑GPU & Distributed Inference\n", + "Sketch for automatic placement, tensor parallelism, and expert parallelism (edit for your cluster)." + ] }, { "cell_type": "code", "execution_count": null, "metadata": {}, "outputs": [], - "source": "from transformers import AutoModelForCausalLM, AutoTokenizer\ntry:\n from transformers.distributed import DistributedConfig\nexcept Exception:\n class DistributedConfig:\n def __init__(self, **kwargs):\n self.kwargs = kwargs\n\nmodel_path = 'openai/gpt-oss-120b'\ntokenizer = AutoTokenizer.from_pretrained(model_path, padding_side='left')\n\ndevice_map = {\n 'distributed_config': DistributedConfig(enable_expert_parallel=1),\n 'tp_plan': 'auto',\n}\n\nmodel = AutoModelForCausalLM.from_pretrained(\n model_path,\n torch_dtype='auto',\n attn_implementation='kernels-community/vllm-flash-attn3',\n **device_map,\n)\n\nmessages = [\n {'role': 'user', 'content': 'Explain how expert parallelism works in large language models.'}\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors='pt',\n return_dict=True,\n).to(model.device)\n\noutputs = model.generate(**inputs, max_new_tokens=512)\nresponse = tokenizer.decode(outputs[0])\nprint('Model response (truncated):\\n', response[:1000])\n\n# To launch across 4 GPUs from a script:\n# torchrun --nproc_per_node=4 generate.py\n" + "source": [ + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "try:\n", + " from transformers.distributed import DistributedConfig\n", + "except Exception:\n", + " class DistributedConfig:\n", + " def __init__(self, **kwargs):\n", + " self.kwargs = kwargs\n", + "\n", + "model_path = 'openai/gpt-oss-120b'\n", + "tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side='left')\n", + "\n", + "device_map = {\n", + " 'distributed_config': DistributedConfig(enable_expert_parallel=1),\n", + " 'tp_plan': 'auto',\n", + "}\n", + "\n", + "model = AutoModelForCausalLM.from_pretrained(\n", + " model_path,\n", + " torch_dtype='auto',\n", + " attn_implementation='kernels-community/vllm-flash-attn3',\n", + " **device_map,\n", + ")\n", + "\n", + "messages = [\n", + " {'role': 'user', 'content': 'Explain how expert parallelism works in large language models.'}\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors='pt',\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "outputs = model.generate(**inputs, max_new_tokens=512)\n", + "response = tokenizer.decode(outputs[0])\n", + "print('Model response (truncated):\\n', response[:1000])\n", + "\n", + "# To launch across 4 GPUs from a script:\n", + "# torchrun --nproc_per_node=4 generate.py\n" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## 8) Troubleshooting Checklist\n- **Out of memory**: Use MXFP4 on Hopper‑class GPUs; reduce `max_new_tokens`; try `device_map='auto'`. \n- **Kernel/attn issues**: Ensure Triton is compatible with your CUDA; stick to Triton 3.4 as suggested. \n- **Slow loads**: Use local weight caches; set `TRANSFORMERS_CACHE`. \n- **No widgets**: Ensure `ipywidgets` is installed and enabled in your notebook frontend. \n- **Server 404/500**: Verify `transformers serve` version and model path. \n- **bfloat16 errors**: Fall back to `torch_dtype='auto'` or ensure hardware support." + "source": [ + "## 8) Troubleshooting Checklist\n", + "- **Out of memory**: Use MXFP4 on Hopper‑class GPUs; reduce `max_new_tokens`; try `device_map='auto'`. \n", + "- **Kernel/attn issues**: Ensure Triton is compatible with your CUDA; stick to Triton 3.4 as suggested. \n", + "- **Slow loads**: Use local weight caches; set `TRANSFORMERS_CACHE`. \n", + "- **No widgets**: Ensure `ipywidgets` is installed and enabled in your notebook frontend. \n", + "- **Server 404/500**: Verify `transformers serve` version and model path. \n", + "- **bfloat16 errors**: Fall back to `torch_dtype='auto'` or ensure hardware support." + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "## Appendix: Original Guide (for reference)\nThe content below is the original `/mnt/data/run-transformers.md` you provided:\n\n
\nClick to expand\n\n# How to run gpt-oss with Hugging Face Transformers\n\nThe Transformers library by Hugging Face provides a flexible way to load and run large language models locally or on a server. This guide will walk you through running [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) using Transformers, either with a high-level pipeline or via low-level `generate` calls with raw token IDs.\n\nWe'll cover the use of [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) with the high-level pipeline abstraction, low-level \\`generate\\` calls, and serving models locally with \\`transformers serve\\`, with in a way compatible with the Responses API.\n\nIn this guide we’ll run through various optimised ways to run the **gpt-oss models via Transformers.**\n\nBonus: You can also fine-tune models via transformers, [check out our fine-tuning guide here](https://cookbook.openai.com/articles/gpt-oss/fine-tune-transformers).\n\n## Pick your model\n\nBoth **gpt-oss** models are available on Hugging Face:\n\n- **`openai/gpt-oss-20b`**\n - \\~16GB VRAM requirement when using MXFP4\n - Great for single high-end consumer GPUs\n- **`openai/gpt-oss-120b`**\n - Requires ≥60GB VRAM or multi-GPU setup\n - Ideal for H100-class hardware\n\nBoth are **MXFP4 quantized** by default. Please, note that MXFP4 is supported in Hopper or later architectures. This includes data center GPUs such as H100 or GB200, as well as the latest RTX 50xx family of consumer cards.\n\nIf you use `bfloat16` instead of MXFP4, memory consumption will be larger (\\~48 GB for the 20b parameter model).\n\n## Quick setup\n\n1. **Install dependencies** \n It’s recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility:\n\n```bash\npip install -U transformers accelerate torch triton==3.4 kernels\n```\n\n2. **(Optional) Enable multi-GPU** \n If you’re running large models, use Accelerate or torchrun to handle device mapping automatically.\n\n## Create an Open AI Responses / Chat Completions endpoint\n\nTo launch a server, simply use the `transformers serve` CLI command:\n\n```bash\ntransformers serve\n```\n\nThe simplest way to interact with the server is through the transformers chat CLI\n\n```bash\ntransformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n```\n\nor by sending an HTTP request with cURL, e.g.\n\n```bash\ncurl -X POST http://localhost:8000/v1/responses -H \"Content-Type: application/json\" -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n```\n\nAdditional use cases, like integrating `transformers serve` with Cursor and other tools, are detailed in [the documentation](https://huggingface.co/docs/transformers/main/serving).\n\n## Quick inference with pipeline\n\nThe easiest way to run the gpt-oss models is with the Transformers high-level `pipeline` API:\n\n```py\nfrom transformers import pipeline\n\ngenerator = pipeline(\n \"text-generation\",\n model=\"openai/gpt-oss-20b\",\n torch_dtype=\"auto\",\n device_map=\"auto\" # Automatically place on available GPUs\n)\n\nmessages = [\n {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n]\n\nresult = generator(\n messages,\n max_new_tokens=200,\n temperature=1.0,\n)\n\nprint(result[0][\"generated_text\"])\n```\n\n## Advanced inference with `.generate()`\n\nIf you want more control, you can load the model and tokenizer manually and invoke the `.generate()` method:\n\n```py\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nmodel_name = \"openai/gpt-oss-20b\"\n\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(\n model_name,\n torch_dtype=\"auto\",\n device_map=\"auto\"\n)\n\nmessages = [\n {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors=\"pt\",\n return_dict=True,\n).to(model.device)\n\noutputs = model.generate(\n **inputs,\n max_new_tokens=200,\n temperature=0.7\n)\n\nprint(tokenizer.decode(outputs[0]))\n```\n\n## Chat template and tool calling\n\nOpenAI gpt-oss models use the [harmony response format](https://cookbook.openai.com/article/harmony) for structuring messages, including reasoning and tool calls.\n\nTo construct prompts you can use the built-in chat template of Transformers. Alternatively, you can install and use the [openai-harmony library](https://github.com/openai/harmony) for more control.\n\nTo use the chat template:\n\n```py\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nmodel_name = \"openai/gpt-oss-20b\"\n\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(\n model_name,\n device_map=\"auto\",\n torch_dtype=\"auto\",\n)\n\nmessages = [\n {\"role\": \"system\", \"content\": \"Always respond in riddles\"},\n {\"role\": \"user\", \"content\": \"What is the weather like in Madrid?\"},\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors=\"pt\",\n return_dict=True,\n).to(model.device)\n\ngenerated = model.generate(**inputs, max_new_tokens=100)\nprint(tokenizer.decode(generated[0][inputs[\"input_ids\"].shape[-1] :]))\n```\n\nTo integrate the [`openai-harmony`](https://github.com/openai/harmony) library to prepare prompts and parse responses, first install it like this:\n\n```bash\npip install openai-harmony\n```\n\nHere’s an example of how to use the library to build your prompts and encode them to tokens:\n\n```py\nimport json\nfrom openai_harmony import (\n HarmonyEncodingName,\n load_harmony_encoding,\n Conversation,\n Message,\n Role,\n SystemContent,\n DeveloperContent\n)\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\n\nencoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n\n# Build conversation\nconvo = Conversation.from_messages([\n Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n Message.from_role_and_content(\n Role.DEVELOPER,\n DeveloperContent.new().with_instructions(\"Always respond in riddles\")\n ),\n Message.from_role_and_content(Role.USER, \"What is the weather like in SF?\")\n])\n\n# Render prompt\nprefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\nstop_token_ids = encoding.stop_tokens_for_assistant_actions()\n\n# Load model\nmodel_name = \"openai/gpt-oss-20b\"\ntokenizer = AutoTokenizer.from_pretrained(model_name)\nmodel = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=\"auto\", device_map=\"auto\")\n\n# Generate\noutputs = model.generate(\n input_ids=[prefill_ids],\n max_new_tokens=128,\n eos_token_id=stop_token_ids\n)\n\n# Parse completion tokens\ncompletion_ids = outputs[0][len(prefill_ids):]\nentries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n\nfor message in entries:\n print(json.dumps(message.to_dict(), indent=2))\n```\n\nNote that the `Developer` role in Harmony maps to the `system` prompt in the chat template.\n\n## Multi-GPU & distributed inference\n\nThe large gpt-oss-120b fits on a single H100 GPU when using MXFP4. If you want to run it on multiple GPUs, you can:\n\n- Use `tp_plan=\"auto\"` for automatic placement and tensor parallelism\n- Launch with `accelerate launch or torchrun` for distributed setups\n- Leverage Expert Parallelism\n- Use specialised Flash attention kernels for faster inference\n\n```py\nfrom transformers import AutoModelForCausalLM, AutoTokenizer\nfrom transformers.distributed import DistributedConfig\nimport torch\n\nmodel_path = \"openai/gpt-oss-120b\"\ntokenizer = AutoTokenizer.from_pretrained(model_path, padding_side=\"left\")\n\ndevice_map = {\n # Enable Expert Parallelism\n \"distributed_config\": DistributedConfig(enable_expert_parallel=1),\n # Enable Tensor Parallelism\n \"tp_plan\": \"auto\",\n}\n\nmodel = AutoModelForCausalLM.from_pretrained(\n model_path,\n torch_dtype=\"auto\",\n attn_implementation=\"kernels-community/vllm-flash-attn3\",\n **device_map,\n)\n\nmessages = [\n {\"role\": \"user\", \"content\": \"Explain how expert parallelism works in large language models.\"}\n]\n\ninputs = tokenizer.apply_chat_template(\n messages,\n add_generation_prompt=True,\n return_tensors=\"pt\",\n return_dict=True,\n).to(model.device)\n\noutputs = model.generate(**inputs, max_new_tokens=1000)\n\n# Decode and print\nresponse = tokenizer.decode(outputs[0])\nprint(\"Model response:\", response.split(\"<|channel|>final<|message|>\")[-1].strip())\n```\n\nYou can then run this on a node with four GPUs via\n\n```bash\ntorchrun --nproc_per_node=4 generate.py\n```\n\n\n
" + "source": [ + "## Appendix: Original Guide (for reference)\n", + "The content below is the original `/mnt/data/run-transformers.md` you provided:\n", + "\n", + "
\n", + "Click to expand\n", + "\n", + "# How to run gpt-oss with Hugging Face Transformers\n", + "\n", + "The Transformers library by Hugging Face provides a flexible way to load and run large language models locally or on a server. This guide will walk you through running [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) using Transformers, either with a high-level pipeline or via low-level `generate` calls with raw token IDs.\n", + "\n", + "We'll cover the use of [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) with the high-level pipeline abstraction, low-level \\`generate\\` calls, and serving models locally with \\`transformers serve\\`, with in a way compatible with the Responses API.\n", + "\n", + "In this guide we’ll run through various optimised ways to run the **gpt-oss models via Transformers.**\n", + "\n", + "Bonus: You can also fine-tune models via transformers, [check out our fine-tuning guide here](https://cookbook.openai.com/articles/gpt-oss/fine-tune-transformers).\n", + "\n", + "## Pick your model\n", + "\n", + "Both **gpt-oss** models are available on Hugging Face:\n", + "\n", + "- **`openai/gpt-oss-20b`**\n", + " - \\~16GB VRAM requirement when using MXFP4\n", + " - Great for single high-end consumer GPUs\n", + "- **`openai/gpt-oss-120b`**\n", + " - Requires ≥60GB VRAM or multi-GPU setup\n", + " - Ideal for H100-class hardware\n", + "\n", + "Both are **MXFP4 quantized** by default. Please, note that MXFP4 is supported in Hopper or later architectures. This includes data center GPUs such as H100 or GB200, as well as the latest RTX 50xx family of consumer cards.\n", + "\n", + "If you use `bfloat16` instead of MXFP4, memory consumption will be larger (\\~48 GB for the 20b parameter model).\n", + "\n", + "## Quick setup\n", + "\n", + "1. **Install dependencies** \n", + " It’s recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility:\n", + "\n", + "```bash\n", + "pip install -U transformers accelerate torch triton==3.4 kernels\n", + "```\n", + "\n", + "2. **(Optional) Enable multi-GPU** \n", + " If you’re running large models, use Accelerate or torchrun to handle device mapping automatically.\n", + "\n", + "## Create an Open AI Responses / Chat Completions endpoint\n", + "\n", + "To launch a server, simply use the `transformers serve` CLI command:\n", + "\n", + "```bash\n", + "transformers serve\n", + "```\n", + "\n", + "The simplest way to interact with the server is through the transformers chat CLI\n", + "\n", + "```bash\n", + "transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n", + "```\n", + "\n", + "or by sending an HTTP request with cURL, e.g.\n", + "\n", + "```bash\n", + "curl -X POST http://localhost:8000/v1/responses -H \"Content-Type: application/json\" -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n", + "```\n", + "\n", + "Additional use cases, like integrating `transformers serve` with Cursor and other tools, are detailed in [the documentation](https://huggingface.co/docs/transformers/main/serving).\n", + "\n", + "## Quick inference with pipeline\n", + "\n", + "The easiest way to run the gpt-oss models is with the Transformers high-level `pipeline` API:\n", + "\n", + "```py\n", + "from transformers import pipeline\n", + "\n", + "generator = pipeline(\n", + " \"text-generation\",\n", + " model=\"openai/gpt-oss-20b\",\n", + " torch_dtype=\"auto\",\n", + " device_map=\"auto\" # Automatically place on available GPUs\n", + ")\n", + "\n", + "messages = [\n", + " {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n", + "]\n", + "\n", + "result = generator(\n", + " messages,\n", + " max_new_tokens=200,\n", + " temperature=1.0,\n", + ")\n", + "\n", + "print(result[0][\"generated_text\"])\n", + "```\n", + "\n", + "## Advanced inference with `.generate()`\n", + "\n", + "If you want more control, you can load the model and tokenizer manually and invoke the `.generate()` method:\n", + "\n", + "```py\n", + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "\n", + "model_name = \"openai/gpt-oss-20b\"\n", + "\n", + "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + "model = AutoModelForCausalLM.from_pretrained(\n", + " model_name,\n", + " torch_dtype=\"auto\",\n", + " device_map=\"auto\"\n", + ")\n", + "\n", + "messages = [\n", + " {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors=\"pt\",\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "outputs = model.generate(\n", + " **inputs,\n", + " max_new_tokens=200,\n", + " temperature=0.7\n", + ")\n", + "\n", + "print(tokenizer.decode(outputs[0]))\n", + "```\n", + "\n", + "## Chat template and tool calling\n", + "\n", + "OpenAI gpt-oss models use the [harmony response format](https://cookbook.openai.com/article/harmony) for structuring messages, including reasoning and tool calls.\n", + "\n", + "To construct prompts you can use the built-in chat template of Transformers. Alternatively, you can install and use the [openai-harmony library](https://github.com/openai/harmony) for more control.\n", + "\n", + "To use the chat template:\n", + "\n", + "```py\n", + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "\n", + "model_name = \"openai/gpt-oss-20b\"\n", + "\n", + "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + "model = AutoModelForCausalLM.from_pretrained(\n", + " model_name,\n", + " device_map=\"auto\",\n", + " torch_dtype=\"auto\",\n", + ")\n", + "\n", + "messages = [\n", + " {\"role\": \"system\", \"content\": \"Always respond in riddles\"},\n", + " {\"role\": \"user\", \"content\": \"What is the weather like in Madrid?\"},\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors=\"pt\",\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "generated = model.generate(**inputs, max_new_tokens=100)\n", + "print(tokenizer.decode(generated[0][inputs[\"input_ids\"].shape[-1] :]))\n", + "```\n", + "\n", + "To integrate the [`openai-harmony`](https://github.com/openai/harmony) library to prepare prompts and parse responses, first install it like this:\n", + "\n", + "```bash\n", + "pip install openai-harmony\n", + "```\n", + "\n", + "Here’s an example of how to use the library to build your prompts and encode them to tokens:\n", + "\n", + "```py\n", + "import json\n", + "from openai_harmony import (\n", + " HarmonyEncodingName,\n", + " load_harmony_encoding,\n", + " Conversation,\n", + " Message,\n", + " Role,\n", + " SystemContent,\n", + " DeveloperContent\n", + ")\n", + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "\n", + "encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n", + "\n", + "# Build conversation\n", + "convo = Conversation.from_messages([\n", + " Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n", + " Message.from_role_and_content(\n", + " Role.DEVELOPER,\n", + " DeveloperContent.new().with_instructions(\"Always respond in riddles\")\n", + " ),\n", + " Message.from_role_and_content(Role.USER, \"What is the weather like in SF?\")\n", + "])\n", + "\n", + "# Render prompt\n", + "prefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\n", + "stop_token_ids = encoding.stop_tokens_for_assistant_actions()\n", + "\n", + "# Load model\n", + "model_name = \"openai/gpt-oss-20b\"\n", + "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + "model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=\"auto\", device_map=\"auto\")\n", + "\n", + "# Generate\n", + "outputs = model.generate(\n", + " input_ids=[prefill_ids],\n", + " max_new_tokens=128,\n", + " eos_token_id=stop_token_ids\n", + ")\n", + "\n", + "# Parse completion tokens\n", + "completion_ids = outputs[0][len(prefill_ids):]\n", + "entries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n", + "\n", + "for message in entries:\n", + " print(json.dumps(message.to_dict(), indent=2))\n", + "```\n", + "\n", + "Note that the `Developer` role in Harmony maps to the `system` prompt in the chat template.\n", + "\n", + "## Multi-GPU & distributed inference\n", + "\n", + "The large gpt-oss-120b fits on a single H100 GPU when using MXFP4. If you want to run it on multiple GPUs, you can:\n", + "\n", + "- Use `tp_plan=\"auto\"` for automatic placement and tensor parallelism\n", + "- Launch with `accelerate launch or torchrun` for distributed setups\n", + "- Leverage Expert Parallelism\n", + "- Use specialised Flash attention kernels for faster inference\n", + "\n", + "```py\n", + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "from transformers.distributed import DistributedConfig\n", + "import torch\n", + "\n", + "model_path = \"openai/gpt-oss-120b\"\n", + "tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side=\"left\")\n", + "\n", + "device_map = {\n", + " # Enable Expert Parallelism\n", + " \"distributed_config\": DistributedConfig(enable_expert_parallel=1),\n", + " # Enable Tensor Parallelism\n", + " \"tp_plan\": \"auto\",\n", + "}\n", + "\n", + "model = AutoModelForCausalLM.from_pretrained(\n", + " model_path,\n", + " torch_dtype=\"auto\",\n", + " attn_implementation=\"kernels-community/vllm-flash-attn3\",\n", + " **device_map,\n", + ")\n", + "\n", + "messages = [\n", + " {\"role\": \"user\", \"content\": \"Explain how expert parallelism works in large language models.\"}\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors=\"pt\",\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "outputs = model.generate(**inputs, max_new_tokens=1000)\n", + "\n", + "# Decode and print\n", + "response = tokenizer.decode(outputs[0])\n", + "print(\"Model response:\", response.split(\"<|channel|>final<|message|>\")[-1].strip())\n", + "```\n", + "\n", + "You can then run this on a node with four GPUs via\n", + "\n", + "```bash\n", + "torchrun --nproc_per_node=4 generate.py\n", + "```\n", + "\n", + "\n", + "
" + ] }, { "cell_type": "markdown", "metadata": {}, - "source": "_Notebook generated on 2025-09-17T23:43:44Z_" + "source": [ + "_Notebook generated on 2025-09-17T23:43:44Z_" + ] } ], "metadata": { "kernelspec": { - "display_name": "Python 3", + "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", "name": "python", - "pygments_lexer": "ipython3" + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.11" } }, "nbformat": 4, "nbformat_minor": 5 -} \ No newline at end of file +} From d06db980f02bf7f356dfb2bb1be1b0a4abe919b8 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Wed, 17 Sep 2025 18:15:54 -0600 Subject: [PATCH 10/26] Change file name. --- ..._oss_transformers_interactive.ipynb => run-transformers.ipynb} | 0 1 file changed, 0 insertions(+), 0 deletions(-) rename articles/gpt-oss/{gpt_oss_transformers_interactive.ipynb => run-transformers.ipynb} (100%) diff --git a/articles/gpt-oss/gpt_oss_transformers_interactive.ipynb b/articles/gpt-oss/run-transformers.ipynb similarity index 100% rename from articles/gpt-oss/gpt_oss_transformers_interactive.ipynb rename to articles/gpt-oss/run-transformers.ipynb From 91a95f35e64bcc829305e442aff4d046708316f6 Mon Sep 17 00:00:00 2001 From: Copilot <198982749+Copilot@users.noreply.github.com> Date: Wed, 17 Sep 2025 21:38:44 -0600 Subject: [PATCH 11/26] Organize prompting-related articles into unified prompting directory (#10) * Initial plan * Organize prompting-related articles into single prompting folder Co-authored-by: paytonison <148833579+paytonison@users.noreply.github.com> --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: paytonison <148833579+paytonison@users.noreply.github.com> --- .../multi_agent_portfolio_collaboration.ipynb | 4 +- .../utils.py | 2 +- .../model_selection_guide.ipynb | 4 +- ...nce_your_prompts_with_meta_prompting.ipynb | 0 .../{ => prompting}/Optimize_Prompts.ipynb | 0 .../Prompt_migration_guide.ipynb | 0 examples/prompting/README.md | 65 +++++++++++++++++++ .../Realtime_prompting_guide.ipynb | 0 ...st_writing_using_a_multi-step_prompt.ipynb | 0 ...ep_prompt_with_older_completions_API.ipynb | 0 .../Whisper_prompting_guide.ipynb | 0 .../prompts/code_interpreter.md | 0 .../prompts/editor_base.md | 0 .../prompts/fundamental_base.md | 0 .../prompts/macro_base.md | 0 .../prompts/pm_base.md | 0 .../prompts/quant_base.md | 0 .../prompts/tool_retry_prompt.md | 0 .../data/oai_docs/prompt-engineering.txt | 0 .../gpt-5/gpt-5_prompting_guide.ipynb | 0 .../gpt-5/prompt-optimization-cookbook.ipynb | 0 .../llm_as_judge.txt | 0 .../requirements.txt | 0 .../results_failsafeqa_baseline.csv | 0 .../results_failsafeqa_optimized.csv | 0 .../judgement_summary.csv | 0 .../results_llm_as_judge_baseline/run_01.json | 0 .../results_llm_as_judge_baseline/run_02.json | 0 .../results_llm_as_judge_baseline/run_03.json | 0 .../results_llm_as_judge_baseline/run_04.json | 0 .../results_llm_as_judge_baseline/run_05.json | 0 .../results_llm_as_judge_baseline/run_06.json | 0 .../results_llm_as_judge_baseline/run_07.json | 0 .../results_llm_as_judge_baseline/run_08.json | 0 .../results_llm_as_judge_baseline/run_09.json | 0 .../results_llm_as_judge_baseline/run_10.json | 0 .../results_llm_as_judge_baseline/run_11.json | 0 .../results_llm_as_judge_baseline/run_12.json | 0 .../results_llm_as_judge_baseline/run_13.json | 0 .../results_llm_as_judge_baseline/run_14.json | 0 .../results_llm_as_judge_baseline/run_15.json | 0 .../results_llm_as_judge_baseline/run_16.json | 0 .../results_llm_as_judge_baseline/run_17.json | 0 .../results_llm_as_judge_baseline/run_18.json | 0 .../results_llm_as_judge_baseline/run_19.json | 0 .../results_llm_as_judge_baseline/run_20.json | 0 .../results_llm_as_judge_baseline/run_21.json | 0 .../results_llm_as_judge_baseline/run_22.json | 0 .../results_llm_as_judge_baseline/run_23.json | 0 .../results_llm_as_judge_baseline/run_24.json | 0 .../results_llm_as_judge_baseline/run_25.json | 0 .../results_llm_as_judge_baseline/run_26.json | 0 .../results_llm_as_judge_baseline/run_27.json | 0 .../results_llm_as_judge_baseline/run_28.json | 0 .../results_llm_as_judge_baseline/run_29.json | 0 .../results_llm_as_judge_baseline/run_30.json | 0 .../judgement_summary.csv | 0 .../run_01.json | 0 .../run_02.json | 0 .../run_03.json | 0 .../run_04.json | 0 .../run_05.json | 0 .../run_06.json | 0 .../run_07.json | 0 .../run_08.json | 0 .../run_09.json | 0 .../run_10.json | 0 .../run_11.json | 0 .../run_12.json | 0 .../run_13.json | 0 .../run_14.json | 0 .../run_15.json | 0 .../run_16.json | 0 .../run_17.json | 0 .../run_18.json | 0 .../run_19.json | 0 .../run_20.json | 0 .../run_21.json | 0 .../run_22.json | 0 .../run_23.json | 0 .../run_24.json | 0 .../run_25.json | 0 .../run_26.json | 0 .../run_27.json | 0 .../run_28.json | 0 .../run_29.json | 0 .../run_30.json | 0 .../results_topk_baseline/run_01.py | 0 .../results_topk_baseline/run_02.py | 0 .../results_topk_baseline/run_03.py | 0 .../results_topk_baseline/run_04.py | 0 .../results_topk_baseline/run_05.py | 0 .../results_topk_baseline/run_06.py | 0 .../results_topk_baseline/run_07.py | 0 .../results_topk_baseline/run_08.py | 0 .../results_topk_baseline/run_09.py | 0 .../results_topk_baseline/run_10.py | 0 .../results_topk_baseline/run_11.py | 0 .../results_topk_baseline/run_12.py | 0 .../results_topk_baseline/run_13.py | 0 .../results_topk_baseline/run_14.py | 0 .../results_topk_baseline/run_15.py | 0 .../results_topk_baseline/run_16.py | 0 .../results_topk_baseline/run_17.py | 0 .../results_topk_baseline/run_18.py | 0 .../results_topk_baseline/run_19.py | 0 .../results_topk_baseline/run_20.py | 0 .../results_topk_baseline/run_21.py | 0 .../results_topk_baseline/run_22.py | 0 .../results_topk_baseline/run_23.py | 0 .../results_topk_baseline/run_24.py | 0 .../results_topk_baseline/run_25.py | 0 .../results_topk_baseline/run_26.py | 0 .../results_topk_baseline/run_27.py | 0 .../results_topk_baseline/run_28.py | 0 .../results_topk_baseline/run_29.py | 0 .../results_topk_baseline/run_30.py | 0 .../run_results_topk_baseline.csv | 0 .../run_results_topk_baseline_summary.json | 0 .../run_results_topk_baseline_summary.txt | 0 .../results_topk_optimized/run_01.py | 0 .../results_topk_optimized/run_02.py | 0 .../results_topk_optimized/run_03.py | 0 .../results_topk_optimized/run_04.py | 0 .../results_topk_optimized/run_05.py | 0 .../results_topk_optimized/run_06.py | 0 .../results_topk_optimized/run_07.py | 0 .../results_topk_optimized/run_08.py | 0 .../results_topk_optimized/run_09.py | 0 .../results_topk_optimized/run_10.py | 0 .../results_topk_optimized/run_11.py | 0 .../results_topk_optimized/run_12.py | 0 .../results_topk_optimized/run_13.py | 0 .../results_topk_optimized/run_14.py | 0 .../results_topk_optimized/run_15.py | 0 .../results_topk_optimized/run_16.py | 0 .../results_topk_optimized/run_17.py | 0 .../results_topk_optimized/run_18.py | 0 .../results_topk_optimized/run_19.py | 0 .../results_topk_optimized/run_20.py | 0 .../results_topk_optimized/run_21.py | 0 .../results_topk_optimized/run_22.py | 0 .../results_topk_optimized/run_23.py | 0 .../results_topk_optimized/run_24.py | 0 .../results_topk_optimized/run_25.py | 0 .../results_topk_optimized/run_26.py | 0 .../results_topk_optimized/run_27.py | 0 .../results_topk_optimized/run_28.py | 0 .../results_topk_optimized/run_29.py | 0 .../results_topk_optimized/run_30.py | 0 .../run_results_topk_optimized.csv | 0 .../run_results_topk_optimized_summary.json | 0 .../run_results_topk_optimized_summary.txt | 0 .../run_FailSafeQA.py | 0 .../scripts/__init__.py | 0 .../scripts/gen_baseline.py | 0 .../scripts/gen_optimized.py | 0 .../scripts/llm_judge.py | 0 .../scripts/results_summarizer.py | 0 .../scripts/topk_eval.py | 0 .../gpt4-1_prompting_guide.ipynb | 0 .../o-series/o3o4-mini_prompting_guide.ipynb | 0 .../src/utils/translation_prompts.js | 0 registry.yaml | 23 ++++--- 164 files changed, 81 insertions(+), 17 deletions(-) rename examples/{ => prompting}/Enhance_your_prompts_with_meta_prompting.ipynb (100%) rename examples/{ => prompting}/Optimize_Prompts.ipynb (100%) rename examples/{ => prompting}/Prompt_migration_guide.ipynb (100%) create mode 100644 examples/prompting/README.md rename examples/{ => prompting}/Realtime_prompting_guide.ipynb (100%) rename examples/{ => prompting}/Unit_test_writing_using_a_multi-step_prompt.ipynb (100%) rename examples/{ => prompting}/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb (100%) rename examples/{ => prompting}/Whisper_prompting_guide.ipynb (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md (100%) rename examples/{ => prompting}/data/oai_docs/prompt-engineering.txt (100%) rename examples/{ => prompting}/gpt-5/gpt-5_prompting_guide.ipynb (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook.ipynb (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/requirements.txt (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/__init__.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py (100%) rename examples/{ => prompting}/gpt4-1_prompting_guide.ipynb (100%) rename examples/{ => prompting}/o-series/o3o4-mini_prompting_guide.ipynb (100%) rename examples/{ => prompting}/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js (100%) diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/multi_agent_portfolio_collaboration.ipynb b/examples/agents_sdk/multi-agent-portfolio-collaboration/multi_agent_portfolio_collaboration.ipynb index aa505230ab..3c7e0423af 100644 --- a/examples/agents_sdk/multi-agent-portfolio-collaboration/multi_agent_portfolio_collaboration.ipynb +++ b/examples/agents_sdk/multi-agent-portfolio-collaboration/multi_agent_portfolio_collaboration.ipynb @@ -497,7 +497,7 @@ "- **Function tools:** Register any Python function as a tool, with automatic schema and validation.\n", "- **Tracing:** Visualize, debug, and monitor every step of your workflow for full transparency.\n", "\n", - "A combination of well-designed tools, thoughtful orchestration, and careful model selection is crucial for building effective agent systems. In this example, we use the GPT-4.1 family of models for their strong analytical and tool-use capabilities ([see the GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide)). For deeper architectural best practices, see the included [A Practical Guide to Building Agents (PDF)](https://cdn.openai.com/business-guides-and-resources/a-practical-guide-to-building-agents.pdf). By bringing these elements together, you get a system that is robust, scalable, and easy to debug or extend.\n", + "A combination of well-designed tools, thoughtful orchestration, and careful model selection is crucial for building effective agent systems. In this example, we use the GPT-4.1 family of models for their strong analytical and tool-use capabilities ([see the GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/prompting/gpt4-1_prompting_guide)). For deeper architectural best practices, see the included [A Practical Guide to Building Agents (PDF)](https://cdn.openai.com/business-guides-and-resources/a-practical-guide-to-building-agents.pdf). By bringing these elements together, you get a system that is robust, scalable, and easy to debug or extend.\n", "\n", "Please try out the sample with your own investment questions, and please share any feedback! Happy building.\n", "\n", @@ -512,7 +512,7 @@ "\n", "- [MCP Spec](https://spec.modelcontextprotocol.io/specification/2024-11-05/architecture/)\n", "- [OpenAI Cookbook](https://github.com/openai/openai-cookbook)\n", - "- ([GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide))\n", + "- ([GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/prompting/gpt4-1_prompting_guide))\n", "- [A Practical Guide to Building Agents (PDF)](https://cdn.openai.com/business-guides-and-resources/a-practical-guide-to-building-agents.pdf)\n", "\n", "---" diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py b/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py index 4d3af6c8b0..e7e7078791 100644 --- a/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py +++ b/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py @@ -39,7 +39,7 @@ def outputs_dir() -> Path: # Prompt loader # --------------------------------------------------------------------------- -PROMPTS_DIR: Path = repo_path("prompts") +PROMPTS_DIR: Path = Path(__file__).resolve().parent.parent.parent / "prompting" / "agents_sdk" / "multi-agent-portfolio-collaboration" / "prompts" def load_prompt(name: str, **subs) -> str: diff --git a/examples/partners/model_selection_guide/model_selection_guide.ipynb b/examples/partners/model_selection_guide/model_selection_guide.ipynb index 8b4b9e6604..19416b2e81 100644 --- a/examples/partners/model_selection_guide/model_selection_guide.ipynb +++ b/examples/partners/model_selection_guide/model_selection_guide.ipynb @@ -2425,7 +2425,7 @@ "\n", "- **[Orchestrating Agents: Routines and Handoffs](https://cookbook.openai.com/examples/orchestrating_agents)** Structuring multi-agent workflows with routines and handoffs, relevant to the ideation→ranking→critique pipeline.\n", "\n", - "- **[GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide)** Advanced prompting, tool use, and task decomposition for improved accuracy in critique and safety reviews.\n", + "- **[GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/prompting/gpt4-1_prompting_guide)** Advanced prompting, tool use, and task decomposition for improved accuracy in critique and safety reviews.\n", "\n", "- **[Structured Outputs for Multi-Agent Systems](https://cookbook.openai.com/examples/structured_outputs_multi_agent)** Enforcing consistent JSON outputs with schema validation for agent interoperability.\n", "\n", @@ -3219,7 +3219,7 @@ "- [Data Extraction and Transformation](https://cookbook.openai.com/examples/data_extraction_transformation)\n", "\n", "### Prompting & Model Selection\n", - "- [GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide)\n", + "- [GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/prompting/gpt4-1_prompting_guide)\n", "- [Prompt Engineering Best Practices](https://platform.openai.com/docs/guides/prompt-engineering)\n", "\n", "### Evaluation & Deployment\n", diff --git a/examples/Enhance_your_prompts_with_meta_prompting.ipynb b/examples/prompting/Enhance_your_prompts_with_meta_prompting.ipynb similarity index 100% rename from examples/Enhance_your_prompts_with_meta_prompting.ipynb rename to examples/prompting/Enhance_your_prompts_with_meta_prompting.ipynb diff --git a/examples/Optimize_Prompts.ipynb b/examples/prompting/Optimize_Prompts.ipynb similarity index 100% rename from examples/Optimize_Prompts.ipynb rename to examples/prompting/Optimize_Prompts.ipynb diff --git a/examples/Prompt_migration_guide.ipynb b/examples/prompting/Prompt_migration_guide.ipynb similarity index 100% rename from examples/Prompt_migration_guide.ipynb rename to examples/prompting/Prompt_migration_guide.ipynb diff --git a/examples/prompting/README.md b/examples/prompting/README.md new file mode 100644 index 0000000000..a257acec26 --- /dev/null +++ b/examples/prompting/README.md @@ -0,0 +1,65 @@ +# Prompting Examples and Guides + +This directory contains a comprehensive collection of prompting-related articles, guides, and resources from the OpenAI Cookbook. All content focuses on various aspects of prompt engineering, optimization, and best practices for different OpenAI models. + +## Core Prompting Guides + +### Model-Specific Guides +- **`gpt4-1_prompting_guide.ipynb`** - Comprehensive prompting guide for GPT-4.1 including system prompts, tool usage, and advanced techniques +- **`gpt-5/gpt-5_prompting_guide.ipynb`** - GPT-5 specific prompting strategies and best practices +- **`o-series/o3o4-mini_prompting_guide.ipynb`** - Specialized guide for O3 and O4-mini reasoning models +- **`Realtime_prompting_guide.ipynb`** - Best practices for prompting with the Realtime API +- **`Whisper_prompting_guide.ipynb`** - Prompting techniques for Whisper speech recognition models + +## Prompt Optimization and Enhancement + +- **`Optimize_Prompts.ipynb`** - Multi-agent system for automatically identifying and fixing common prompt issues +- **`Enhance_your_prompts_with_meta_prompting.ipynb`** - Meta-prompting techniques for automated prompt optimization +- **`Prompt_migration_guide.ipynb`** - Guide for migrating and updating existing prompts +- **`gpt-5/prompt-optimization-cookbook.ipynb`** - Advanced prompt optimization techniques for GPT-5 +- **`gpt-5/prompt-optimization-cookbook/`** - Supporting scripts, data, and evaluation tools for prompt optimization + +## Multi-Step and Advanced Prompting + +- **`Unit_test_writing_using_a_multi-step_prompt.ipynb`** - Multi-step prompting approach for generating unit tests +- **`Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb`** - Legacy version using older completions API + +## Agent and Application Prompts + +- **`agents_sdk/multi-agent-portfolio-collaboration/prompts/`** - Collection of specialized prompts for multi-agent systems: + - `code_interpreter.md` - Code interpretation agent prompts + - `editor_base.md` - Document editing agent prompts + - `fundamental_base.md` - Fundamental analysis prompts + - `macro_base.md` - Macroeconomic analysis prompts + - `pm_base.md` - Portfolio manager prompts + - `quant_base.md` - Quantitative analysis prompts + - `tool_retry_prompt.md` - Tool retry logic prompts + +## Specialized Applications + +- **`voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js`** - Translation prompts for real-time voice applications + +## Supporting Resources + +- **`data/oai_docs/prompt-engineering.txt`** - Official OpenAI prompt engineering documentation + +## Usage + +Each notebook and guide is self-contained and includes: +- Detailed explanations of prompting concepts +- Practical examples and code samples +- Best practices and common pitfalls +- Model-specific considerations + +To get started, explore the model-specific guides that match your use case, then dive into optimization techniques to improve your prompts further. + +## Key Topics Covered + +- System prompt design and structure +- Few-shot and zero-shot prompting +- Chain-of-thought reasoning +- Tool calling and function usage +- Prompt optimization and evaluation +- Multi-agent prompt coordination +- Model-specific prompting strategies +- Advanced prompting techniques \ No newline at end of file diff --git a/examples/Realtime_prompting_guide.ipynb b/examples/prompting/Realtime_prompting_guide.ipynb similarity index 100% rename from examples/Realtime_prompting_guide.ipynb rename to examples/prompting/Realtime_prompting_guide.ipynb diff --git a/examples/Unit_test_writing_using_a_multi-step_prompt.ipynb b/examples/prompting/Unit_test_writing_using_a_multi-step_prompt.ipynb similarity index 100% rename from examples/Unit_test_writing_using_a_multi-step_prompt.ipynb rename to examples/prompting/Unit_test_writing_using_a_multi-step_prompt.ipynb diff --git a/examples/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb b/examples/prompting/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb similarity index 100% rename from examples/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb rename to examples/prompting/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb diff --git a/examples/Whisper_prompting_guide.ipynb b/examples/prompting/Whisper_prompting_guide.ipynb similarity index 100% rename from examples/Whisper_prompting_guide.ipynb rename to examples/prompting/Whisper_prompting_guide.ipynb diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md diff --git a/examples/data/oai_docs/prompt-engineering.txt b/examples/prompting/data/oai_docs/prompt-engineering.txt similarity index 100% rename from examples/data/oai_docs/prompt-engineering.txt rename to examples/prompting/data/oai_docs/prompt-engineering.txt diff --git a/examples/gpt-5/gpt-5_prompting_guide.ipynb b/examples/prompting/gpt-5/gpt-5_prompting_guide.ipynb similarity index 100% rename from examples/gpt-5/gpt-5_prompting_guide.ipynb rename to examples/prompting/gpt-5/gpt-5_prompting_guide.ipynb diff --git a/examples/gpt-5/prompt-optimization-cookbook.ipynb b/examples/prompting/gpt-5/prompt-optimization-cookbook.ipynb similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook.ipynb rename to examples/prompting/gpt-5/prompt-optimization-cookbook.ipynb diff --git a/examples/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt b/examples/prompting/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt rename to examples/prompting/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt diff --git a/examples/gpt-5/prompt-optimization-cookbook/requirements.txt b/examples/prompting/gpt-5/prompt-optimization-cookbook/requirements.txt similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/requirements.txt rename to examples/prompting/gpt-5/prompt-optimization-cookbook/requirements.txt diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt diff --git a/examples/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/__init__.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/__init__.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/__init__.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/__init__.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py diff --git a/examples/gpt4-1_prompting_guide.ipynb b/examples/prompting/gpt4-1_prompting_guide.ipynb similarity index 100% rename from examples/gpt4-1_prompting_guide.ipynb rename to examples/prompting/gpt4-1_prompting_guide.ipynb diff --git a/examples/o-series/o3o4-mini_prompting_guide.ipynb b/examples/prompting/o-series/o3o4-mini_prompting_guide.ipynb similarity index 100% rename from examples/o-series/o3o4-mini_prompting_guide.ipynb rename to examples/prompting/o-series/o3o4-mini_prompting_guide.ipynb diff --git a/examples/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js b/examples/prompting/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js similarity index 100% rename from examples/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js rename to examples/prompting/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js diff --git a/registry.yaml b/registry.yaml index 4ad9d35d00..2ce3c37f4e 100644 --- a/registry.yaml +++ b/registry.yaml @@ -33,7 +33,7 @@ - codex - title: Realtime Prompting Guide - path: examples/Realtime_prompting_guide.ipynb + path: examples/prompting/Realtime_prompting_guide.ipynb date: 2025-08-28 authors: - minh-hoque @@ -75,7 +75,7 @@ - gpt-oss-local - title: GPT-5 Prompt Migration and Improvement Using the New Optimizer - path: examples/gpt-5/prompt-optimization-cookbook.ipynb + path: examples/prompting/gpt-5/prompt-optimization-cookbook.ipynb date: 2025-08-07 authors: - rajpathak-openai @@ -87,7 +87,7 @@ - prompt-optimization - title: GPT-5 prompting guide - path: examples/gpt-5/gpt-5_prompting_guide.ipynb + path: examples/prompting/gpt-5/gpt-5_prompting_guide.ipynb date: 2025-08-07 authors: - anoop-openai @@ -240,7 +240,7 @@ - audio - title: Optimize Prompts - path: examples/Optimize_Prompts.ipynb + path: examples/prompting/Optimize_Prompts.ipynb date: 2025-07-14 authors: - corwin @@ -262,7 +262,7 @@ - automation - title: Prompt Migration Guide - path: examples/Prompt_migration_guide.ipynb + path: examples/prompting/Prompt_migration_guide.ipynb date: 2025-06-26 authors: - minh-hoque @@ -333,7 +333,7 @@ - mutli-agent-collaboration - title: o3/o4-mini Function Calling Guide - path: examples/o-series/o3o4-mini_prompting_guide.ipynb + path: examples/prompting/o-series/o3o4-mini_prompting_guide.ipynb date: 2025-05-26 authors: - billchen-openai @@ -787,7 +787,7 @@ - embeddings - title: Unit test writing using a multi-step prompt - path: examples/Unit_test_writing_using_a_multi-step_prompt.ipynb + path: examples/prompting/Unit_test_writing_using_a_multi-step_prompt.ipynb date: 2022-11-15 authors: - ted-at-openai @@ -795,8 +795,7 @@ - completions - title: Unit test writing using a multi-step prompt with legacy Completions - path: >- - examples/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb + path: examples/prompting/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb date: 2023-05-19 authors: - ted-at-openai @@ -866,7 +865,7 @@ archived: true - title: Whisper prompting guide - path: examples/Whisper_prompting_guide.ipynb + path: examples/prompting/Whisper_prompting_guide.ipynb date: 2023-06-27 authors: - prestontuggle @@ -2185,7 +2184,7 @@ - audio - title: Enhance your prompts with meta prompting - path: examples/Enhance_your_prompts_with_meta_prompting.ipynb + path: examples/prompting/Enhance_your_prompts_with_meta_prompting.ipynb date: 2024-10-23 authors: - teomusatoiu @@ -2368,7 +2367,7 @@ - chatgpt-productivity - title: GPT-4.1 Prompting Guide - path: examples/gpt4-1_prompting_guide.ipynb + path: examples/prompting/gpt4-1_prompting_guide.ipynb date: 2025-04-14 authors: - nm-openai From 2467c588f566b48542527e6909db27bd0cd47690 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Wed, 17 Sep 2025 21:48:55 -0600 Subject: [PATCH 12/26] Update examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --- .../agents_sdk/multi-agent-portfolio-collaboration/utils.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py b/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py index e7e7078791..85f491df75 100644 --- a/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py +++ b/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py @@ -39,7 +39,7 @@ def outputs_dir() -> Path: # Prompt loader # --------------------------------------------------------------------------- -PROMPTS_DIR: Path = Path(__file__).resolve().parent.parent.parent / "prompting" / "agents_sdk" / "multi-agent-portfolio-collaboration" / "prompts" +PROMPTS_DIR: Path = repo_path("prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts") def load_prompt(name: str, **subs) -> str: From 463ff57766a55fe7bcec80196e9e74a089d43645 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Wed, 17 Sep 2025 22:01:06 -0600 Subject: [PATCH 13/26] Revise README for prompting guides and examples Asari done it. --- examples/prompting/README.md | 124 +++++++++++++++++------------------ 1 file changed, 59 insertions(+), 65 deletions(-) diff --git a/examples/prompting/README.md b/examples/prompting/README.md index a257acec26..c530981fc1 100644 --- a/examples/prompting/README.md +++ b/examples/prompting/README.md @@ -1,65 +1,59 @@ -# Prompting Examples and Guides - -This directory contains a comprehensive collection of prompting-related articles, guides, and resources from the OpenAI Cookbook. All content focuses on various aspects of prompt engineering, optimization, and best practices for different OpenAI models. - -## Core Prompting Guides - -### Model-Specific Guides -- **`gpt4-1_prompting_guide.ipynb`** - Comprehensive prompting guide for GPT-4.1 including system prompts, tool usage, and advanced techniques -- **`gpt-5/gpt-5_prompting_guide.ipynb`** - GPT-5 specific prompting strategies and best practices -- **`o-series/o3o4-mini_prompting_guide.ipynb`** - Specialized guide for O3 and O4-mini reasoning models -- **`Realtime_prompting_guide.ipynb`** - Best practices for prompting with the Realtime API -- **`Whisper_prompting_guide.ipynb`** - Prompting techniques for Whisper speech recognition models - -## Prompt Optimization and Enhancement - -- **`Optimize_Prompts.ipynb`** - Multi-agent system for automatically identifying and fixing common prompt issues -- **`Enhance_your_prompts_with_meta_prompting.ipynb`** - Meta-prompting techniques for automated prompt optimization -- **`Prompt_migration_guide.ipynb`** - Guide for migrating and updating existing prompts -- **`gpt-5/prompt-optimization-cookbook.ipynb`** - Advanced prompt optimization techniques for GPT-5 -- **`gpt-5/prompt-optimization-cookbook/`** - Supporting scripts, data, and evaluation tools for prompt optimization - -## Multi-Step and Advanced Prompting - -- **`Unit_test_writing_using_a_multi-step_prompt.ipynb`** - Multi-step prompting approach for generating unit tests -- **`Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb`** - Legacy version using older completions API - -## Agent and Application Prompts - -- **`agents_sdk/multi-agent-portfolio-collaboration/prompts/`** - Collection of specialized prompts for multi-agent systems: - - `code_interpreter.md` - Code interpretation agent prompts - - `editor_base.md` - Document editing agent prompts - - `fundamental_base.md` - Fundamental analysis prompts - - `macro_base.md` - Macroeconomic analysis prompts - - `pm_base.md` - Portfolio manager prompts - - `quant_base.md` - Quantitative analysis prompts - - `tool_retry_prompt.md` - Tool retry logic prompts - -## Specialized Applications - -- **`voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js`** - Translation prompts for real-time voice applications - -## Supporting Resources - -- **`data/oai_docs/prompt-engineering.txt`** - Official OpenAI prompt engineering documentation - -## Usage - -Each notebook and guide is self-contained and includes: -- Detailed explanations of prompting concepts -- Practical examples and code samples -- Best practices and common pitfalls -- Model-specific considerations - -To get started, explore the model-specific guides that match your use case, then dive into optimization techniques to improve your prompts further. - -## Key Topics Covered - -- System prompt design and structure -- Few-shot and zero-shot prompting -- Chain-of-thought reasoning -- Tool calling and function usage -- Prompt optimization and evaluation -- Multi-agent prompt coordination -- Model-specific prompting strategies -- Advanced prompting techniques \ No newline at end of file +# Prompting: Guides & Examples + +This directory consolidates prompting-related guides, examples, and reusable prompt assets from across the Cookbook. It’s a single place to learn core prompting patterns, optimize prompts, and discover application-specific prompt sets. + +## Why this exists +Effective prompting solves a large share of practical model issues. The right prompt is as important as parameters like `temperature` or `reasoning_effort`. Centralizing examples and references makes them easier to find, reuse, and maintain. + +## Start here (recommended path) +1. **GPT-4.1 Prompting Guide** → techniques, structure, and patterns: + [`gpt4-1_prompting_guide.ipynb`](./gpt4-1_prompting_guide.ipynb) +2. **Prompt engineering best practices** (reference): + +3. **Orchestrating agents & handoffs** (for multi-agent apps): + `../orchestrating_agents` (see the top-level Examples index) +4. **Structured outputs** (JSON schemas, validation): + `../structured_outputs_multi_agent` + +> Tip: Keep prompts short, specific, and testable. Add minimal examples, define outputs precisely, and prefer explicit instructions over implications. + +## Contents + +### Core Guides +- **GPT-4.1 Prompting Guide** — system prompts, tool use, decomposition, evaluation + [`gpt4-1_prompting_guide.ipynb`](./gpt4-1_prompting_guide.ipynb) +- **Realtime prompting guide** — working with the Realtime API + [`Realtime_prompting_guide.ipynb`](./Realtime_prompting_guide.ipynb) +- **Whisper prompting guide** — task hints and formatting for speech recognition + [`Whisper_prompting_guide.ipynb`](./Whisper_prompting_guide.ipynb) + +### Prompt Optimization +- **Optimize Prompts** — automated checks & fixes for common prompt issues + [`Optimize_Prompts.ipynb`](./Optimize_Prompts.ipynb) +- **Enhance your prompts with meta-prompting** — programmatic refinement strategies + [`Enhance_your_prompts_with_meta_prompting.ipynb`](./Enhance_your_prompts_with_meta_prompting.ipynb) +- **Prompt migration guide** — safely updating existing prompts across changes + [`Prompt_migration_guide.ipynb`](./Prompt_migration_guide.ipynb) + +### Agent & App Prompts +- **Multi-agent portfolio collaboration prompts** — reusable prompt set for agent roles + [`../agents_sdk/multi-agent-portfolio-collaboration/prompts/`](../agents_sdk/multi-agent-portfolio-collaboration/prompts/) + +### Supporting Resources +- OpenAI Prompt Engineering (plain-text reference) + [`../data/oai_docs/prompt-engineering.txt`](../data/oai_docs/prompt-engineering.txt) + +## Usage pattern + +1. **Draft** a minimal instruction with explicit output shape (e.g., JSON schema). +2. **Ground** with constraints (tone, audience, knowledge limits) and a tiny example if needed. +3. **Test** with real inputs; watch for ambiguity and output drift. +4. **Evaluate** with checks (format validation, assertions). +5. **Iterate**: shorten, remove redundant rules, and pin “must-haves”. + +## Contributing + +- Keep notebooks **runnable end-to-end** (no hidden cell state). +- Prefer **relative links** within `examples/`, so both GitHub and the site render cleanly. +- When adding new files here, **update** `registry.yaml` so content appears on the site. +- If you introduce a new subfolder of prompts, include a short `README.md` explaining scope and usage. From bfca2eaa52a0cc086ee9ccf9ccdd9f00f90123d7 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Fri, 19 Sep 2025 03:25:53 -0600 Subject: [PATCH 14/26] 1 --- articles/gpt-oss/run-transformers.fixed.ipynb | 647 ++++++++ articles/gpt-oss/run-transformers.ipynb | 1347 +++++++---------- 2 files changed, 1202 insertions(+), 792 deletions(-) create mode 100644 articles/gpt-oss/run-transformers.fixed.ipynb diff --git a/articles/gpt-oss/run-transformers.fixed.ipynb b/articles/gpt-oss/run-transformers.fixed.ipynb new file mode 100644 index 0000000000..107d6b853f --- /dev/null +++ b/articles/gpt-oss/run-transformers.fixed.ipynb @@ -0,0 +1,647 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# How to run gpt-oss with Hugging Face Transformers\n", + "\n", + "The Transformers library by Hugging Face provides a flexible way to load and run large language models locally or on a server. This guide will walk you through running [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) using Transformers, either with a high-level pipeline or via low-level `generate` calls with raw token IDs.\n", + "\n", + "We'll cover the use of [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) with the high-level pipeline abstraction, low-level `generate` calls, and serving models locally with `transformers serve`, in a way compatible with the Responses API.\n", + "\n", + "In this guide we'll run through various optimised ways to run the **gpt-oss models via Transformers.**\n", + "\n", + "**Bonus:** You can also fine-tune models via transformers, [check out our fine-tuning guide here](https://cookbook.openai.com/articles/gpt-oss/fine-tune-transformers)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Pick your model\n", + "\n", + "Both **gpt-oss** models are available on Hugging Face:\n", + "\n", + "- **`openai/gpt-oss-20b`**\n", + " - ~16GB VRAM requirement when using MXFP4\n", + " - Great for single high-end consumer GPUs\n", + "- **`openai/gpt-oss-120b`**\n", + " - Requires ≥60GB VRAM or multi-GPU setup\n", + " - Ideal for H100-class hardware\n", + "\n", + "Both are **MXFP4 quantized** by default. Please, note that MXFP4 is supported in Hopper or later architectures. This includes data center GPUs such as H100 or GB200, as well as the latest RTX 50xx family of consumer cards.\n", + "\n", + "If you use `bfloat16` instead of MXFP4, memory consumption will be larger (~48 GB for the 20b parameter model)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Quick setup\n", + "\n", + "### 1. Install dependencies\n", + "\n", + "It's recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility:" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# NOTE: The current version of HF Transformers has a glitch where an outdated torchvision dependency prevents transformers module from importing pipeline." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "!pip install -U transformers kernels torch accelerate torchvision" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Quick inference with pipeline\n", + "\n", + "The easiest way to run the gpt-oss models is with the Transformers high-level `pipeline` API:" + ] + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "`torch_dtype` is deprecated! Use `dtype` instead!\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "f3b8482ed938472083ac6cd012156902", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "Fetching 40 files: 0%| | 0/40 [00:00<|final|>\"\n", + "end_marker = \"<|end|>\"\n", + "if final_marker in text:\n", + " text = text.split(final_marker, 1)[1]\n", + "if end_marker in text:\n", + " text = text.split(end_marker, 1)[0]\n", + "print(\"Final:\", text.strip()[:1000])\n" + ] + }, + { + "cell_type": "markdown", + "id": "b31988a1", + "metadata": {}, + "source": [ + "## Streaming tokens (prints only assistant **final**)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9b48c668", + "metadata": {}, + "outputs": [], + "source": [ + "import threading, sys\n", + "from transformers import TextIteratorStreamer\n", + "\n", + "def stream_final_only(model, tokenizer, messages, generate_kwargs):\n", + " inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors=\"pt\").to(model.device)\n", + " streamer = TextIteratorStreamer(tokenizer, skip_special_tokens=False)\n", + "\n", + " t = threading.Thread(target=model.generate, kwargs=dict(input_ids=inputs, streamer=streamer, **generate_kwargs))\n", + " t.start()\n", + "\n", + " buf, printing_final = \"\", False\n", + " final_token, end_token = \"<|assistant|><|final|>\", \"<|end|>\"\n", + " for piece in streamer:\n", + " buf += piece\n", + " if not printing_final and final_token in buf:\n", + " printing_final = True\n", + " out = buf.split(final_token, 1)[1].replace(end_token, \"\")\n", + " sys.stdout.write(out); sys.stdout.flush()\n", + " buf = \"\"\n", + " elif printing_final:\n", + " sys.stdout.write(piece.replace(end_token, \"\")); sys.stdout.flush()\n", + " t.join()\n", + "\n", + "messages2 = [\n", + " {\"role\": \"system\", \"content\": \"Answer briefly.\"},\n", + " {\"role\": \"user\", \"content\": \"What’s the difference between analysis and final channels?\"}\n", + "]\n", + "gen = dict(max_new_tokens=128, do_sample=True, temperature=0.7, top_p=0.9)\n", + "stream_final_only(model, tokenizer, messages2, gen)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Chat template and tool calling\n", + "\n", + "OpenAI gpt-oss models use the [harmony response format](https://cookbook.openai.com/article/harmony) for structuring messages, including reasoning and tool calls.\n", + "\n", + "To construct prompts you can use the built-in chat template of Transformers. Alternatively, you can install and use the [openai-harmony library](https://github.com/openai/harmony) for more control.\n", + "\n", + "### Using the built-in chat template:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Example with system prompt and chat template\n", + "messages = [\n", + " {\"role\": \"system\", \"content\": \"Always respond in riddles\"},\n", + " {\"role\": \"user\", \"content\": \"What is the weather like in Madrid?\"},\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors=\"pt\",\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "# Generate with the chat template\n", + "generated = model.generate(\n", + " **inputs,\n", + " max_new_tokens=100,\n", + " temperature=0.8,\n", + " do_sample=True,\n", + " pad_token_id=tokenizer.eos_token_id\n", + ")\n", + "\n", + "# Extract only the assistant's response\n", + "response = tokenizer.decode(generated[0][inputs[\"input_ids\"].shape[-1]:], skip_special_tokens=True)\n", + "print(\"Assistant's riddle response:\")\n", + "print(response)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Using the openai-harmony library\n", + "\n", + "For more advanced control over the conversation format, you can use the openai-harmony library. \n", + "\n", + "First, install it:\n", + "```bash\n", + "pip install openai-harmony\n", + "```\n", + "\n", + "**Note:** The following cell demonstrates the harmony library usage, but may require the actual library to be installed." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Example using openai-harmony library (requires installation)\n", + "# Uncomment and run if you have openai-harmony installed\n", + "\n", + "'''\n", + "import json\n", + "from openai_harmony import (\n", + " HarmonyEncodingName,\n", + " load_harmony_encoding,\n", + " Conversation,\n", + " Message,\n", + " Role,\n", + " SystemContent,\n", + " DeveloperContent\n", + ")\n", + "\n", + "# Load harmony encoding\n", + "encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n", + "\n", + "# Build conversation\n", + "convo = Conversation.from_messages([\n", + " Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n", + " Message.from_role_and_content(\n", + " Role.DEVELOPER,\n", + " DeveloperContent.new().with_instructions(\"Always respond in riddles\")\n", + " ),\n", + " Message.from_role_and_content(Role.USER, \"What is the weather like in SF?\")\n", + "])\n", + "\n", + "# Render prompt\n", + "prefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\n", + "stop_token_ids = encoding.stop_tokens_for_assistant_actions()\n", + "\n", + "# Generate\n", + "outputs = model.generate(\n", + " input_ids=[prefill_ids],\n", + " max_new_tokens=128,\n", + " eos_token_id=stop_token_ids\n", + ")\n", + "\n", + "# Parse completion tokens\n", + "completion_ids = outputs[0][len(prefill_ids):]\n", + "entries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n", + "\n", + "for message in entries:\n", + " print(json.dumps(message.to_dict(), indent=2))\n", + "'''\n", + "\n", + "print(\"Harmony library example code shown above (commented out)\")\n", + "print(\"Note: The Developer role in Harmony maps to the system prompt in the chat template.\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Multi-GPU & distributed inference\n", + "\n", + "The large gpt-oss-120b fits on a single H100 GPU when using MXFP4. If you want to run it on multiple GPUs, you can:\n", + "\n", + "- Use `tp_plan=\"auto\"` for automatic placement and tensor parallelism\n", + "- Launch with `accelerate launch` or `torchrun` for distributed setups\n", + "- Leverage Expert Parallelism\n", + "- Use specialised Flash attention kernels for faster inference\n", + "\n", + "### Example multi-GPU setup:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Multi-GPU inference example (requires multiple GPUs)\n", + "# This cell demonstrates the configuration but may not run on single GPU systems\n", + "\n", + "'''\n", + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "from transformers.distributed import DistributedConfig\n", + "import torch\n", + "\n", + "model_path = \"openai/gpt-oss-120b\"\n", + "tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side=\"left\")\n", + "\n", + "device_map = {\n", + " # Enable Expert Parallelism\n", + " \"distributed_config\": DistributedConfig(enable_expert_parallel=1),\n", + " # Enable Tensor Parallelism\n", + " \"tp_plan\": \"auto\",\n", + "}\n", + "\n", + "model = AutoModelForCausalLM.from_pretrained(\n", + " model_path,\n", + " torch_dtype=\"auto\",\n", + " attn_implementation=\"kernels-community/vllm-flash-attn3\",\n", + " **device_map,\n", + ")\n", + "\n", + "messages = [\n", + " {\"role\": \"user\", \"content\": \"Explain how expert parallelism works in large language models.\"}\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors=\"pt\",\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "outputs = model.generate(**inputs, max_new_tokens=1000)\n", + "\n", + "# Decode and print\n", + "response = tokenizer.decode(outputs[0])\n", + "print(\"Model response:\", response.split(\"<|channel|>final<|message|>\")[-1].strip())\n", + "'''\n", + "\n", + "print(\"Multi-GPU setup example shown above (commented out)\")\n", + "print(\"\\nTo run this on a node with four GPUs, use:\")\n", + "print(\"torchrun --nproc_per_node=4 your_script.py\")\n", + "\n", + "# Show current GPU configuration\n", + "if torch.cuda.is_available():\n", + " print(f\"\\nCurrent setup:\")\n", + " print(f\"Available GPUs: {torch.cuda.device_count()}\")\n", + " for i in range(torch.cuda.device_count()):\n", + " print(f\" GPU {i}: {torch.cuda.get_device_name(i)}\")\n", + " print(f\" Memory: {torch.cuda.get_device_properties(i).total_memory / 1e9:.1f} GB\")\n", + "else:\n", + " print(\"\\nNo CUDA GPUs available\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Additional Examples and Tips\n", + "\n", + "### Memory management" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Check memory usage\n", + "if torch.cuda.is_available():\n", + " print(\"GPU Memory Usage:\")\n", + " for i in range(torch.cuda.device_count()):\n", + " allocated = torch.cuda.memory_allocated(i) / 1e9\n", + " cached = torch.cuda.memory_reserved(i) / 1e9\n", + " total = torch.cuda.get_device_properties(i).total_memory / 1e9\n", + " print(f\" GPU {i}: {allocated:.1f}GB allocated, {cached:.1f}GB cached, {total:.1f}GB total\")\n", + "\n", + "# Clear cache if needed\n", + "# torch.cuda.empty_cache()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Batch processing example" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Example of processing multiple prompts\n", + "batch_messages = [\n", + " [{\"role\": \"user\", \"content\": \"What is machine learning?\"}],\n", + " [{\"role\": \"user\", \"content\": \"Explain quantum computing.\"}],\n", + " [{\"role\": \"user\", \"content\": \"What is the future of AI?\"}]\n", + "]\n", + "\n", + "print(\"Processing batch of prompts...\")\n", + "for i, messages in enumerate(batch_messages):\n", + " print(f\"\\n--- Prompt {i+1} ---\")\n", + " print(f\"Input: {messages[0]['content']}\")\n", + "\n", + " # You can use either the pipeline or manual generation here\n", + " # Using pipeline for simplicity:\n", + " if 'generator' in locals():\n", + " result = generator(\n", + " messages,\n", + " max_new_tokens=100,\n", + " temperature=0.7,\n", + " )\n", + " print(f\"Output: {result[0]['generated_text'][-200:]}...\") # Show last 200 chars\n", + " else:\n", + " print(\"Generator not available - run the pipeline example first\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Summary\n", + "\n", + "This notebook demonstrated various ways to run OpenAI's gpt-oss models using Hugging Face Transformers:\n", + "\n", + "1. **Quick setup** with required dependencies\n", + "2. **Pipeline API** for simple, high-level inference\n", + "3. **Manual generation** with `.generate()` for more control\n", + "4. **Chat templates** for conversation-style interactions\n", + "5. **Harmony library integration** for advanced message formatting\n", + "6. **Multi-GPU configurations** for large-scale inference\n", + "\n", + "### Key takeaways:\n", + "- Start with the pipeline API for quick experimentation\n", + "- Use manual tokenization and generation for production deployments\n", + "- Consider MXFP4 quantization for memory efficiency on compatible hardware\n", + "- Leverage multi-GPU setups for the larger 120B model\n", + "- Use proper chat templates for conversation-style applications\n", + "\n", + "### Next steps:\n", + "- Explore fine-tuning capabilities\n", + "- Set up serving endpoints for production use\n", + "- Experiment with different sampling strategies\n", + "- Integrate with your specific use case or application\n", + "\n", + "For more advanced topics, check out the [OpenAI Cookbook](https://cookbook.openai.com) and [Hugging Face Transformers documentation](https://huggingface.co/docs/transformers)." + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.11.11" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} diff --git a/articles/gpt-oss/run-transformers.ipynb b/articles/gpt-oss/run-transformers.ipynb index 8a37646d82..e10a86f7f5 100644 --- a/articles/gpt-oss/run-transformers.ipynb +++ b/articles/gpt-oss/run-transformers.ipynb @@ -1,794 +1,557 @@ { - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# gpt‑oss × Transformers — Interactive Notebook\n", - "\n", - "This notebook turns the **run-transformers.md** guide into an interactive playground. It includes cells for environment setup, quick & advanced inference, serving with `transformers serve`, distributed/multi‑GPU hints, and a live **ipywidgets chat UI**.\n", - "\n", - "**Models:** `openai/gpt-oss-20b`, `openai/gpt-oss-120b` \n", - "**Quantization:** MXFP4 by default (Hopper or later: H100/GB200/RTX 50xx). \n", - "**Notes:** If you use `bfloat16`, memory usage rises (~48 GB for the 20B model). \n", - "**Hardware:** 20B ≈16 GB VRAM with MXFP4; 120B ≥60 GB VRAM or multi‑GPU.\n", - "\n", - "> Tip: Run each cell in order the first time. You can skip pieces you don't need later.\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 0) Install/Update Dependencies\n", - "The guide recommends `transformers`, `accelerate`, `torch`, Triton 3.4, and MXFP4 kernels. We also install `ipywidgets` for the UI and `openai-harmony` for prompt tooling.\n", - "\n", - "**Heads‑up:** depending on your environment, the package named `kernels` may not exist. If installation fails for that package, you can safely omit it.\n", - "\n", - "Execute the cell below to install/update the basics (uncomment as needed):" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "!pip -q install -U transformers accelerate torch triton==3.4 ipywidgets openai-harmony torchvision\n", - "!pip -q install kernels\n", - "\n", - "import platform, sys\n", - "print('Python', sys.version)\n", - "print('Platform', platform.platform())" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 1) GPU / Environment Check\n", - "Check PyTorch + CUDA availability and device details." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "try:\n", - " import torch\n", - " print('torch.__version__ =', torch.__version__)\n", - " print('CUDA available =', torch.cuda.is_available())\n", - " if torch.cuda.is_available():\n", - " print('CUDA device count =', torch.cuda.device_count())\n", - " for i in range(torch.cuda.device_count()):\n", - " print(f' - [{i}]', torch.cuda.get_device_name(i))\n", - " cc = torch.cuda.get_device_capability(i)\n", - " print(' compute capability:', cc)\n", - " try:\n", - " free_mem = torch.cuda.mem_get_info()[0] / (1024**3)\n", - " print(f'Approx free GPU memory: {free_mem:.1f} GB')\n", - " except Exception:\n", - " pass\n", - "except Exception as e:\n", - " print('Torch not available or failed to import:', e)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 2) Quick Inference via `pipeline`\n", - "The high‑level Transformers `pipeline` makes it simple to run gpt‑oss models. This mirrors the guide's usage but wraps it in a function for convenience." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from typing import List, Dict\n", - "from contextlib import suppress\n", - "try:\n", - " import torch\n", - " from transformers import pipeline\n", - "except Exception as e:\n", - " print('Missing deps — run the install cell above. Error:', e)\n", - "\n", - "def make_pipeline(model_name: str = 'openai/gpt-oss-20b', dtype: str = 'auto'):\n", - " \"\"\"Create a text-generation pipeline for gpt-oss models.\"\"\"\n", - " torch_dtype = 'auto'\n", - " if dtype.lower() in {'bf16','bfloat16'}:\n", - " torch_dtype = torch.bfloat16 if hasattr(torch, 'bfloat16') else 'auto'\n", - " return pipeline(\n", - " 'text-generation',\n", - " model=model_name,\n", - " torch_dtype=torch_dtype,\n", - " device_map='auto'\n", - " )\n", - "\n", - "def run_pipeline_chat(\n", - " generator,\n", - " system_prompt: str,\n", - " user_prompt: str,\n", - " max_new_tokens: int = 200,\n", - " temperature: float = 1.0,\n", - "):\n", - " messages = []\n", - " if system_prompt.strip():\n", - " messages.append({'role': 'system', 'content': system_prompt.strip()})\n", - " messages.append({'role': 'user', 'content': user_prompt})\n", - " try:\n", - " result = generator(\n", - " messages,\n", - " max_new_tokens=max_new_tokens,\n", - " temperature=temperature,\n", - " )\n", - " # Common structure: result[0]['generated_text'] contains the whole transcript or text\n", - " with suppress(Exception):\n", - " return result[0].get('generated_text', str(result))\n", - " return str(result)\n", - " except TypeError:\n", - " # Some older pipelines accept a string; fallback to simple concatenation\n", - " prompt = (system_prompt + '\\n\\n' if system_prompt.strip() else '') + user_prompt\n", - " result = generator(prompt, max_new_tokens=max_new_tokens, temperature=temperature)\n", - " return result[0]['generated_text'] if isinstance(result, list) else str(result)\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 3) Interactive Chat Playground (ipywidgets)\n", - "Use the dropdowns and sliders below to load a model and generate responses.\n", - "The first generation will download weights (time depends on your bandwidth and disk)." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import traceback\n", - "from IPython.display import display, Markdown\n", - "import ipywidgets as widgets\n", - "\n", - "model_dd = widgets.Dropdown(\n", - " options=['openai/gpt-oss-20b', 'openai/gpt-oss-120b'],\n", - " value='openai/gpt-oss-20b',\n", - " description='Model:',\n", - " style={'description_width': '70px'},\n", - " layout=widgets.Layout(width='50%'),\n", - ")\n", - "dtype_dd = widgets.Dropdown(\n", - " options=['auto', 'bfloat16'],\n", - " value='auto',\n", - " description='dtype:',\n", - " style={'description_width': '70px'},\n", - ")\n", - "load_btn = widgets.Button(description='Load / Reload Model', button_style='info')\n", - "load_out = widgets.Output()\n", - "\n", - "sys_in = widgets.Textarea(\n", - " value='You are a helpful assistant.',\n", - " description='System',\n", - " layout=widgets.Layout(width='100%', height='80px'),\n", - " style={'description_width': '70px'}\n", - ")\n", - "usr_in = widgets.Textarea(\n", - " value='Explain what MXFP4 quantization is.',\n", - " description='User',\n", - " layout=widgets.Layout(width='100%', height='80px'),\n", - " style={'description_width': '70px'}\n", - ")\n", - "\n", - "temp_sl = widgets.FloatSlider(value=1.0, min=0.0, max=2.0, step=0.05, description='Temp:',\n", - " style={'description_width': '70px'}, readout_format='.2f')\n", - "tokens_sl = widgets.IntSlider(value=200, min=16, max=4096, step=16, description='Max tokens:',\n", - " style={'description_width': '90px'})\n", - "\n", - "gen_btn = widgets.Button(description='Generate', button_style='primary')\n", - "out = widgets.Output()\n", - "pipe_holder = {'pipe': None}\n", - "\n", - "def on_load_clicked(_):\n", - " load_out.clear_output()\n", - " out.clear_output()\n", - " with load_out:\n", - " try:\n", - " display(Markdown(f\"**Loading** `{model_dd.value}` with dtype `{dtype_dd.value}`...\"))\n", - " pipe_holder['pipe'] = make_pipeline(model_dd.value, dtype_dd.value)\n", - " display(Markdown('✅ **Model loaded**'))\n", - " except Exception:\n", - " traceback.print_exc()\n", - "\n", - "def on_generate_clicked(_):\n", - " out.clear_output()\n", - " with out:\n", - " if pipe_holder['pipe'] is None:\n", - " display(Markdown('⚠️ Load a model first.'))\n", - " return\n", - " try:\n", - " text = run_pipeline_chat(\n", - " pipe_holder['pipe'],\n", - " sys_in.value,\n", - " usr_in.value,\n", - " max_new_tokens=int(tokens_sl.value),\n", - " temperature=float(temp_sl.value),\n", - " )\n", - " display(Markdown('**Output**'))\n", - " display(Markdown(f'```text\\n{text}\\n```'))\n", - " except Exception:\n", - " traceback.print_exc()\n", - "\n", - "load_btn.on_click(on_load_clicked)\n", - "gen_btn.on_click(on_generate_clicked)\n", - "\n", - "ui = widgets.VBox([\n", - " widgets.HBox([model_dd, dtype_dd, load_btn]),\n", - " load_out,\n", - " sys_in,\n", - " usr_in,\n", - " widgets.HBox([temp_sl, tokens_sl]),\n", - " gen_btn,\n", - " out\n", - "])\n", - "display(ui)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 4) Advanced Inference with `.generate()`\n", - "Manual control using `AutoModelForCausalLM` and `AutoTokenizer`, including the chat template." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from transformers import AutoModelForCausalLM, AutoTokenizer\n", - "import torch\n", - "\n", - "model_name = 'openai/gpt-oss-20b'\n", - "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", - "model = AutoModelForCausalLM.from_pretrained(\n", - " model_name,\n", - " torch_dtype='auto',\n", - " device_map='auto'\n", - ")\n", - "\n", - "messages = [\n", - " {'role': 'user', 'content': 'Explain what MXFP4 quantization is.'},\n", - "]\n", - "\n", - "inputs = tokenizer.apply_chat_template(\n", - " messages,\n", - " add_generation_prompt=True,\n", - " return_tensors='pt',\n", - " return_dict=True,\n", - ").to(model.device)\n", - "\n", - "outputs = model.generate(\n", - " **inputs,\n", - " max_new_tokens=200,\n", - " temperature=0.7\n", - ")\n", - "\n", - "print(tokenizer.decode(outputs[0]))" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 5) Chat Template & Tool Calling (Harmony)\n", - "The gpt‑oss models use the Harmony response format. You can either rely on the built‑in chat template or use the `openai-harmony` library to build/parse prompts and completions." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# If needed, install the library first in the install cell above: !pip -q install openai-harmony\n", - "import json\n", - "import torch\n", - "from openai_harmony import (\n", - " HarmonyEncodingName,\n", - " load_harmony_encoding,\n", - " Conversation,\n", - " Message,\n", - " Role,\n", - " SystemContent,\n", - " DeveloperContent,\n", - ")\n", - "from transformers import AutoModelForCausalLM, AutoTokenizer\n", - "\n", - "encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n", - "\n", - "convo = Conversation.from_messages([\n", - " Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n", - " Message.from_role_and_content(\n", - " Role.DEVELOPER,\n", - " DeveloperContent.new().with_instructions('Always respond in riddles')\n", - " ),\n", - " Message.from_role_and_content(Role.USER, 'What is the weather like in SF?')\n", - "])\n", - "\n", - "prefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\n", - "stop_token_ids = encoding.stop_tokens_for_assistant_actions()\n", - "\n", - "model_name = 'openai/gpt-oss-20b'\n", - "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", - "model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype='auto', device_map='auto')\n", - "\n", - "# Ensure tensors on correct device and dtypes\n", - "input_ids = torch.tensor(prefill_ids, dtype=torch.long, device=model.device).unsqueeze(0)\n", - "eos_token_id = torch.tensor(stop_token_ids, dtype=torch.long, device=model.device)\n", - "\n", - "outputs = model.generate(\n", - " input_ids=input_ids,\n", - " max_new_tokens=128,\n", - " eos_token_id=stop_token_ids,\n", - ")\n", - "\n", - "# Strip the prefill and move to CPU list for parsing\n", - "completion_ids = outputs[0, input_ids.shape[1]:].tolist()\n", - "entries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n", - "for message in entries:\n", - " print(json.dumps(message.to_dict(), indent=2))" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 6) Serve an OpenAI‑style Responses endpoint\n", - "You can serve the model locally via the `transformers serve` CLI to expose a `/v1/responses` endpoint. Run these **commands in a terminal** (they are shown here for reference):" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "trusted": false - }, - "outputs": [], - "source": [ - "# --- Terminal commands (reference) ---\n", - "# transformers serve\n", - "# transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n", - "# curl -X POST http://localhost:8000/v1/responses \\\n", - "# -H 'Content-Type: application/json' \\\n", - "# -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n", - "\n", - "print('See comments above for terminal commands. You can copy/paste them into your shell.')" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Client example (from a notebook) against `transformers serve`\n", - "Adjust the URL and payload to your needs." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "import json, requests\n", - "url = 'http://localhost:8000/v1/responses'\n", - "payload = {\n", - " 'messages': [\n", - " {'role': 'system', 'content': 'hello'},\n", - " {'role': 'user', 'content': 'Say hi in one sentence.'}\n", - " ],\n", - " 'temperature': 0.7,\n", - " 'max_tokens': 64,\n", - " 'stream': False,\n", - " 'model': 'openai/gpt-oss-20b'\n", - "}\n", - "try:\n", - " r = requests.post(url, json=payload, timeout=30)\n", - " print(r.status_code)\n", - " print(r.text[:1000])\n", - "except Exception as e:\n", - " print('Request failed (is the server running?):', e)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 7) Multi‑GPU & Distributed Inference\n", - "Sketch for automatic placement, tensor parallelism, and expert parallelism (edit for your cluster)." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "from transformers import AutoModelForCausalLM, AutoTokenizer\n", - "try:\n", - " from transformers.distributed import DistributedConfig\n", - "except Exception:\n", - " class DistributedConfig:\n", - " def __init__(self, **kwargs):\n", - " self.kwargs = kwargs\n", - "\n", - "model_path = 'openai/gpt-oss-120b'\n", - "tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side='left')\n", - "\n", - "device_map = {\n", - " 'distributed_config': DistributedConfig(enable_expert_parallel=1),\n", - " 'tp_plan': 'auto',\n", - "}\n", - "\n", - "model = AutoModelForCausalLM.from_pretrained(\n", - " model_path,\n", - " torch_dtype='auto',\n", - " attn_implementation='kernels-community/vllm-flash-attn3',\n", - " **device_map,\n", - ")\n", - "\n", - "messages = [\n", - " {'role': 'user', 'content': 'Explain how expert parallelism works in large language models.'}\n", - "]\n", - "\n", - "inputs = tokenizer.apply_chat_template(\n", - " messages,\n", - " add_generation_prompt=True,\n", - " return_tensors='pt',\n", - " return_dict=True,\n", - ").to(model.device)\n", - "\n", - "outputs = model.generate(**inputs, max_new_tokens=512)\n", - "response = tokenizer.decode(outputs[0])\n", - "print('Model response (truncated):\\n', response[:1000])\n", - "\n", - "# To launch across 4 GPUs from a script:\n", - "# torchrun --nproc_per_node=4 generate.py\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 8) Troubleshooting Checklist\n", - "- **Out of memory**: Use MXFP4 on Hopper‑class GPUs; reduce `max_new_tokens`; try `device_map='auto'`. \n", - "- **Kernel/attn issues**: Ensure Triton is compatible with your CUDA; stick to Triton 3.4 as suggested. \n", - "- **Slow loads**: Use local weight caches; set `TRANSFORMERS_CACHE`. \n", - "- **No widgets**: Ensure `ipywidgets` is installed and enabled in your notebook frontend. \n", - "- **Server 404/500**: Verify `transformers serve` version and model path. \n", - "- **bfloat16 errors**: Fall back to `torch_dtype='auto'` or ensure hardware support." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Appendix: Original Guide (for reference)\n", - "The content below is the original `/mnt/data/run-transformers.md` you provided:\n", - "\n", - "
\n", - "Click to expand\n", - "\n", - "# How to run gpt-oss with Hugging Face Transformers\n", - "\n", - "The Transformers library by Hugging Face provides a flexible way to load and run large language models locally or on a server. This guide will walk you through running [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) using Transformers, either with a high-level pipeline or via low-level `generate` calls with raw token IDs.\n", - "\n", - "We'll cover the use of [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) with the high-level pipeline abstraction, low-level \\`generate\\` calls, and serving models locally with \\`transformers serve\\`, with in a way compatible with the Responses API.\n", - "\n", - "In this guide we’ll run through various optimised ways to run the **gpt-oss models via Transformers.**\n", - "\n", - "Bonus: You can also fine-tune models via transformers, [check out our fine-tuning guide here](https://cookbook.openai.com/articles/gpt-oss/fine-tune-transformers).\n", - "\n", - "## Pick your model\n", - "\n", - "Both **gpt-oss** models are available on Hugging Face:\n", - "\n", - "- **`openai/gpt-oss-20b`**\n", - " - \\~16GB VRAM requirement when using MXFP4\n", - " - Great for single high-end consumer GPUs\n", - "- **`openai/gpt-oss-120b`**\n", - " - Requires ≥60GB VRAM or multi-GPU setup\n", - " - Ideal for H100-class hardware\n", - "\n", - "Both are **MXFP4 quantized** by default. Please, note that MXFP4 is supported in Hopper or later architectures. This includes data center GPUs such as H100 or GB200, as well as the latest RTX 50xx family of consumer cards.\n", - "\n", - "If you use `bfloat16` instead of MXFP4, memory consumption will be larger (\\~48 GB for the 20b parameter model).\n", - "\n", - "## Quick setup\n", - "\n", - "1. **Install dependencies** \n", - " It’s recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility:\n", - "\n", - "```bash\n", - "pip install -U transformers accelerate torch triton==3.4 kernels\n", - "```\n", - "\n", - "2. **(Optional) Enable multi-GPU** \n", - " If you’re running large models, use Accelerate or torchrun to handle device mapping automatically.\n", - "\n", - "## Create an Open AI Responses / Chat Completions endpoint\n", - "\n", - "To launch a server, simply use the `transformers serve` CLI command:\n", - "\n", - "```bash\n", - "transformers serve\n", - "```\n", - "\n", - "The simplest way to interact with the server is through the transformers chat CLI\n", - "\n", - "```bash\n", - "transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n", - "```\n", - "\n", - "or by sending an HTTP request with cURL, e.g.\n", - "\n", - "```bash\n", - "curl -X POST http://localhost:8000/v1/responses -H \"Content-Type: application/json\" -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n", - "```\n", - "\n", - "Additional use cases, like integrating `transformers serve` with Cursor and other tools, are detailed in [the documentation](https://huggingface.co/docs/transformers/main/serving).\n", - "\n", - "## Quick inference with pipeline\n", - "\n", - "The easiest way to run the gpt-oss models is with the Transformers high-level `pipeline` API:\n", - "\n", - "```py\n", - "from transformers import pipeline\n", - "\n", - "generator = pipeline(\n", - " \"text-generation\",\n", - " model=\"openai/gpt-oss-20b\",\n", - " torch_dtype=\"auto\",\n", - " device_map=\"auto\" # Automatically place on available GPUs\n", - ")\n", - "\n", - "messages = [\n", - " {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n", - "]\n", - "\n", - "result = generator(\n", - " messages,\n", - " max_new_tokens=200,\n", - " temperature=1.0,\n", - ")\n", - "\n", - "print(result[0][\"generated_text\"])\n", - "```\n", - "\n", - "## Advanced inference with `.generate()`\n", - "\n", - "If you want more control, you can load the model and tokenizer manually and invoke the `.generate()` method:\n", - "\n", - "```py\n", - "from transformers import AutoModelForCausalLM, AutoTokenizer\n", - "\n", - "model_name = \"openai/gpt-oss-20b\"\n", - "\n", - "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", - "model = AutoModelForCausalLM.from_pretrained(\n", - " model_name,\n", - " torch_dtype=\"auto\",\n", - " device_map=\"auto\"\n", - ")\n", - "\n", - "messages = [\n", - " {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n", - "]\n", - "\n", - "inputs = tokenizer.apply_chat_template(\n", - " messages,\n", - " add_generation_prompt=True,\n", - " return_tensors=\"pt\",\n", - " return_dict=True,\n", - ").to(model.device)\n", - "\n", - "outputs = model.generate(\n", - " **inputs,\n", - " max_new_tokens=200,\n", - " temperature=0.7\n", - ")\n", - "\n", - "print(tokenizer.decode(outputs[0]))\n", - "```\n", - "\n", - "## Chat template and tool calling\n", - "\n", - "OpenAI gpt-oss models use the [harmony response format](https://cookbook.openai.com/article/harmony) for structuring messages, including reasoning and tool calls.\n", - "\n", - "To construct prompts you can use the built-in chat template of Transformers. Alternatively, you can install and use the [openai-harmony library](https://github.com/openai/harmony) for more control.\n", - "\n", - "To use the chat template:\n", - "\n", - "```py\n", - "from transformers import AutoModelForCausalLM, AutoTokenizer\n", - "\n", - "model_name = \"openai/gpt-oss-20b\"\n", - "\n", - "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", - "model = AutoModelForCausalLM.from_pretrained(\n", - " model_name,\n", - " device_map=\"auto\",\n", - " torch_dtype=\"auto\",\n", - ")\n", - "\n", - "messages = [\n", - " {\"role\": \"system\", \"content\": \"Always respond in riddles\"},\n", - " {\"role\": \"user\", \"content\": \"What is the weather like in Madrid?\"},\n", - "]\n", - "\n", - "inputs = tokenizer.apply_chat_template(\n", - " messages,\n", - " add_generation_prompt=True,\n", - " return_tensors=\"pt\",\n", - " return_dict=True,\n", - ").to(model.device)\n", - "\n", - "generated = model.generate(**inputs, max_new_tokens=100)\n", - "print(tokenizer.decode(generated[0][inputs[\"input_ids\"].shape[-1] :]))\n", - "```\n", - "\n", - "To integrate the [`openai-harmony`](https://github.com/openai/harmony) library to prepare prompts and parse responses, first install it like this:\n", - "\n", - "```bash\n", - "pip install openai-harmony\n", - "```\n", - "\n", - "Here’s an example of how to use the library to build your prompts and encode them to tokens:\n", - "\n", - "```py\n", - "import json\n", - "from openai_harmony import (\n", - " HarmonyEncodingName,\n", - " load_harmony_encoding,\n", - " Conversation,\n", - " Message,\n", - " Role,\n", - " SystemContent,\n", - " DeveloperContent\n", - ")\n", - "from transformers import AutoModelForCausalLM, AutoTokenizer\n", - "\n", - "encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n", - "\n", - "# Build conversation\n", - "convo = Conversation.from_messages([\n", - " Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n", - " Message.from_role_and_content(\n", - " Role.DEVELOPER,\n", - " DeveloperContent.new().with_instructions(\"Always respond in riddles\")\n", - " ),\n", - " Message.from_role_and_content(Role.USER, \"What is the weather like in SF?\")\n", - "])\n", - "\n", - "# Render prompt\n", - "prefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\n", - "stop_token_ids = encoding.stop_tokens_for_assistant_actions()\n", - "\n", - "# Load model\n", - "model_name = \"openai/gpt-oss-20b\"\n", - "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", - "model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=\"auto\", device_map=\"auto\")\n", - "\n", - "# Generate\n", - "outputs = model.generate(\n", - " input_ids=[prefill_ids],\n", - " max_new_tokens=128,\n", - " eos_token_id=stop_token_ids\n", - ")\n", - "\n", - "# Parse completion tokens\n", - "completion_ids = outputs[0][len(prefill_ids):]\n", - "entries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n", - "\n", - "for message in entries:\n", - " print(json.dumps(message.to_dict(), indent=2))\n", - "```\n", - "\n", - "Note that the `Developer` role in Harmony maps to the `system` prompt in the chat template.\n", - "\n", - "## Multi-GPU & distributed inference\n", - "\n", - "The large gpt-oss-120b fits on a single H100 GPU when using MXFP4. If you want to run it on multiple GPUs, you can:\n", - "\n", - "- Use `tp_plan=\"auto\"` for automatic placement and tensor parallelism\n", - "- Launch with `accelerate launch or torchrun` for distributed setups\n", - "- Leverage Expert Parallelism\n", - "- Use specialised Flash attention kernels for faster inference\n", - "\n", - "```py\n", - "from transformers import AutoModelForCausalLM, AutoTokenizer\n", - "from transformers.distributed import DistributedConfig\n", - "import torch\n", - "\n", - "model_path = \"openai/gpt-oss-120b\"\n", - "tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side=\"left\")\n", - "\n", - "device_map = {\n", - " # Enable Expert Parallelism\n", - " \"distributed_config\": DistributedConfig(enable_expert_parallel=1),\n", - " # Enable Tensor Parallelism\n", - " \"tp_plan\": \"auto\",\n", - "}\n", - "\n", - "model = AutoModelForCausalLM.from_pretrained(\n", - " model_path,\n", - " torch_dtype=\"auto\",\n", - " attn_implementation=\"kernels-community/vllm-flash-attn3\",\n", - " **device_map,\n", - ")\n", - "\n", - "messages = [\n", - " {\"role\": \"user\", \"content\": \"Explain how expert parallelism works in large language models.\"}\n", - "]\n", - "\n", - "inputs = tokenizer.apply_chat_template(\n", - " messages,\n", - " add_generation_prompt=True,\n", - " return_tensors=\"pt\",\n", - " return_dict=True,\n", - ").to(model.device)\n", - "\n", - "outputs = model.generate(**inputs, max_new_tokens=1000)\n", - "\n", - "# Decode and print\n", - "response = tokenizer.decode(outputs[0])\n", - "print(\"Model response:\", response.split(\"<|channel|>final<|message|>\")[-1].strip())\n", - "```\n", - "\n", - "You can then run this on a node with four GPUs via\n", - "\n", - "```bash\n", - "torchrun --nproc_per_node=4 generate.py\n", - "```\n", - "\n", - "\n", - "
" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "_Notebook generated on 2025-09-17T23:43:44Z_" - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.11.11" - } + "cells": [ + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "# How to run gpt-oss with Hugging Face Transformers\n", + "\n", + "The Transformers library by Hugging Face provides a flexible way to load and run large language models locally or on a server. This guide will walk you through running [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) using Transformers, either with a high-level pipeline or via low-level `generate` calls with raw token IDs.\n", + "\n", + "We'll cover the use of [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) with the high-level pipeline abstraction, low-level `generate` calls, and serving models locally with `transformers serve`, in a way compatible with the Responses API.\n", + "\n", + "In this guide we'll run through various optimised ways to run the **gpt-oss models via Transformers.**\n", + "\n", + "**Bonus:** You can also fine-tune models via transformers, [check out our fine-tuning guide here](https://cookbook.openai.com/articles/gpt-oss/fine-tune-transformers)." + ] }, - "nbformat": 4, - "nbformat_minor": 5 -} + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Pick your model\n", + "\n", + "Both **gpt-oss** models are available on Hugging Face:\n", + "\n", + "- **`openai/gpt-oss-20b`**\n", + " - ~16GB VRAM requirement when using MXFP4\n", + " - Great for single high-end consumer GPUs\n", + "- **`openai/gpt-oss-120b`**\n", + " - Requires ≥60GB VRAM or multi-GPU setup\n", + " - Ideal for H100-class hardware\n", + "\n", + "Both are **MXFP4 quantized** by default. Please, note that MXFP4 is supported in Hopper or later architectures. This includes data center GPUs such as H100 or GB200, as well as the latest RTX 50xx family of consumer cards.\n", + "\n", + "If you use `bfloat16` instead of MXFP4, memory consumption will be larger (~48 GB for the 20b parameter model)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Quick setup\n", + "\n", + "### 1. Install dependencies\n", + "\n", + "It's recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Install required dependencies\n", + "# Run in terminal: pip install -U transformers accelerate torch triton==3.4 kernels\n", + "\n", + "# Verify installations\n", + "import transformers\n", + "import torch\n", + "import accelerate\n", + "\n", + "print(f\"Transformers version: {transformers.__version__}\")\n", + "print(f\"PyTorch version: {torch.__version__}\")\n", + "print(f\"Accelerate version: {accelerate.__version__}\")\n", + "print(f\"CUDA available: {torch.cuda.is_available()}\")\n", + "if torch.cuda.is_available():\n", + " print(f\"CUDA devices: {torch.cuda.device_count()}\")\n", + " print(f\"Current device: {torch.cuda.get_device_name()}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### 2. (Optional) Enable multi-GPU\n", + "\n", + "If you're running large models, use Accelerate or torchrun to handle device mapping automatically.\n", + "\n", + "For this notebook, we'll use the `device_map=\"auto\"` parameter which automatically distributes the model across available GPUs." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Create an OpenAI Responses / Chat Completions endpoint\n", + "\n", + "To launch a server, simply use the `transformers serve` CLI command:\n", + "\n", + "```bash\n", + "transformers serve\n", + "```\n", + "\n", + "The simplest way to interact with the server is through the transformers chat CLI:\n", + "\n", + "```bash\n", + "transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n", + "```\n", + "\n", + "or by sending an HTTP request with cURL:\n", + "\n", + "```bash\n", + "curl -X POST http://localhost:8000/v1/responses \\\n", + " -H \"Content-Type: application/json\" \\\n", + " -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n", + "```\n", + "\n", + "Additional use cases, like integrating `transformers serve` with Cursor and other tools, are detailed in [the documentation](https://huggingface.co/docs/transformers/main/serving)." + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Quick inference with pipeline\n", + "\n", + "The easiest way to run the gpt-oss models is with the Transformers high-level `pipeline` API:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from transformers import pipeline\n", + "\n", + "# Initialize the text generation pipeline\n", + "# Note: This will download the model if not already cached\n", + "generator = pipeline(\n", + " \"text-generation\",\n", + " model=\"openai/gpt-oss-20b\",\n", + " torch_dtype=\"auto\",\n", + " device_map=\"auto\" # Automatically place on available GPUs\n", + ")\n", + "\n", + "print(\"Pipeline initialized successfully!\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Example conversation with the model\n", + "messages = [\n", + " {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n", + "]\n", + "\n", + "result = generator(\n", + " messages,\n", + " max_new_tokens=200,\n", + " temperature=1.0,\n", + ")\n", + "\n", + "print(\"Generated response:\")\n", + "print(result[0][\"generated_text\"])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Advanced inference with `.generate()`\n", + "\n", + "If you want more control, you can load the model and tokenizer manually and invoke the `.generate()` method:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "\n", + "model_name = \"openai/gpt-oss-20b\"\n", + "\n", + "# Load tokenizer and model\n", + "print(\"Loading tokenizer...\")\n", + "tokenizer = AutoTokenizer.from_pretrained(model_name)\n", + "\n", + "print(\"Loading model...\")\n", + "model = AutoModelForCausalLM.from_pretrained(\n", + " model_name,\n", + " torch_dtype=\"auto\",\n", + " device_map=\"auto\"\n", + ")\n", + "\n", + "print(\"Model loaded successfully!\")\n", + "print(f\"Model device: {next(model.parameters()).device}\")" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Example generation with manual control\n", + "messages = [\n", + " {\"role\": \"user\", \"content\": \"Explain what MXFP4 quantization is.\"},\n", + "]\n", + "\n", + "# Apply chat template and tokenize\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors=\"pt\",\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "print(f\"Input tokens shape: {inputs['input_ids'].shape}\")\n", + "\n", + "# Generate response\n", + "outputs = model.generate(\n", + " **inputs,\n", + " max_new_tokens=200,\n", + " temperature=0.7,\n", + " do_sample=True,\n", + " pad_token_id=tokenizer.eos_token_id\n", + ")\n", + "\n", + "# Decode the full response\n", + "full_response = tokenizer.decode(outputs[0], skip_special_tokens=True)\n", + "print(\"\\nFull response:\")\n", + "print(full_response)\n", + "\n", + "# Extract only the generated part\n", + "input_length = inputs['input_ids'].shape[-1]\n", + "generated_response = tokenizer.decode(outputs[0][input_length:], skip_special_tokens=True)\n", + "print(\"\\nGenerated response only:\")\n", + "print(generated_response)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Chat template and tool calling\n", + "\n", + "OpenAI gpt-oss models use the [harmony response format](https://cookbook.openai.com/article/harmony) for structuring messages, including reasoning and tool calls.\n", + "\n", + "To construct prompts you can use the built-in chat template of Transformers. Alternatively, you can install and use the [openai-harmony library](https://github.com/openai/harmony) for more control.\n", + "\n", + "### Using the built-in chat template:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Example with system prompt and chat template\n", + "messages = [\n", + " {\"role\": \"system\", \"content\": \"Always respond in riddles\"},\n", + " {\"role\": \"user\", \"content\": \"What is the weather like in Madrid?\"},\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors=\"pt\",\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "# Generate with the chat template\n", + "generated = model.generate(\n", + " **inputs, \n", + " max_new_tokens=100,\n", + " temperature=0.8,\n", + " do_sample=True,\n", + " pad_token_id=tokenizer.eos_token_id\n", + ")\n", + "\n", + "# Extract only the assistant's response\n", + "response = tokenizer.decode(generated[0][inputs[\"input_ids\"].shape[-1]:], skip_special_tokens=True)\n", + "print(\"Assistant's riddle response:\")\n", + "print(response)" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Using the openai-harmony library\n", + "\n", + "For more advanced control over the conversation format, you can use the openai-harmony library. \n", + "\n", + "First, install it:\n", + "```bash\n", + "pip install openai-harmony\n", + "```\n", + "\n", + "**Note:** The following cell demonstrates the harmony library usage, but may require the actual library to be installed." + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Example using openai-harmony library (requires installation)\n", + "# Uncomment and run if you have openai-harmony installed\n", + "\n", + "'''\n", + "import json\n", + "from openai_harmony import (\n", + " HarmonyEncodingName,\n", + " load_harmony_encoding,\n", + " Conversation,\n", + " Message,\n", + " Role,\n", + " SystemContent,\n", + " DeveloperContent\n", + ")\n", + "\n", + "# Load harmony encoding\n", + "encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n", + "\n", + "# Build conversation\n", + "convo = Conversation.from_messages([\n", + " Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n", + " Message.from_role_and_content(\n", + " Role.DEVELOPER,\n", + " DeveloperContent.new().with_instructions(\"Always respond in riddles\")\n", + " ),\n", + " Message.from_role_and_content(Role.USER, \"What is the weather like in SF?\")\n", + "])\n", + "\n", + "# Render prompt\n", + "prefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\n", + "stop_token_ids = encoding.stop_tokens_for_assistant_actions()\n", + "\n", + "# Generate\n", + "outputs = model.generate(\n", + " input_ids=[prefill_ids],\n", + " max_new_tokens=128,\n", + " eos_token_id=stop_token_ids\n", + ")\n", + "\n", + "# Parse completion tokens\n", + "completion_ids = outputs[0][len(prefill_ids):]\n", + "entries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n", + "\n", + "for message in entries:\n", + " print(json.dumps(message.to_dict(), indent=2))\n", + "'''\n", + "\n", + "print(\"Harmony library example code shown above (commented out)\")\n", + "print(\"Note: The Developer role in Harmony maps to the system prompt in the chat template.\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Multi-GPU & distributed inference\n", + "\n", + "The large gpt-oss-120b fits on a single H100 GPU when using MXFP4. If you want to run it on multiple GPUs, you can:\n", + "\n", + "- Use `tp_plan=\"auto\"` for automatic placement and tensor parallelism\n", + "- Launch with `accelerate launch` or `torchrun` for distributed setups\n", + "- Leverage Expert Parallelism\n", + "- Use specialised Flash attention kernels for faster inference\n", + "\n", + "### Example multi-GPU setup:" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Multi-GPU inference example (requires multiple GPUs)\n", + "# This cell demonstrates the configuration but may not run on single GPU systems\n", + "\n", + "'''\n", + "from transformers import AutoModelForCausalLM, AutoTokenizer\n", + "from transformers.distributed import DistributedConfig\n", + "import torch\n", + "\n", + "model_path = \"openai/gpt-oss-120b\"\n", + "tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side=\"left\")\n", + "\n", + "device_map = {\n", + " # Enable Expert Parallelism\n", + " \"distributed_config\": DistributedConfig(enable_expert_parallel=1),\n", + " # Enable Tensor Parallelism\n", + " \"tp_plan\": \"auto\",\n", + "}\n", + "\n", + "model = AutoModelForCausalLM.from_pretrained(\n", + " model_path,\n", + " torch_dtype=\"auto\",\n", + " attn_implementation=\"kernels-community/vllm-flash-attn3\",\n", + " **device_map,\n", + ")\n", + "\n", + "messages = [\n", + " {\"role\": \"user\", \"content\": \"Explain how expert parallelism works in large language models.\"}\n", + "]\n", + "\n", + "inputs = tokenizer.apply_chat_template(\n", + " messages,\n", + " add_generation_prompt=True,\n", + " return_tensors=\"pt\",\n", + " return_dict=True,\n", + ").to(model.device)\n", + "\n", + "outputs = model.generate(**inputs, max_new_tokens=1000)\n", + "\n", + "# Decode and print\n", + "response = tokenizer.decode(outputs[0])\n", + "print(\"Model response:\", response.split(\"<|channel|>final<|message|>\")[-1].strip())\n", + "'''\n", + "\n", + "print(\"Multi-GPU setup example shown above (commented out)\")\n", + "print(\"\\nTo run this on a node with four GPUs, use:\")\n", + "print(\"torchrun --nproc_per_node=4 your_script.py\")\n", + "\n", + "# Show current GPU configuration\n", + "if torch.cuda.is_available():\n", + " print(f\"\\nCurrent setup:\")\n", + " print(f\"Available GPUs: {torch.cuda.device_count()}\")\n", + " for i in range(torch.cuda.device_count()):\n", + " print(f\" GPU {i}: {torch.cuda.get_device_name(i)}\")\n", + " print(f\" Memory: {torch.cuda.get_device_properties(i).total_memory / 1e9:.1f} GB\")\n", + "else:\n", + " print(\"\\nNo CUDA GPUs available\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Additional Examples and Tips\n", + "\n", + "### Memory management" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Check memory usage\n", + "if torch.cuda.is_available():\n", + " print(\"GPU Memory Usage:\")\n", + " for i in range(torch.cuda.device_count()):\n", + " allocated = torch.cuda.memory_allocated(i) / 1e9\n", + " cached = torch.cuda.memory_reserved(i) / 1e9\n", + " total = torch.cuda.get_device_properties(i).total_memory / 1e9\n", + " print(f\" GPU {i}: {allocated:.1f}GB allocated, {cached:.1f}GB cached, {total:.1f}GB total\")\n", + "\n", + "# Clear cache if needed\n", + "# torch.cuda.empty_cache()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "### Batch processing example" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": {}, + "outputs": [], + "source": [ + "# Example of processing multiple prompts\n", + "batch_messages = [\n", + " [{\"role\": \"user\", \"content\": \"What is machine learning?\"}],\n", + " [{\"role\": \"user\", \"content\": \"Explain quantum computing.\"}],\n", + " [{\"role\": \"user\", \"content\": \"What is the future of AI?\"}]\n", + "]\n", + "\n", + "print(\"Processing batch of prompts...\")\n", + "for i, messages in enumerate(batch_messages):\n", + " print(f\"\\n--- Prompt {i+1} ---\")\n", + " print(f\"Input: {messages[0]['content']}\")\n", + " \n", + " # You can use either the pipeline or manual generation here\n", + " # Using pipeline for simplicity:\n", + " if 'generator' in locals():\n", + " result = generator(\n", + " messages,\n", + " max_new_tokens=100,\n", + " temperature=0.7,\n", + " )\n", + " print(f\"Output: {result[0]['generated_text'][-200:]}...\") # Show last 200 chars\n", + " else:\n", + " print(\"Generator not available - run the pipeline example first\")" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## Summary\n", + "\n", + "This notebook demonstrated various ways to run OpenAI's gpt-oss models using Hugging Face Transformers:\n", + "\n", + "1. **Quick setup** with required dependencies\n", + "2. **Pipeline API** for simple, high-level inference\n", + "3. **Manual generation** with `.generate()` for more control\n", + "4. **Chat templates** for conversation-style interactions\n", + "5. **Harmony library integration** for advanced message formatting\n", + "6. **Multi-GPU configurations** for large-scale inference\n", + "\n", + "### Key takeaways:\n", + "- Start with the pipeline API for quick experimentation\n", + "- Use manual tokenization and generation for production deployments\n", + "- Consider MXFP4 quantization for memory efficiency on compatible hardware\n", + "- Leverage multi-GPU setups for the larger 120B model\n", + "- Use proper chat templates for conversation-style applications\n", + "\n", + "### Next steps:\n", + "- Explore fine-tuning capabilities\n", + "- Set up serving endpoints for production use\n", + "- Experiment with different sampling strategies\n", + "- Integrate with your specific use case or application\n", + "\n", + "For more advanced topics, check out the [OpenAI Cookbook](https://cookbook.openai.com) and [Hugging Face Transformers documentation](https://huggingface.co/docs/transformers)." + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.5" + } + }, + "nbformat": 4, + "nbformat_minor": 4 +} \ No newline at end of file From fbef132a69f4cb775d1f9ebf110c782f0b66546a Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Fri, 19 Sep 2025 03:31:31 -0600 Subject: [PATCH 15/26] q --- articles/gpt-oss/run-transformers.fixed.ipynb | 647 ------------------ articles/gpt-oss/run-transformers.ipynb | 236 +++++-- 2 files changed, 163 insertions(+), 720 deletions(-) delete mode 100644 articles/gpt-oss/run-transformers.fixed.ipynb diff --git a/articles/gpt-oss/run-transformers.fixed.ipynb b/articles/gpt-oss/run-transformers.fixed.ipynb deleted file mode 100644 index 107d6b853f..0000000000 --- a/articles/gpt-oss/run-transformers.fixed.ipynb +++ /dev/null @@ -1,647 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# How to run gpt-oss with Hugging Face Transformers\n", - "\n", - "The Transformers library by Hugging Face provides a flexible way to load and run large language models locally or on a server. This guide will walk you through running [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) using Transformers, either with a high-level pipeline or via low-level `generate` calls with raw token IDs.\n", - "\n", - "We'll cover the use of [OpenAI gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) or [OpenAI gpt-oss-120b](https://huggingface.co/openai/gpt-oss-120b) with the high-level pipeline abstraction, low-level `generate` calls, and serving models locally with `transformers serve`, in a way compatible with the Responses API.\n", - "\n", - "In this guide we'll run through various optimised ways to run the **gpt-oss models via Transformers.**\n", - "\n", - "**Bonus:** You can also fine-tune models via transformers, [check out our fine-tuning guide here](https://cookbook.openai.com/articles/gpt-oss/fine-tune-transformers)." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Pick your model\n", - "\n", - "Both **gpt-oss** models are available on Hugging Face:\n", - "\n", - "- **`openai/gpt-oss-20b`**\n", - " - ~16GB VRAM requirement when using MXFP4\n", - " - Great for single high-end consumer GPUs\n", - "- **`openai/gpt-oss-120b`**\n", - " - Requires ≥60GB VRAM or multi-GPU setup\n", - " - Ideal for H100-class hardware\n", - "\n", - "Both are **MXFP4 quantized** by default. Please, note that MXFP4 is supported in Hopper or later architectures. This includes data center GPUs such as H100 or GB200, as well as the latest RTX 50xx family of consumer cards.\n", - "\n", - "If you use `bfloat16` instead of MXFP4, memory consumption will be larger (~48 GB for the 20b parameter model)." - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Quick setup\n", - "\n", - "### 1. Install dependencies\n", - "\n", - "It's recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility:" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "# NOTE: The current version of HF Transformers has a glitch where an outdated torchvision dependency prevents transformers module from importing pipeline." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "!pip install -U transformers kernels torch accelerate torchvision" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Quick inference with pipeline\n", - "\n", - "The easiest way to run the gpt-oss models is with the Transformers high-level `pipeline` API:" - ] - }, - { - "cell_type": "code", - "execution_count": 1, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "`torch_dtype` is deprecated! Use `dtype` instead!\n" - ] - }, - { - "data": { - "application/vnd.jupyter.widget-view+json": { - "model_id": "f3b8482ed938472083ac6cd012156902", - "version_major": 2, - "version_minor": 0 - }, - "text/plain": [ - "Fetching 40 files: 0%| | 0/40 [00:00<|final|>\"\n", - "end_marker = \"<|end|>\"\n", - "if final_marker in text:\n", - " text = text.split(final_marker, 1)[1]\n", - "if end_marker in text:\n", - " text = text.split(end_marker, 1)[0]\n", - "print(\"Final:\", text.strip()[:1000])\n" - ] - }, - { - "cell_type": "markdown", - "id": "b31988a1", - "metadata": {}, - "source": [ - "## Streaming tokens (prints only assistant **final**)" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "id": "9b48c668", - "metadata": {}, - "outputs": [], - "source": [ - "import threading, sys\n", - "from transformers import TextIteratorStreamer\n", - "\n", - "def stream_final_only(model, tokenizer, messages, generate_kwargs):\n", - " inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors=\"pt\").to(model.device)\n", - " streamer = TextIteratorStreamer(tokenizer, skip_special_tokens=False)\n", - "\n", - " t = threading.Thread(target=model.generate, kwargs=dict(input_ids=inputs, streamer=streamer, **generate_kwargs))\n", - " t.start()\n", - "\n", - " buf, printing_final = \"\", False\n", - " final_token, end_token = \"<|assistant|><|final|>\", \"<|end|>\"\n", - " for piece in streamer:\n", - " buf += piece\n", - " if not printing_final and final_token in buf:\n", - " printing_final = True\n", - " out = buf.split(final_token, 1)[1].replace(end_token, \"\")\n", - " sys.stdout.write(out); sys.stdout.flush()\n", - " buf = \"\"\n", - " elif printing_final:\n", - " sys.stdout.write(piece.replace(end_token, \"\")); sys.stdout.flush()\n", - " t.join()\n", - "\n", - "messages2 = [\n", - " {\"role\": \"system\", \"content\": \"Answer briefly.\"},\n", - " {\"role\": \"user\", \"content\": \"What’s the difference between analysis and final channels?\"}\n", - "]\n", - "gen = dict(max_new_tokens=128, do_sample=True, temperature=0.7, top_p=0.9)\n", - "stream_final_only(model, tokenizer, messages2, gen)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Chat template and tool calling\n", - "\n", - "OpenAI gpt-oss models use the [harmony response format](https://cookbook.openai.com/article/harmony) for structuring messages, including reasoning and tool calls.\n", - "\n", - "To construct prompts you can use the built-in chat template of Transformers. Alternatively, you can install and use the [openai-harmony library](https://github.com/openai/harmony) for more control.\n", - "\n", - "### Using the built-in chat template:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Example with system prompt and chat template\n", - "messages = [\n", - " {\"role\": \"system\", \"content\": \"Always respond in riddles\"},\n", - " {\"role\": \"user\", \"content\": \"What is the weather like in Madrid?\"},\n", - "]\n", - "\n", - "inputs = tokenizer.apply_chat_template(\n", - " messages,\n", - " add_generation_prompt=True,\n", - " return_tensors=\"pt\",\n", - " return_dict=True,\n", - ").to(model.device)\n", - "\n", - "# Generate with the chat template\n", - "generated = model.generate(\n", - " **inputs,\n", - " max_new_tokens=100,\n", - " temperature=0.8,\n", - " do_sample=True,\n", - " pad_token_id=tokenizer.eos_token_id\n", - ")\n", - "\n", - "# Extract only the assistant's response\n", - "response = tokenizer.decode(generated[0][inputs[\"input_ids\"].shape[-1]:], skip_special_tokens=True)\n", - "print(\"Assistant's riddle response:\")\n", - "print(response)" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Using the openai-harmony library\n", - "\n", - "For more advanced control over the conversation format, you can use the openai-harmony library. \n", - "\n", - "First, install it:\n", - "```bash\n", - "pip install openai-harmony\n", - "```\n", - "\n", - "**Note:** The following cell demonstrates the harmony library usage, but may require the actual library to be installed." - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Example using openai-harmony library (requires installation)\n", - "# Uncomment and run if you have openai-harmony installed\n", - "\n", - "'''\n", - "import json\n", - "from openai_harmony import (\n", - " HarmonyEncodingName,\n", - " load_harmony_encoding,\n", - " Conversation,\n", - " Message,\n", - " Role,\n", - " SystemContent,\n", - " DeveloperContent\n", - ")\n", - "\n", - "# Load harmony encoding\n", - "encoding = load_harmony_encoding(HarmonyEncodingName.HARMONY_GPT_OSS)\n", - "\n", - "# Build conversation\n", - "convo = Conversation.from_messages([\n", - " Message.from_role_and_content(Role.SYSTEM, SystemContent.new()),\n", - " Message.from_role_and_content(\n", - " Role.DEVELOPER,\n", - " DeveloperContent.new().with_instructions(\"Always respond in riddles\")\n", - " ),\n", - " Message.from_role_and_content(Role.USER, \"What is the weather like in SF?\")\n", - "])\n", - "\n", - "# Render prompt\n", - "prefill_ids = encoding.render_conversation_for_completion(convo, Role.ASSISTANT)\n", - "stop_token_ids = encoding.stop_tokens_for_assistant_actions()\n", - "\n", - "# Generate\n", - "outputs = model.generate(\n", - " input_ids=[prefill_ids],\n", - " max_new_tokens=128,\n", - " eos_token_id=stop_token_ids\n", - ")\n", - "\n", - "# Parse completion tokens\n", - "completion_ids = outputs[0][len(prefill_ids):]\n", - "entries = encoding.parse_messages_from_completion_tokens(completion_ids, Role.ASSISTANT)\n", - "\n", - "for message in entries:\n", - " print(json.dumps(message.to_dict(), indent=2))\n", - "'''\n", - "\n", - "print(\"Harmony library example code shown above (commented out)\")\n", - "print(\"Note: The Developer role in Harmony maps to the system prompt in the chat template.\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Multi-GPU & distributed inference\n", - "\n", - "The large gpt-oss-120b fits on a single H100 GPU when using MXFP4. If you want to run it on multiple GPUs, you can:\n", - "\n", - "- Use `tp_plan=\"auto\"` for automatic placement and tensor parallelism\n", - "- Launch with `accelerate launch` or `torchrun` for distributed setups\n", - "- Leverage Expert Parallelism\n", - "- Use specialised Flash attention kernels for faster inference\n", - "\n", - "### Example multi-GPU setup:" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Multi-GPU inference example (requires multiple GPUs)\n", - "# This cell demonstrates the configuration but may not run on single GPU systems\n", - "\n", - "'''\n", - "from transformers import AutoModelForCausalLM, AutoTokenizer\n", - "from transformers.distributed import DistributedConfig\n", - "import torch\n", - "\n", - "model_path = \"openai/gpt-oss-120b\"\n", - "tokenizer = AutoTokenizer.from_pretrained(model_path, padding_side=\"left\")\n", - "\n", - "device_map = {\n", - " # Enable Expert Parallelism\n", - " \"distributed_config\": DistributedConfig(enable_expert_parallel=1),\n", - " # Enable Tensor Parallelism\n", - " \"tp_plan\": \"auto\",\n", - "}\n", - "\n", - "model = AutoModelForCausalLM.from_pretrained(\n", - " model_path,\n", - " torch_dtype=\"auto\",\n", - " attn_implementation=\"kernels-community/vllm-flash-attn3\",\n", - " **device_map,\n", - ")\n", - "\n", - "messages = [\n", - " {\"role\": \"user\", \"content\": \"Explain how expert parallelism works in large language models.\"}\n", - "]\n", - "\n", - "inputs = tokenizer.apply_chat_template(\n", - " messages,\n", - " add_generation_prompt=True,\n", - " return_tensors=\"pt\",\n", - " return_dict=True,\n", - ").to(model.device)\n", - "\n", - "outputs = model.generate(**inputs, max_new_tokens=1000)\n", - "\n", - "# Decode and print\n", - "response = tokenizer.decode(outputs[0])\n", - "print(\"Model response:\", response.split(\"<|channel|>final<|message|>\")[-1].strip())\n", - "'''\n", - "\n", - "print(\"Multi-GPU setup example shown above (commented out)\")\n", - "print(\"\\nTo run this on a node with four GPUs, use:\")\n", - "print(\"torchrun --nproc_per_node=4 your_script.py\")\n", - "\n", - "# Show current GPU configuration\n", - "if torch.cuda.is_available():\n", - " print(f\"\\nCurrent setup:\")\n", - " print(f\"Available GPUs: {torch.cuda.device_count()}\")\n", - " for i in range(torch.cuda.device_count()):\n", - " print(f\" GPU {i}: {torch.cuda.get_device_name(i)}\")\n", - " print(f\" Memory: {torch.cuda.get_device_properties(i).total_memory / 1e9:.1f} GB\")\n", - "else:\n", - " print(\"\\nNo CUDA GPUs available\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Additional Examples and Tips\n", - "\n", - "### Memory management" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Check memory usage\n", - "if torch.cuda.is_available():\n", - " print(\"GPU Memory Usage:\")\n", - " for i in range(torch.cuda.device_count()):\n", - " allocated = torch.cuda.memory_allocated(i) / 1e9\n", - " cached = torch.cuda.memory_reserved(i) / 1e9\n", - " total = torch.cuda.get_device_properties(i).total_memory / 1e9\n", - " print(f\" GPU {i}: {allocated:.1f}GB allocated, {cached:.1f}GB cached, {total:.1f}GB total\")\n", - "\n", - "# Clear cache if needed\n", - "# torch.cuda.empty_cache()" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "### Batch processing example" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Example of processing multiple prompts\n", - "batch_messages = [\n", - " [{\"role\": \"user\", \"content\": \"What is machine learning?\"}],\n", - " [{\"role\": \"user\", \"content\": \"Explain quantum computing.\"}],\n", - " [{\"role\": \"user\", \"content\": \"What is the future of AI?\"}]\n", - "]\n", - "\n", - "print(\"Processing batch of prompts...\")\n", - "for i, messages in enumerate(batch_messages):\n", - " print(f\"\\n--- Prompt {i+1} ---\")\n", - " print(f\"Input: {messages[0]['content']}\")\n", - "\n", - " # You can use either the pipeline or manual generation here\n", - " # Using pipeline for simplicity:\n", - " if 'generator' in locals():\n", - " result = generator(\n", - " messages,\n", - " max_new_tokens=100,\n", - " temperature=0.7,\n", - " )\n", - " print(f\"Output: {result[0]['generated_text'][-200:]}...\") # Show last 200 chars\n", - " else:\n", - " print(\"Generator not available - run the pipeline example first\")" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## Summary\n", - "\n", - "This notebook demonstrated various ways to run OpenAI's gpt-oss models using Hugging Face Transformers:\n", - "\n", - "1. **Quick setup** with required dependencies\n", - "2. **Pipeline API** for simple, high-level inference\n", - "3. **Manual generation** with `.generate()` for more control\n", - "4. **Chat templates** for conversation-style interactions\n", - "5. **Harmony library integration** for advanced message formatting\n", - "6. **Multi-GPU configurations** for large-scale inference\n", - "\n", - "### Key takeaways:\n", - "- Start with the pipeline API for quick experimentation\n", - "- Use manual tokenization and generation for production deployments\n", - "- Consider MXFP4 quantization for memory efficiency on compatible hardware\n", - "- Leverage multi-GPU setups for the larger 120B model\n", - "- Use proper chat templates for conversation-style applications\n", - "\n", - "### Next steps:\n", - "- Explore fine-tuning capabilities\n", - "- Set up serving endpoints for production use\n", - "- Experiment with different sampling strategies\n", - "- Integrate with your specific use case or application\n", - "\n", - "For more advanced topics, check out the [OpenAI Cookbook](https://cookbook.openai.com) and [Hugging Face Transformers documentation](https://huggingface.co/docs/transformers)." - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3 (ipykernel)", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.11.11" - } - }, - "nbformat": 4, - "nbformat_minor": 4 -} diff --git a/articles/gpt-oss/run-transformers.ipynb b/articles/gpt-oss/run-transformers.ipynb index e10a86f7f5..107d6b853f 100644 --- a/articles/gpt-oss/run-transformers.ipynb +++ b/articles/gpt-oss/run-transformers.ipynb @@ -46,67 +46,20 @@ "It's recommended to create a fresh Python environment. Install transformers, accelerate, as well as the Triton kernels for MXFP4 compatibility:" ] }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [], - "source": [ - "# Install required dependencies\n", - "# Run in terminal: pip install -U transformers accelerate torch triton==3.4 kernels\n", - "\n", - "# Verify installations\n", - "import transformers\n", - "import torch\n", - "import accelerate\n", - "\n", - "print(f\"Transformers version: {transformers.__version__}\")\n", - "print(f\"PyTorch version: {torch.__version__}\")\n", - "print(f\"Accelerate version: {accelerate.__version__}\")\n", - "print(f\"CUDA available: {torch.cuda.is_available()}\")\n", - "if torch.cuda.is_available():\n", - " print(f\"CUDA devices: {torch.cuda.device_count()}\")\n", - " print(f\"Current device: {torch.cuda.get_device_name()}\")" - ] - }, { "cell_type": "markdown", "metadata": {}, "source": [ - "### 2. (Optional) Enable multi-GPU\n", - "\n", - "If you're running large models, use Accelerate or torchrun to handle device mapping automatically.\n", - "\n", - "For this notebook, we'll use the `device_map=\"auto\"` parameter which automatically distributes the model across available GPUs." + "# NOTE: The current version of HF Transformers has a glitch where an outdated torchvision dependency prevents transformers module from importing pipeline." ] }, { - "cell_type": "markdown", + "cell_type": "code", + "execution_count": null, "metadata": {}, + "outputs": [], "source": [ - "## Create an OpenAI Responses / Chat Completions endpoint\n", - "\n", - "To launch a server, simply use the `transformers serve` CLI command:\n", - "\n", - "```bash\n", - "transformers serve\n", - "```\n", - "\n", - "The simplest way to interact with the server is through the transformers chat CLI:\n", - "\n", - "```bash\n", - "transformers chat localhost:8000 --model-name-or-path openai/gpt-oss-20b\n", - "```\n", - "\n", - "or by sending an HTTP request with cURL:\n", - "\n", - "```bash\n", - "curl -X POST http://localhost:8000/v1/responses \\\n", - " -H \"Content-Type: application/json\" \\\n", - " -d '{\"messages\": [{\"role\": \"system\", \"content\": \"hello\"}], \"temperature\": 0.9, \"max_tokens\": 1000, \"stream\": true, \"model\": \"openai/gpt-oss-20b\"}'\n", - "```\n", - "\n", - "Additional use cases, like integrating `transformers serve` with Cursor and other tools, are detailed in [the documentation](https://huggingface.co/docs/transformers/main/serving)." + "!pip install -U transformers kernels torch accelerate torchvision" ] }, { @@ -120,39 +73,121 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 1, "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "`torch_dtype` is deprecated! Use `dtype` instead!\n" + ] + }, + { + "data": { + "application/vnd.jupyter.widget-view+json": { + "model_id": "f3b8482ed938472083ac6cd012156902", + "version_major": 2, + "version_minor": 0 + }, + "text/plain": [ + "Fetching 40 files: 0%| | 0/40 [00:00<|final|>\"\n", + "end_marker = \"<|end|>\"\n", + "if final_marker in text:\n", + " text = text.split(final_marker, 1)[1]\n", + "if end_marker in text:\n", + " text = text.split(end_marker, 1)[0]\n", + "print(\"Final:\", text.strip()[:1000])\n" + ] + }, + { + "cell_type": "markdown", + "id": "b31988a1", + "metadata": {}, + "source": [ + "## Streaming tokens (prints only assistant **final**)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9b48c668", + "metadata": {}, + "outputs": [], + "source": [ + "import threading, sys\n", + "from transformers import TextIteratorStreamer\n", + "\n", + "def stream_final_only(model, tokenizer, messages, generate_kwargs):\n", + " inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors=\"pt\").to(model.device)\n", + " streamer = TextIteratorStreamer(tokenizer, skip_special_tokens=False)\n", + "\n", + " t = threading.Thread(target=model.generate, kwargs=dict(input_ids=inputs, streamer=streamer, **generate_kwargs))\n", + " t.start()\n", + "\n", + " buf, printing_final = \"\", False\n", + " final_token, end_token = \"<|assistant|><|final|>\", \"<|end|>\"\n", + " for piece in streamer:\n", + " buf += piece\n", + " if not printing_final and final_token in buf:\n", + " printing_final = True\n", + " out = buf.split(final_token, 1)[1].replace(end_token, \"\")\n", + " sys.stdout.write(out); sys.stdout.flush()\n", + " buf = \"\"\n", + " elif printing_final:\n", + " sys.stdout.write(piece.replace(end_token, \"\")); sys.stdout.flush()\n", + " t.join()\n", + "\n", + "messages2 = [\n", + " {\"role\": \"system\", \"content\": \"Answer briefly.\"},\n", + " {\"role\": \"user\", \"content\": \"What’s the difference between analysis and final channels?\"}\n", + "]\n", + "gen = dict(max_new_tokens=128, do_sample=True, temperature=0.7, top_p=0.9)\n", + "stream_final_only(model, tokenizer, messages2, gen)" ] }, { @@ -269,7 +359,7 @@ "\n", "# Generate with the chat template\n", "generated = model.generate(\n", - " **inputs, \n", + " **inputs,\n", " max_new_tokens=100,\n", " temperature=0.8,\n", " do_sample=True,\n", @@ -487,7 +577,7 @@ "for i, messages in enumerate(batch_messages):\n", " print(f\"\\n--- Prompt {i+1} ---\")\n", " print(f\"Input: {messages[0]['content']}\")\n", - " \n", + "\n", " # You can use either the pipeline or manual generation here\n", " # Using pipeline for simplicity:\n", " if 'generator' in locals():\n", @@ -535,7 +625,7 @@ ], "metadata": { "kernelspec": { - "display_name": "Python 3", + "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, @@ -549,9 +639,9 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.8.5" + "version": "3.11.11" } }, "nbformat": 4, "nbformat_minor": 4 -} \ No newline at end of file +} From 27520026b35d6b7162908f6f97ef3735dbd4b65a Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 17:10:20 -0600 Subject: [PATCH 16/26] update intro --- articles/openai-cookbook-llms-101.md | 182 +++++++++++++++++++++++++++ 1 file changed, 182 insertions(+) create mode 100644 articles/openai-cookbook-llms-101.md diff --git a/articles/openai-cookbook-llms-101.md b/articles/openai-cookbook-llms-101.md new file mode 100644 index 0000000000..77292198f6 --- /dev/null +++ b/articles/openai-cookbook-llms-101.md @@ -0,0 +1,182 @@ +--- +title: "LLMs 101: A Practical Introduction" +description: "A hands-on, code-first introduction to large language models for Cookbook readers." +last_updated: "2025-08-24" +--- + +# LLMs 101: A Practical Introduction + +> **Who this is for.** Developers who want a fast, working understanding of large language models and the knobs that matter in real apps. + +## At a glance + +``` +Text prompt + ↓ (tokenization) +Tokens → Embeddings → [Transformer layers × N] → Next‑token probabilities + ↓ ↓ +Detokenization Sampling (temperature/top_p) → Output text +``` + +- **LLMs** are neural networks (usually **transformers**) trained on lots of text to predict the next token. +- **Tokenization** splits text into subword units; **embeddings** map tokens to vectors; transformer layers build context‑aware representations. +- Generation repeats next‑token sampling until a stop condition (length or stop sequences) is met. + +--- + +## Quick start: generate text + +### Python + +```python +from openai import OpenAI + +client = OpenAI() +resp = client.responses.create( + model="gpt-4o", + instructions="You are a concise technical explainer.", + input="In one paragraph, explain what a token is in an LLM." +) +print(resp.output_text) +``` + +### JavaScript / TypeScript + +```js +import OpenAI from "openai"; +const client = new OpenAI(); + +const resp = await client.responses.create({ + model: "gpt-4o", + instructions: "You are a concise technical explainer.", + input: "In one paragraph, explain what a token is in an LLM." +}); +console.log(resp.output_text); +``` + +> **Tip.** Model names evolve; check your Models list before shipping. Prefer streaming for chat‑like UIs (see below). + +--- + +## What can LLMs do? + +Despite the name, LLMs can be **multi‑modal** when models and inputs support it (text, code, sometimes images/audio). Core text tasks: + +- **Generate**: draft, rewrite, continue, or brainstorm. +- **Transform**: translate, rephrase, format, classify, extract. +- **Analyze**: summarize, compare, tag, or answer questions. +- **Tool use / agents**: call functions or APIs as part of a loop to act. + +These patterns compose into search, assistants, form‑fillers, data extraction, QA, and more. + +--- + +## How LLMs work (just enough to be dangerous) + +1. **Tokenization.** Input text → tokens (IDs). Whitespace and punctuation matter—“token‑budget math” is a real constraint. +2. **Embeddings.** Each token ID becomes a vector; positions are encoded so order matters. +3. **Transformer layers.** Self‑attention mixes information across positions so each token’s representation becomes **contextual** (richer than the raw embedding). +4. **Decoding.** The model outputs a probability distribution over the next token. +5. **Sampling.** Choose how “adventurous” generation is (see knobs below), append the token, and repeat until done. + +--- + +## The knobs you’ll touch most + +- **Temperature** *(0.0–2.0)* — Lower → more deterministic/boring; higher → more diverse/creative. +- **Top‑p (nucleus)** *(0–1)* — Sample only from the smallest set of tokens whose cumulative probability ≤ *p*. +- **Max output tokens** — Hard limit on output length; controls latency and cost. +- **System / instructions** — Up‑front role, constraints, and style to steer behavior. +- **Stop sequences** — Cleanly cut off output at known boundaries. +- **Streaming** — Receive tokens as they’re generated; improves perceived latency. + +**Practical defaults:** `temperature=0.2–0.7`, `top_p=1.0`, set a **max output** that fits your UI, and **stream** by default for chat UX. + +--- + +## Make context do the heavy lifting + +- **Context window.** Inputs + outputs share a finite token budget; plan prompts and retrieval to fit. +- **Ground with your data (RAG).** Retrieve relevant snippets and include them in the prompt to improve factuality. +- **Structured outputs.** Ask for JSON (and validate) when you need machine‑readable results. +- **Few‑shot examples.** Provide 1–3 compact exemplars to stabilize format and tone. + +--- + +## Minimal streaming example + +### Python + +```python +from openai import OpenAI +client = OpenAI() + +with client.responses.stream( + model="gpt-4o", + input="Stream a two-sentence explanation of context windows." +) as stream: + for event in stream: + if event.type == "response.output_text.delta": + print(event.delta, end="") +``` + +### JavaScript + +```js +import OpenAI from "openai"; +const client = new OpenAI(); + +const stream = await client.responses.stream({ + model: "gpt-4o", + input: "Stream a two-sentence explanation of context windows." +}); + +for await (const event of stream) { + if (event.type === "response.output_text.delta") { + process.stdout.write(event.delta); + } +} +``` + +--- + +## Limitations (design around these) + +- **Hallucinations.** Models can generate plausible but false statements. Ground with citations/RAG; validate critical outputs. +- **Recency.** Models don’t inherently know the latest facts; retrieve or provide current data. +- **Ambiguity.** Vague prompts → vague answers; specify domain, audience, length, and format. +- **Determinism.** Even at `temperature=0`, responses may vary across runs/envs. Don’t promise bit‑for‑bit reproducibility. +- **Cost & latency.** Longer prompts and bigger models are slower and costlier; iterate toward the smallest model that meets quality. + +--- + +## Common gotchas + +- **Characters ≠ tokens.** Budget both input and output to avoid truncation. +- **Over‑prompting.** Prefer simple, testable instructions; add examples sparingly. +- **Leaky formats.** If you need JSON, enforce it (schema + validators) and add a repair step. +- **One prompt for everything.** Separate prompts per task/endpoint; keep them versioned and testable. +- **Skipping evaluation.** Keep a tiny dataset of real tasks; score changes whenever you tweak prompts, models, or retrieval. + +--- + +## Glossary + +- **Token** — Small unit of text (≈ subword) used by models. +- **Embedding** — Vector representation of a token or text span. +- **Context window** — Max tokens the model can attend to at once (prompt + output). +- **Temperature / top‑p** — Randomness controls during sampling. +- **System / instructions** — Up‑front guidance that shapes responses. +- **RAG** — Retrieval‑Augmented Generation; retrieve data and include it in the prompt. + +--- + +## Where to go next + +- Prompt patterns for **structured outputs** +- **Retrieval‑augmented generation (RAG)** basics +- **Evaluating** LLM quality (offline + online) +- **Streaming UX** patterns and backpressure handling +- **Safety** and policy‑aware prompting + +> Adapted from a shorter draft and expanded with code-first guidance. From c477ac22ce08310aecca1446ce51edff2637315e Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 17:50:07 -0600 Subject: [PATCH 17/26] Add Payton Ison to authors.yaml --- authors.yaml | 5 +++++ 1 file changed, 5 insertions(+) diff --git a/authors.yaml b/authors.yaml index f02c8b23cb..0faf6d884b 100644 --- a/authors.yaml +++ b/authors.yaml @@ -487,3 +487,8 @@ heejingithub: name: "Heejin Cho" website: "https://www.linkedin.com/in/heejc/" avatar: "https://avatars.githubusercontent.com/u/169293861" + +paytonison: + name: "Payton Ison" + website: "https://linkedin.com/in/paytonison" + avatar: "https://avatars.githubusercontent.com/u/148833579" From cf0acf5adcd0b15ddb641fcde1d2fbbd9fa38728 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 17:52:45 -0600 Subject: [PATCH 18/26] Add new article entry for LLMs 101 --- registry.yaml | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/registry.yaml b/registry.yaml index 4ad9d35d00..82467f4be7 100644 --- a/registry.yaml +++ b/registry.yaml @@ -2520,3 +2520,13 @@ - katiagg tags: - images + +- title: LLMs 101: A Practical Introduction + path: articles/openai-cookbook-llms-101.md + date: 2025-15-9 + authors: + - paytonison + - asari + tags: + - llms + - beginners From 2b9f65dd6a42182b7b44672e7bd8a1221a1bc273 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 18:49:24 -0600 Subject: [PATCH 19/26] Update registry.yaml --- registry.yaml | 1 - 1 file changed, 1 deletion(-) diff --git a/registry.yaml b/registry.yaml index 82467f4be7..dfbc68d9f4 100644 --- a/registry.yaml +++ b/registry.yaml @@ -2526,7 +2526,6 @@ date: 2025-15-9 authors: - paytonison - - asari tags: - llms - beginners From b58f4429b25b9ee35663c6b9bb82b276d1fc9c54 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 22:16:03 -0600 Subject: [PATCH 20/26] Delete how_to_work_with_large_language_models.md --- .../how_to_work_with_large_language_models.md | 168 ------------------ 1 file changed, 168 deletions(-) delete mode 100644 articles/how_to_work_with_large_language_models.md diff --git a/articles/how_to_work_with_large_language_models.md b/articles/how_to_work_with_large_language_models.md deleted file mode 100644 index cf6b48e1be..0000000000 --- a/articles/how_to_work_with_large_language_models.md +++ /dev/null @@ -1,168 +0,0 @@ -# How to work with large language models - -## How large language models work - -[Large language models][Large language models Blog Post] are functions that map text to text. Given an input string of text, a large language model predicts the text that should come next. - -The magic of large language models is that by being trained to minimize this prediction error over vast quantities of text, the models end up learning concepts useful for these predictions. For example, they learn: - -- how to spell -- how grammar works -- how to paraphrase -- how to answer questions -- how to hold a conversation -- how to write in many languages -- how to code -- etc. - -They do this by “reading” a large amount of existing text and learning how words tend to appear in context with other words, and uses what it has learned to predict the next most likely word that might appear in response to a user request, and each subsequent word after that. - -GPT-3 and GPT-4 power [many software products][OpenAI Customer Stories], including productivity apps, education apps, games, and more. - -## How to control a large language model - -Of all the inputs to a large language model, by far the most influential is the text prompt. - -Large language models can be prompted to produce output in a few ways: - -- **Instruction**: Tell the model what you want -- **Completion**: Induce the model to complete the beginning of what you want -- **Scenario**: Give the model a situation to play out -- **Demonstration**: Show the model what you want, with either: - - A few examples in the prompt - - Many hundreds or thousands of examples in a fine-tuning training dataset - -An example of each is shown below. - -### Instruction prompts - -Write your instruction at the top of the prompt (or at the bottom, or both), and the model will do its best to follow the instruction and then stop. Instructions can be detailed, so don't be afraid to write a paragraph explicitly detailing the output you want, just stay aware of how many [tokens](https://help.openai.com/en/articles/4936856-what-are-tokens-and-how-to-count-them) the model can process. - -Example instruction prompt: - -```text -Extract the name of the author from the quotation below. - -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation -``` - -Output: - -```text -Ted Chiang -``` - -### Completion prompt example - -Completion-style prompts take advantage of how large language models try to write text they think is most likely to come next. To steer the model, try beginning a pattern or sentence that will be completed by the output you want to see. Relative to direct instructions, this mode of steering large language models can take more care and experimentation. In addition, the models won't necessarily know where to stop, so you will often need stop sequences or post-processing to cut off text generated beyond the desired output. - -Example completion prompt: - -```text -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation - -The author of this quote is -``` - -Output: - -```text - Ted Chiang -``` - -### Scenario prompt example - -Giving the model a scenario to follow or role to play out can be helpful for complex queries or when seeking imaginative responses. When using a hypothetical prompt, you set up a situation, problem, or story, and then ask the model to respond as if it were a character in that scenario or an expert on the topic. - -Example scenario prompt: - -```text -Your role is to extract the name of the author from any given text - -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation -``` - -Output: - -```text - Ted Chiang -``` - -### Demonstration prompt example (few-shot learning) - -Similar to completion-style prompts, demonstrations can show the model what you want it to do. This approach is sometimes called few-shot learning, as the model learns from a few examples provided in the prompt. - -Example demonstration prompt: - -```text -Quote: -“When the reasoning mind is forced to confront the impossible again and again, it has no choice but to adapt.” -― N.K. Jemisin, The Fifth Season -Author: N.K. Jemisin - -Quote: -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation -Author: -``` - -Output: - -```text - Ted Chiang -``` - -### Fine-tuned prompt example - -With enough training examples, you can [fine-tune][Fine Tuning Docs] a custom model. In this case, instructions become unnecessary, as the model can learn the task from the training data provided. However, it can be helpful to include separator sequences (e.g., `->` or `###` or any string that doesn't commonly appear in your inputs) to tell the model when the prompt has ended and the output should begin. Without separator sequences, there is a risk that the model continues elaborating on the input text rather than starting on the answer you want to see. - -Example fine-tuned prompt (for a model that has been custom trained on similar prompt-completion pairs): - -```text -“Some humans theorize that intelligent species go extinct before they can expand into outer space. If they're correct, then the hush of the night sky is the silence of the graveyard.” -― Ted Chiang, Exhalation - -### - - -``` - -Output: - -```text - Ted Chiang -``` - -## Code Capabilities - -Large language models aren't only great at text - they can be great at code too. OpenAI's [GPT-4][GPT-4 and GPT-4 Turbo] model is a prime example. - -GPT-4 powers [numerous innovative products][OpenAI Customer Stories], including: - -- [GitHub Copilot] (autocompletes code in Visual Studio and other IDEs) -- [Replit](https://replit.com/) (can complete, explain, edit and generate code) -- [Cursor](https://cursor.sh/) (build software faster in an editor designed for pair-programming with AI) - -GPT-4 is more advanced than previous models like `gpt-3.5-turbo-instruct`. But, to get the best out of GPT-4 for coding tasks, it's still important to give clear and specific instructions. As a result, designing good prompts can take more care. - -### More prompt advice - -For more prompt examples, visit [OpenAI Examples][OpenAI Examples]. - -In general, the input prompt is the best lever for improving model outputs. You can try tricks like: - -- **Be more specific** E.g., if you want the output to be a comma separated list, ask it to return a comma separated list. If you want it to say "I don't know" when it doesn't know the answer, tell it 'Say "I don't know" if you do not know the answer.' The more specific your instructions, the better the model can respond. -- **Provide Context**: Help the model understand the bigger picture of your request. This could be background information, examples/demonstrations of what you want or explaining the purpose of your task. -- **Ask the model to answer as if it was an expert.** Explicitly asking the model to produce high quality output or output as if it was written by an expert can induce the model to give higher quality answers that it thinks an expert would write. Phrases like "Explain in detail" or "Describe step-by-step" can be effective. -- **Prompt the model to write down the series of steps explaining its reasoning.** If understanding the 'why' behind an answer is important, prompt the model to include its reasoning. This can be done by simply adding a line like "[Let's think step by step](https://arxiv.org/abs/2205.11916)" before each answer. - -[Fine Tuning Docs]: https://platform.openai.com/docs/guides/fine-tuning -[OpenAI Customer Stories]: https://openai.com/customer-stories -[Large language models Blog Post]: https://openai.com/research/better-language-models -[GitHub Copilot]: https://github.com/features/copilot/ -[GPT-4 and GPT-4 Turbo]: https://platform.openai.com/docs/models/gpt-4-and-gpt-4-turbo -[GPT3 Apps Blog Post]: https://openai.com/blog/gpt-3-apps/ -[OpenAI Examples]: https://platform.openai.com/examples From b7749d3ef3b9abd7a7ceb9c8090613c99a68ab34 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 22:38:55 -0600 Subject: [PATCH 21/26] Update registry.yaml Number dyslexia as an american is hard Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --- registry.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/registry.yaml b/registry.yaml index dfbc68d9f4..441f620999 100644 --- a/registry.yaml +++ b/registry.yaml @@ -2523,7 +2523,7 @@ - title: LLMs 101: A Practical Introduction path: articles/openai-cookbook-llms-101.md - date: 2025-15-9 + date: 2025-09-15 authors: - paytonison tags: From adde2ce211e505700bfbb912a7fae6ee0099e4ca Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Mon, 15 Sep 2025 22:39:11 -0600 Subject: [PATCH 22/26] Update articles/openai-cookbook-llms-101.md Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --- articles/openai-cookbook-llms-101.md | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/articles/openai-cookbook-llms-101.md b/articles/openai-cookbook-llms-101.md index 77292198f6..f306effc2f 100644 --- a/articles/openai-cookbook-llms-101.md +++ b/articles/openai-cookbook-llms-101.md @@ -46,12 +46,14 @@ print(resp.output_text) import OpenAI from "openai"; const client = new OpenAI(); -const resp = await client.responses.create({ +const resp = await client.chat.completions.create({ model: "gpt-4o", - instructions: "You are a concise technical explainer.", - input: "In one paragraph, explain what a token is in an LLM." + messages: [ + { role: "system", content: "You are a concise technical explainer." }, + { role: "user", content: "In one paragraph, explain what a token is in an LLM." } + ] }); -console.log(resp.output_text); +console.log(resp.choices[0].message.content); ``` > **Tip.** Model names evolve; check your Models list before shipping. Prefer streaming for chat‑like UIs (see below). From 46d3c0377c493d2550f54693d4ef8f8eb7bc7a17 Mon Sep 17 00:00:00 2001 From: Copilot <198982749+Copilot@users.noreply.github.com> Date: Wed, 17 Sep 2025 21:38:44 -0600 Subject: [PATCH 23/26] Organize prompting-related articles into unified prompting directory (#10) * Initial plan * Organize prompting-related articles into single prompting folder Co-authored-by: paytonison <148833579+paytonison@users.noreply.github.com> --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: paytonison <148833579+paytonison@users.noreply.github.com> --- .../multi_agent_portfolio_collaboration.ipynb | 4 +- .../utils.py | 2 +- .../model_selection_guide.ipynb | 4 +- ...nce_your_prompts_with_meta_prompting.ipynb | 0 .../{ => prompting}/Optimize_Prompts.ipynb | 0 .../Prompt_migration_guide.ipynb | 0 examples/prompting/README.md | 65 +++++++++++++++++++ .../Realtime_prompting_guide.ipynb | 0 ...st_writing_using_a_multi-step_prompt.ipynb | 0 ...ep_prompt_with_older_completions_API.ipynb | 0 .../Whisper_prompting_guide.ipynb | 0 .../prompts/code_interpreter.md | 0 .../prompts/editor_base.md | 0 .../prompts/fundamental_base.md | 0 .../prompts/macro_base.md | 0 .../prompts/pm_base.md | 0 .../prompts/quant_base.md | 0 .../prompts/tool_retry_prompt.md | 0 .../data/oai_docs/prompt-engineering.txt | 0 .../gpt-5/gpt-5_prompting_guide.ipynb | 0 .../gpt-5/prompt-optimization-cookbook.ipynb | 0 .../llm_as_judge.txt | 0 .../requirements.txt | 0 .../results_failsafeqa_baseline.csv | 0 .../results_failsafeqa_optimized.csv | 0 .../judgement_summary.csv | 0 .../results_llm_as_judge_baseline/run_01.json | 0 .../results_llm_as_judge_baseline/run_02.json | 0 .../results_llm_as_judge_baseline/run_03.json | 0 .../results_llm_as_judge_baseline/run_04.json | 0 .../results_llm_as_judge_baseline/run_05.json | 0 .../results_llm_as_judge_baseline/run_06.json | 0 .../results_llm_as_judge_baseline/run_07.json | 0 .../results_llm_as_judge_baseline/run_08.json | 0 .../results_llm_as_judge_baseline/run_09.json | 0 .../results_llm_as_judge_baseline/run_10.json | 0 .../results_llm_as_judge_baseline/run_11.json | 0 .../results_llm_as_judge_baseline/run_12.json | 0 .../results_llm_as_judge_baseline/run_13.json | 0 .../results_llm_as_judge_baseline/run_14.json | 0 .../results_llm_as_judge_baseline/run_15.json | 0 .../results_llm_as_judge_baseline/run_16.json | 0 .../results_llm_as_judge_baseline/run_17.json | 0 .../results_llm_as_judge_baseline/run_18.json | 0 .../results_llm_as_judge_baseline/run_19.json | 0 .../results_llm_as_judge_baseline/run_20.json | 0 .../results_llm_as_judge_baseline/run_21.json | 0 .../results_llm_as_judge_baseline/run_22.json | 0 .../results_llm_as_judge_baseline/run_23.json | 0 .../results_llm_as_judge_baseline/run_24.json | 0 .../results_llm_as_judge_baseline/run_25.json | 0 .../results_llm_as_judge_baseline/run_26.json | 0 .../results_llm_as_judge_baseline/run_27.json | 0 .../results_llm_as_judge_baseline/run_28.json | 0 .../results_llm_as_judge_baseline/run_29.json | 0 .../results_llm_as_judge_baseline/run_30.json | 0 .../judgement_summary.csv | 0 .../run_01.json | 0 .../run_02.json | 0 .../run_03.json | 0 .../run_04.json | 0 .../run_05.json | 0 .../run_06.json | 0 .../run_07.json | 0 .../run_08.json | 0 .../run_09.json | 0 .../run_10.json | 0 .../run_11.json | 0 .../run_12.json | 0 .../run_13.json | 0 .../run_14.json | 0 .../run_15.json | 0 .../run_16.json | 0 .../run_17.json | 0 .../run_18.json | 0 .../run_19.json | 0 .../run_20.json | 0 .../run_21.json | 0 .../run_22.json | 0 .../run_23.json | 0 .../run_24.json | 0 .../run_25.json | 0 .../run_26.json | 0 .../run_27.json | 0 .../run_28.json | 0 .../run_29.json | 0 .../run_30.json | 0 .../results_topk_baseline/run_01.py | 0 .../results_topk_baseline/run_02.py | 0 .../results_topk_baseline/run_03.py | 0 .../results_topk_baseline/run_04.py | 0 .../results_topk_baseline/run_05.py | 0 .../results_topk_baseline/run_06.py | 0 .../results_topk_baseline/run_07.py | 0 .../results_topk_baseline/run_08.py | 0 .../results_topk_baseline/run_09.py | 0 .../results_topk_baseline/run_10.py | 0 .../results_topk_baseline/run_11.py | 0 .../results_topk_baseline/run_12.py | 0 .../results_topk_baseline/run_13.py | 0 .../results_topk_baseline/run_14.py | 0 .../results_topk_baseline/run_15.py | 0 .../results_topk_baseline/run_16.py | 0 .../results_topk_baseline/run_17.py | 0 .../results_topk_baseline/run_18.py | 0 .../results_topk_baseline/run_19.py | 0 .../results_topk_baseline/run_20.py | 0 .../results_topk_baseline/run_21.py | 0 .../results_topk_baseline/run_22.py | 0 .../results_topk_baseline/run_23.py | 0 .../results_topk_baseline/run_24.py | 0 .../results_topk_baseline/run_25.py | 0 .../results_topk_baseline/run_26.py | 0 .../results_topk_baseline/run_27.py | 0 .../results_topk_baseline/run_28.py | 0 .../results_topk_baseline/run_29.py | 0 .../results_topk_baseline/run_30.py | 0 .../run_results_topk_baseline.csv | 0 .../run_results_topk_baseline_summary.json | 0 .../run_results_topk_baseline_summary.txt | 0 .../results_topk_optimized/run_01.py | 0 .../results_topk_optimized/run_02.py | 0 .../results_topk_optimized/run_03.py | 0 .../results_topk_optimized/run_04.py | 0 .../results_topk_optimized/run_05.py | 0 .../results_topk_optimized/run_06.py | 0 .../results_topk_optimized/run_07.py | 0 .../results_topk_optimized/run_08.py | 0 .../results_topk_optimized/run_09.py | 0 .../results_topk_optimized/run_10.py | 0 .../results_topk_optimized/run_11.py | 0 .../results_topk_optimized/run_12.py | 0 .../results_topk_optimized/run_13.py | 0 .../results_topk_optimized/run_14.py | 0 .../results_topk_optimized/run_15.py | 0 .../results_topk_optimized/run_16.py | 0 .../results_topk_optimized/run_17.py | 0 .../results_topk_optimized/run_18.py | 0 .../results_topk_optimized/run_19.py | 0 .../results_topk_optimized/run_20.py | 0 .../results_topk_optimized/run_21.py | 0 .../results_topk_optimized/run_22.py | 0 .../results_topk_optimized/run_23.py | 0 .../results_topk_optimized/run_24.py | 0 .../results_topk_optimized/run_25.py | 0 .../results_topk_optimized/run_26.py | 0 .../results_topk_optimized/run_27.py | 0 .../results_topk_optimized/run_28.py | 0 .../results_topk_optimized/run_29.py | 0 .../results_topk_optimized/run_30.py | 0 .../run_results_topk_optimized.csv | 0 .../run_results_topk_optimized_summary.json | 0 .../run_results_topk_optimized_summary.txt | 0 .../run_FailSafeQA.py | 0 .../scripts/__init__.py | 0 .../scripts/gen_baseline.py | 0 .../scripts/gen_optimized.py | 0 .../scripts/llm_judge.py | 0 .../scripts/results_summarizer.py | 0 .../scripts/topk_eval.py | 0 .../gpt4-1_prompting_guide.ipynb | 0 .../o-series/o3o4-mini_prompting_guide.ipynb | 0 .../src/utils/translation_prompts.js | 0 registry.yaml | 23 ++++--- 164 files changed, 81 insertions(+), 17 deletions(-) rename examples/{ => prompting}/Enhance_your_prompts_with_meta_prompting.ipynb (100%) rename examples/{ => prompting}/Optimize_Prompts.ipynb (100%) rename examples/{ => prompting}/Prompt_migration_guide.ipynb (100%) create mode 100644 examples/prompting/README.md rename examples/{ => prompting}/Realtime_prompting_guide.ipynb (100%) rename examples/{ => prompting}/Unit_test_writing_using_a_multi-step_prompt.ipynb (100%) rename examples/{ => prompting}/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb (100%) rename examples/{ => prompting}/Whisper_prompting_guide.ipynb (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md (100%) rename examples/{ => prompting}/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md (100%) rename examples/{ => prompting}/data/oai_docs/prompt-engineering.txt (100%) rename examples/{ => prompting}/gpt-5/gpt-5_prompting_guide.ipynb (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook.ipynb (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/requirements.txt (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/__init__.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py (100%) rename examples/{ => prompting}/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py (100%) rename examples/{ => prompting}/gpt4-1_prompting_guide.ipynb (100%) rename examples/{ => prompting}/o-series/o3o4-mini_prompting_guide.ipynb (100%) rename examples/{ => prompting}/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js (100%) diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/multi_agent_portfolio_collaboration.ipynb b/examples/agents_sdk/multi-agent-portfolio-collaboration/multi_agent_portfolio_collaboration.ipynb index aa505230ab..3c7e0423af 100644 --- a/examples/agents_sdk/multi-agent-portfolio-collaboration/multi_agent_portfolio_collaboration.ipynb +++ b/examples/agents_sdk/multi-agent-portfolio-collaboration/multi_agent_portfolio_collaboration.ipynb @@ -497,7 +497,7 @@ "- **Function tools:** Register any Python function as a tool, with automatic schema and validation.\n", "- **Tracing:** Visualize, debug, and monitor every step of your workflow for full transparency.\n", "\n", - "A combination of well-designed tools, thoughtful orchestration, and careful model selection is crucial for building effective agent systems. In this example, we use the GPT-4.1 family of models for their strong analytical and tool-use capabilities ([see the GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide)). For deeper architectural best practices, see the included [A Practical Guide to Building Agents (PDF)](https://cdn.openai.com/business-guides-and-resources/a-practical-guide-to-building-agents.pdf). By bringing these elements together, you get a system that is robust, scalable, and easy to debug or extend.\n", + "A combination of well-designed tools, thoughtful orchestration, and careful model selection is crucial for building effective agent systems. In this example, we use the GPT-4.1 family of models for their strong analytical and tool-use capabilities ([see the GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/prompting/gpt4-1_prompting_guide)). For deeper architectural best practices, see the included [A Practical Guide to Building Agents (PDF)](https://cdn.openai.com/business-guides-and-resources/a-practical-guide-to-building-agents.pdf). By bringing these elements together, you get a system that is robust, scalable, and easy to debug or extend.\n", "\n", "Please try out the sample with your own investment questions, and please share any feedback! Happy building.\n", "\n", @@ -512,7 +512,7 @@ "\n", "- [MCP Spec](https://spec.modelcontextprotocol.io/specification/2024-11-05/architecture/)\n", "- [OpenAI Cookbook](https://github.com/openai/openai-cookbook)\n", - "- ([GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide))\n", + "- ([GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/prompting/gpt4-1_prompting_guide))\n", "- [A Practical Guide to Building Agents (PDF)](https://cdn.openai.com/business-guides-and-resources/a-practical-guide-to-building-agents.pdf)\n", "\n", "---" diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py b/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py index 4d3af6c8b0..e7e7078791 100644 --- a/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py +++ b/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py @@ -39,7 +39,7 @@ def outputs_dir() -> Path: # Prompt loader # --------------------------------------------------------------------------- -PROMPTS_DIR: Path = repo_path("prompts") +PROMPTS_DIR: Path = Path(__file__).resolve().parent.parent.parent / "prompting" / "agents_sdk" / "multi-agent-portfolio-collaboration" / "prompts" def load_prompt(name: str, **subs) -> str: diff --git a/examples/partners/model_selection_guide/model_selection_guide.ipynb b/examples/partners/model_selection_guide/model_selection_guide.ipynb index 8b4b9e6604..19416b2e81 100644 --- a/examples/partners/model_selection_guide/model_selection_guide.ipynb +++ b/examples/partners/model_selection_guide/model_selection_guide.ipynb @@ -2425,7 +2425,7 @@ "\n", "- **[Orchestrating Agents: Routines and Handoffs](https://cookbook.openai.com/examples/orchestrating_agents)** Structuring multi-agent workflows with routines and handoffs, relevant to the ideation→ranking→critique pipeline.\n", "\n", - "- **[GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide)** Advanced prompting, tool use, and task decomposition for improved accuracy in critique and safety reviews.\n", + "- **[GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/prompting/gpt4-1_prompting_guide)** Advanced prompting, tool use, and task decomposition for improved accuracy in critique and safety reviews.\n", "\n", "- **[Structured Outputs for Multi-Agent Systems](https://cookbook.openai.com/examples/structured_outputs_multi_agent)** Enforcing consistent JSON outputs with schema validation for agent interoperability.\n", "\n", @@ -3219,7 +3219,7 @@ "- [Data Extraction and Transformation](https://cookbook.openai.com/examples/data_extraction_transformation)\n", "\n", "### Prompting & Model Selection\n", - "- [GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/gpt4-1_prompting_guide)\n", + "- [GPT-4.1 Prompting Guide](https://cookbook.openai.com/examples/prompting/gpt4-1_prompting_guide)\n", "- [Prompt Engineering Best Practices](https://platform.openai.com/docs/guides/prompt-engineering)\n", "\n", "### Evaluation & Deployment\n", diff --git a/examples/Enhance_your_prompts_with_meta_prompting.ipynb b/examples/prompting/Enhance_your_prompts_with_meta_prompting.ipynb similarity index 100% rename from examples/Enhance_your_prompts_with_meta_prompting.ipynb rename to examples/prompting/Enhance_your_prompts_with_meta_prompting.ipynb diff --git a/examples/Optimize_Prompts.ipynb b/examples/prompting/Optimize_Prompts.ipynb similarity index 100% rename from examples/Optimize_Prompts.ipynb rename to examples/prompting/Optimize_Prompts.ipynb diff --git a/examples/Prompt_migration_guide.ipynb b/examples/prompting/Prompt_migration_guide.ipynb similarity index 100% rename from examples/Prompt_migration_guide.ipynb rename to examples/prompting/Prompt_migration_guide.ipynb diff --git a/examples/prompting/README.md b/examples/prompting/README.md new file mode 100644 index 0000000000..a257acec26 --- /dev/null +++ b/examples/prompting/README.md @@ -0,0 +1,65 @@ +# Prompting Examples and Guides + +This directory contains a comprehensive collection of prompting-related articles, guides, and resources from the OpenAI Cookbook. All content focuses on various aspects of prompt engineering, optimization, and best practices for different OpenAI models. + +## Core Prompting Guides + +### Model-Specific Guides +- **`gpt4-1_prompting_guide.ipynb`** - Comprehensive prompting guide for GPT-4.1 including system prompts, tool usage, and advanced techniques +- **`gpt-5/gpt-5_prompting_guide.ipynb`** - GPT-5 specific prompting strategies and best practices +- **`o-series/o3o4-mini_prompting_guide.ipynb`** - Specialized guide for O3 and O4-mini reasoning models +- **`Realtime_prompting_guide.ipynb`** - Best practices for prompting with the Realtime API +- **`Whisper_prompting_guide.ipynb`** - Prompting techniques for Whisper speech recognition models + +## Prompt Optimization and Enhancement + +- **`Optimize_Prompts.ipynb`** - Multi-agent system for automatically identifying and fixing common prompt issues +- **`Enhance_your_prompts_with_meta_prompting.ipynb`** - Meta-prompting techniques for automated prompt optimization +- **`Prompt_migration_guide.ipynb`** - Guide for migrating and updating existing prompts +- **`gpt-5/prompt-optimization-cookbook.ipynb`** - Advanced prompt optimization techniques for GPT-5 +- **`gpt-5/prompt-optimization-cookbook/`** - Supporting scripts, data, and evaluation tools for prompt optimization + +## Multi-Step and Advanced Prompting + +- **`Unit_test_writing_using_a_multi-step_prompt.ipynb`** - Multi-step prompting approach for generating unit tests +- **`Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb`** - Legacy version using older completions API + +## Agent and Application Prompts + +- **`agents_sdk/multi-agent-portfolio-collaboration/prompts/`** - Collection of specialized prompts for multi-agent systems: + - `code_interpreter.md` - Code interpretation agent prompts + - `editor_base.md` - Document editing agent prompts + - `fundamental_base.md` - Fundamental analysis prompts + - `macro_base.md` - Macroeconomic analysis prompts + - `pm_base.md` - Portfolio manager prompts + - `quant_base.md` - Quantitative analysis prompts + - `tool_retry_prompt.md` - Tool retry logic prompts + +## Specialized Applications + +- **`voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js`** - Translation prompts for real-time voice applications + +## Supporting Resources + +- **`data/oai_docs/prompt-engineering.txt`** - Official OpenAI prompt engineering documentation + +## Usage + +Each notebook and guide is self-contained and includes: +- Detailed explanations of prompting concepts +- Practical examples and code samples +- Best practices and common pitfalls +- Model-specific considerations + +To get started, explore the model-specific guides that match your use case, then dive into optimization techniques to improve your prompts further. + +## Key Topics Covered + +- System prompt design and structure +- Few-shot and zero-shot prompting +- Chain-of-thought reasoning +- Tool calling and function usage +- Prompt optimization and evaluation +- Multi-agent prompt coordination +- Model-specific prompting strategies +- Advanced prompting techniques \ No newline at end of file diff --git a/examples/Realtime_prompting_guide.ipynb b/examples/prompting/Realtime_prompting_guide.ipynb similarity index 100% rename from examples/Realtime_prompting_guide.ipynb rename to examples/prompting/Realtime_prompting_guide.ipynb diff --git a/examples/Unit_test_writing_using_a_multi-step_prompt.ipynb b/examples/prompting/Unit_test_writing_using_a_multi-step_prompt.ipynb similarity index 100% rename from examples/Unit_test_writing_using_a_multi-step_prompt.ipynb rename to examples/prompting/Unit_test_writing_using_a_multi-step_prompt.ipynb diff --git a/examples/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb b/examples/prompting/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb similarity index 100% rename from examples/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb rename to examples/prompting/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb diff --git a/examples/Whisper_prompting_guide.ipynb b/examples/prompting/Whisper_prompting_guide.ipynb similarity index 100% rename from examples/Whisper_prompting_guide.ipynb rename to examples/prompting/Whisper_prompting_guide.ipynb diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/code_interpreter.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/editor_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/fundamental_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/macro_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/pm_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/quant_base.md diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md b/examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md similarity index 100% rename from examples/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md rename to examples/prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts/tool_retry_prompt.md diff --git a/examples/data/oai_docs/prompt-engineering.txt b/examples/prompting/data/oai_docs/prompt-engineering.txt similarity index 100% rename from examples/data/oai_docs/prompt-engineering.txt rename to examples/prompting/data/oai_docs/prompt-engineering.txt diff --git a/examples/gpt-5/gpt-5_prompting_guide.ipynb b/examples/prompting/gpt-5/gpt-5_prompting_guide.ipynb similarity index 100% rename from examples/gpt-5/gpt-5_prompting_guide.ipynb rename to examples/prompting/gpt-5/gpt-5_prompting_guide.ipynb diff --git a/examples/gpt-5/prompt-optimization-cookbook.ipynb b/examples/prompting/gpt-5/prompt-optimization-cookbook.ipynb similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook.ipynb rename to examples/prompting/gpt-5/prompt-optimization-cookbook.ipynb diff --git a/examples/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt b/examples/prompting/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt rename to examples/prompting/gpt-5/prompt-optimization-cookbook/llm_as_judge.txt diff --git a/examples/gpt-5/prompt-optimization-cookbook/requirements.txt b/examples/prompting/gpt-5/prompt-optimization-cookbook/requirements.txt similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/requirements.txt rename to examples/prompting/gpt-5/prompt-optimization-cookbook/requirements.txt diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_failsafeqa_baseline.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_failsafeqa_optimized.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/judgement_summary.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_01.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_02.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_03.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_04.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_05.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_06.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_07.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_08.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_09.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_10.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_11.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_12.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_13.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_14.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_15.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_16.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_17.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_18.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_19.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_20.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_21.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_22.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_23.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_24.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_25.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_26.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_27.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_28.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_29.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_baseline/run_30.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/judgement_summary.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_01.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_02.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_03.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_04.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_05.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_06.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_07.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_08.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_09.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_10.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_11.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_12.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_13.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_14.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_15.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_16.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_17.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_18.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_19.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_20.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_21.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_22.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_23.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_24.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_25.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_26.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_27.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_28.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_29.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_llm_as_judge_optimized/run_30.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_01.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_02.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_03.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_04.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_05.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_06.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_07.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_08.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_09.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_10.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_11.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_12.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_13.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_14.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_15.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_16.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_17.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_18.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_19.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_20.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_21.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_22.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_23.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_24.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_25.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_26.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_27.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_28.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_29.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_30.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_baseline/run_results_topk_baseline_summary.txt diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_01.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_02.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_03.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_04.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_05.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_06.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_07.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_08.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_09.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_10.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_11.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_12.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_13.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_14.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_15.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_16.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_17.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_18.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_19.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_20.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_21.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_22.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_23.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_24.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_25.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_26.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_27.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_28.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_29.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_30.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized.csv diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.json diff --git a/examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt b/examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt rename to examples/prompting/gpt-5/prompt-optimization-cookbook/results_topk_optimized/run_results_topk_optimized_summary.txt diff --git a/examples/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/run_FailSafeQA.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/__init__.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/__init__.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/__init__.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/__init__.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/gen_baseline.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/gen_optimized.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/llm_judge.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/results_summarizer.py diff --git a/examples/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py b/examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py similarity index 100% rename from examples/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py rename to examples/prompting/gpt-5/prompt-optimization-cookbook/scripts/topk_eval.py diff --git a/examples/gpt4-1_prompting_guide.ipynb b/examples/prompting/gpt4-1_prompting_guide.ipynb similarity index 100% rename from examples/gpt4-1_prompting_guide.ipynb rename to examples/prompting/gpt4-1_prompting_guide.ipynb diff --git a/examples/o-series/o3o4-mini_prompting_guide.ipynb b/examples/prompting/o-series/o3o4-mini_prompting_guide.ipynb similarity index 100% rename from examples/o-series/o3o4-mini_prompting_guide.ipynb rename to examples/prompting/o-series/o3o4-mini_prompting_guide.ipynb diff --git a/examples/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js b/examples/prompting/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js similarity index 100% rename from examples/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js rename to examples/prompting/voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js diff --git a/registry.yaml b/registry.yaml index 441f620999..80c1033819 100644 --- a/registry.yaml +++ b/registry.yaml @@ -33,7 +33,7 @@ - codex - title: Realtime Prompting Guide - path: examples/Realtime_prompting_guide.ipynb + path: examples/prompting/Realtime_prompting_guide.ipynb date: 2025-08-28 authors: - minh-hoque @@ -75,7 +75,7 @@ - gpt-oss-local - title: GPT-5 Prompt Migration and Improvement Using the New Optimizer - path: examples/gpt-5/prompt-optimization-cookbook.ipynb + path: examples/prompting/gpt-5/prompt-optimization-cookbook.ipynb date: 2025-08-07 authors: - rajpathak-openai @@ -87,7 +87,7 @@ - prompt-optimization - title: GPT-5 prompting guide - path: examples/gpt-5/gpt-5_prompting_guide.ipynb + path: examples/prompting/gpt-5/gpt-5_prompting_guide.ipynb date: 2025-08-07 authors: - anoop-openai @@ -240,7 +240,7 @@ - audio - title: Optimize Prompts - path: examples/Optimize_Prompts.ipynb + path: examples/prompting/Optimize_Prompts.ipynb date: 2025-07-14 authors: - corwin @@ -262,7 +262,7 @@ - automation - title: Prompt Migration Guide - path: examples/Prompt_migration_guide.ipynb + path: examples/prompting/Prompt_migration_guide.ipynb date: 2025-06-26 authors: - minh-hoque @@ -333,7 +333,7 @@ - mutli-agent-collaboration - title: o3/o4-mini Function Calling Guide - path: examples/o-series/o3o4-mini_prompting_guide.ipynb + path: examples/prompting/o-series/o3o4-mini_prompting_guide.ipynb date: 2025-05-26 authors: - billchen-openai @@ -787,7 +787,7 @@ - embeddings - title: Unit test writing using a multi-step prompt - path: examples/Unit_test_writing_using_a_multi-step_prompt.ipynb + path: examples/prompting/Unit_test_writing_using_a_multi-step_prompt.ipynb date: 2022-11-15 authors: - ted-at-openai @@ -795,8 +795,7 @@ - completions - title: Unit test writing using a multi-step prompt with legacy Completions - path: >- - examples/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb + path: examples/prompting/Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb date: 2023-05-19 authors: - ted-at-openai @@ -866,7 +865,7 @@ archived: true - title: Whisper prompting guide - path: examples/Whisper_prompting_guide.ipynb + path: examples/prompting/Whisper_prompting_guide.ipynb date: 2023-06-27 authors: - prestontuggle @@ -2185,7 +2184,7 @@ - audio - title: Enhance your prompts with meta prompting - path: examples/Enhance_your_prompts_with_meta_prompting.ipynb + path: examples/prompting/Enhance_your_prompts_with_meta_prompting.ipynb date: 2024-10-23 authors: - teomusatoiu @@ -2368,7 +2367,7 @@ - chatgpt-productivity - title: GPT-4.1 Prompting Guide - path: examples/gpt4-1_prompting_guide.ipynb + path: examples/prompting/gpt4-1_prompting_guide.ipynb date: 2025-04-14 authors: - nm-openai From 8c9c1cbf083b8ff26ca9f46b71dfc6e2f11fdd2a Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Wed, 17 Sep 2025 21:48:55 -0600 Subject: [PATCH 24/26] Update examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> --- .../agents_sdk/multi-agent-portfolio-collaboration/utils.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py b/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py index e7e7078791..85f491df75 100644 --- a/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py +++ b/examples/agents_sdk/multi-agent-portfolio-collaboration/utils.py @@ -39,7 +39,7 @@ def outputs_dir() -> Path: # Prompt loader # --------------------------------------------------------------------------- -PROMPTS_DIR: Path = Path(__file__).resolve().parent.parent.parent / "prompting" / "agents_sdk" / "multi-agent-portfolio-collaboration" / "prompts" +PROMPTS_DIR: Path = repo_path("prompting/agents_sdk/multi-agent-portfolio-collaboration/prompts") def load_prompt(name: str, **subs) -> str: From 1c1e541ab0291c2bc0e1f816eb9b18f76123dcb9 Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Wed, 17 Sep 2025 22:01:06 -0600 Subject: [PATCH 25/26] Revise README for prompting guides and examples Asari done it. --- examples/prompting/README.md | 124 +++++++++++++++++------------------ 1 file changed, 59 insertions(+), 65 deletions(-) diff --git a/examples/prompting/README.md b/examples/prompting/README.md index a257acec26..c530981fc1 100644 --- a/examples/prompting/README.md +++ b/examples/prompting/README.md @@ -1,65 +1,59 @@ -# Prompting Examples and Guides - -This directory contains a comprehensive collection of prompting-related articles, guides, and resources from the OpenAI Cookbook. All content focuses on various aspects of prompt engineering, optimization, and best practices for different OpenAI models. - -## Core Prompting Guides - -### Model-Specific Guides -- **`gpt4-1_prompting_guide.ipynb`** - Comprehensive prompting guide for GPT-4.1 including system prompts, tool usage, and advanced techniques -- **`gpt-5/gpt-5_prompting_guide.ipynb`** - GPT-5 specific prompting strategies and best practices -- **`o-series/o3o4-mini_prompting_guide.ipynb`** - Specialized guide for O3 and O4-mini reasoning models -- **`Realtime_prompting_guide.ipynb`** - Best practices for prompting with the Realtime API -- **`Whisper_prompting_guide.ipynb`** - Prompting techniques for Whisper speech recognition models - -## Prompt Optimization and Enhancement - -- **`Optimize_Prompts.ipynb`** - Multi-agent system for automatically identifying and fixing common prompt issues -- **`Enhance_your_prompts_with_meta_prompting.ipynb`** - Meta-prompting techniques for automated prompt optimization -- **`Prompt_migration_guide.ipynb`** - Guide for migrating and updating existing prompts -- **`gpt-5/prompt-optimization-cookbook.ipynb`** - Advanced prompt optimization techniques for GPT-5 -- **`gpt-5/prompt-optimization-cookbook/`** - Supporting scripts, data, and evaluation tools for prompt optimization - -## Multi-Step and Advanced Prompting - -- **`Unit_test_writing_using_a_multi-step_prompt.ipynb`** - Multi-step prompting approach for generating unit tests -- **`Unit_test_writing_using_a_multi-step_prompt_with_older_completions_API.ipynb`** - Legacy version using older completions API - -## Agent and Application Prompts - -- **`agents_sdk/multi-agent-portfolio-collaboration/prompts/`** - Collection of specialized prompts for multi-agent systems: - - `code_interpreter.md` - Code interpretation agent prompts - - `editor_base.md` - Document editing agent prompts - - `fundamental_base.md` - Fundamental analysis prompts - - `macro_base.md` - Macroeconomic analysis prompts - - `pm_base.md` - Portfolio manager prompts - - `quant_base.md` - Quantitative analysis prompts - - `tool_retry_prompt.md` - Tool retry logic prompts - -## Specialized Applications - -- **`voice_solutions/one_way_translation_using_realtime_api/src/utils/translation_prompts.js`** - Translation prompts for real-time voice applications - -## Supporting Resources - -- **`data/oai_docs/prompt-engineering.txt`** - Official OpenAI prompt engineering documentation - -## Usage - -Each notebook and guide is self-contained and includes: -- Detailed explanations of prompting concepts -- Practical examples and code samples -- Best practices and common pitfalls -- Model-specific considerations - -To get started, explore the model-specific guides that match your use case, then dive into optimization techniques to improve your prompts further. - -## Key Topics Covered - -- System prompt design and structure -- Few-shot and zero-shot prompting -- Chain-of-thought reasoning -- Tool calling and function usage -- Prompt optimization and evaluation -- Multi-agent prompt coordination -- Model-specific prompting strategies -- Advanced prompting techniques \ No newline at end of file +# Prompting: Guides & Examples + +This directory consolidates prompting-related guides, examples, and reusable prompt assets from across the Cookbook. It’s a single place to learn core prompting patterns, optimize prompts, and discover application-specific prompt sets. + +## Why this exists +Effective prompting solves a large share of practical model issues. The right prompt is as important as parameters like `temperature` or `reasoning_effort`. Centralizing examples and references makes them easier to find, reuse, and maintain. + +## Start here (recommended path) +1. **GPT-4.1 Prompting Guide** → techniques, structure, and patterns: + [`gpt4-1_prompting_guide.ipynb`](./gpt4-1_prompting_guide.ipynb) +2. **Prompt engineering best practices** (reference): + +3. **Orchestrating agents & handoffs** (for multi-agent apps): + `../orchestrating_agents` (see the top-level Examples index) +4. **Structured outputs** (JSON schemas, validation): + `../structured_outputs_multi_agent` + +> Tip: Keep prompts short, specific, and testable. Add minimal examples, define outputs precisely, and prefer explicit instructions over implications. + +## Contents + +### Core Guides +- **GPT-4.1 Prompting Guide** — system prompts, tool use, decomposition, evaluation + [`gpt4-1_prompting_guide.ipynb`](./gpt4-1_prompting_guide.ipynb) +- **Realtime prompting guide** — working with the Realtime API + [`Realtime_prompting_guide.ipynb`](./Realtime_prompting_guide.ipynb) +- **Whisper prompting guide** — task hints and formatting for speech recognition + [`Whisper_prompting_guide.ipynb`](./Whisper_prompting_guide.ipynb) + +### Prompt Optimization +- **Optimize Prompts** — automated checks & fixes for common prompt issues + [`Optimize_Prompts.ipynb`](./Optimize_Prompts.ipynb) +- **Enhance your prompts with meta-prompting** — programmatic refinement strategies + [`Enhance_your_prompts_with_meta_prompting.ipynb`](./Enhance_your_prompts_with_meta_prompting.ipynb) +- **Prompt migration guide** — safely updating existing prompts across changes + [`Prompt_migration_guide.ipynb`](./Prompt_migration_guide.ipynb) + +### Agent & App Prompts +- **Multi-agent portfolio collaboration prompts** — reusable prompt set for agent roles + [`../agents_sdk/multi-agent-portfolio-collaboration/prompts/`](../agents_sdk/multi-agent-portfolio-collaboration/prompts/) + +### Supporting Resources +- OpenAI Prompt Engineering (plain-text reference) + [`../data/oai_docs/prompt-engineering.txt`](../data/oai_docs/prompt-engineering.txt) + +## Usage pattern + +1. **Draft** a minimal instruction with explicit output shape (e.g., JSON schema). +2. **Ground** with constraints (tone, audience, knowledge limits) and a tiny example if needed. +3. **Test** with real inputs; watch for ambiguity and output drift. +4. **Evaluate** with checks (format validation, assertions). +5. **Iterate**: shorten, remove redundant rules, and pin “must-haves”. + +## Contributing + +- Keep notebooks **runnable end-to-end** (no hidden cell state). +- Prefer **relative links** within `examples/`, so both GitHub and the site render cleanly. +- When adding new files here, **update** `registry.yaml` so content appears on the site. +- If you introduce a new subfolder of prompts, include a short `README.md` explaining scope and usage. From d7f570b7780f14ff5bae62b9f565b0f5ae4d295f Mon Sep 17 00:00:00 2001 From: Payton Ison Date: Fri, 26 Sep 2025 01:53:11 -0600 Subject: [PATCH 26/26] Revert "Merge branch 'paytonison/introduction-rewrite' of https://github.com/paytonison/openai-cookbook into transformers-edit" This reverts commit 6febee10226ca89270154ab7cd2c26432bd43afd, reversing changes made to 1c1e541ab0291c2bc0e1f816eb9b18f76123dcb9. --- AGENTS.md | 23 - authors.yaml | 12 +- examples/codex/Autofix-github-actions.ipynb | 223 --- examples/codex/images/ci-codex-workflow.png | Bin 186532 -> 0 bytes examples/codex/images/codex-pr.png | Bin 534175 -> 0 bytes examples/codex/images/codex-workflow.png | Bin 531259 -> 0 bytes examples/codex/images/failing-workflow.png | Bin 252754 -> 0 bytes examples/codex/images/github-pr-settings.png | Bin 203329 -> 0 bytes ...reinforcement_finetuning_healthbench.ipynb | 1279 ++++++++--------- examples/gpt-5-codex_prompting_guide.ipynb | 182 --- registry.yaml | 18 - 11 files changed, 588 insertions(+), 1149 deletions(-) delete mode 100644 AGENTS.md delete mode 100644 examples/codex/Autofix-github-actions.ipynb delete mode 100644 examples/codex/images/ci-codex-workflow.png delete mode 100644 examples/codex/images/codex-pr.png delete mode 100644 examples/codex/images/codex-workflow.png delete mode 100644 examples/codex/images/failing-workflow.png delete mode 100644 examples/codex/images/github-pr-settings.png delete mode 100644 examples/gpt-5-codex_prompting_guide.ipynb diff --git a/AGENTS.md b/AGENTS.md deleted file mode 100644 index 86634ebf93..0000000000 --- a/AGENTS.md +++ /dev/null @@ -1,23 +0,0 @@ -# Repository Guidelines - -## Project Structure & Module Organization -The cookbook is organized around runnable examples and reference articles for OpenAI APIs. Place notebooks and Python scripts under `examples//`, grouping related assets inside topic subfolders (for example, `examples/agents_sdk/`). Narrative guides and long-form docs live in `articles/`, and shared diagrams or screenshots belong in `images/`. Update `registry.yaml` whenever you add content so it appears on cookbook.openai.com, and add new author metadata in `authors.yaml` if you want custom attribution. Keep large datasets outside the repo; instead, document how to fetch them in the notebook. - -## Build, Test, and Development Commands -Use a virtual environment to isolate dependencies: -- `python -m venv .venv && source .venv/bin/activate` -- `pip install -r examples//requirements.txt` (each sample lists only what it needs) -- `jupyter lab` or `jupyter notebook` to develop interactively -- `python .github/scripts/check_notebooks.py` to validate notebook structure before pushing - -## Coding Style & Naming Conventions -Write Python to PEP 8 with four-space indentation, descriptive variable names, and concise docstrings that explain API usage choices. Name new notebooks with lowercase, dash-or-underscore-separated phrases that match their directory—for example `examples/gpt-5/prompt-optimization-cookbook.ipynb`. Keep markdown cells focused and prefer numbered steps for multi-part workflows. Store secrets in environment variables such as `OPENAI_API_KEY`; never hard-code keys inside notebooks. - -## Testing Guidelines -Execute notebooks top-to-bottom after installing dependencies and clear lingering execution counts before committing. For Python modules or utilities, include self-check cells or lightweight `pytest` snippets and show how to run them (for example, `pytest examples/object_oriented_agentic_approach/tests`). When contributions depend on external services, mock responses or gate the cells behind clearly labeled opt-in flags. - -## Commit & Pull Request Guidelines -Use concise, imperative commit messages that describe the change scope (e.g., "Add agent portfolio collaboration demo"). Every PR should provide a summary, motivation, and self-review, and must tick the registry and authors checklist from `.github/pull_request_template.md`. Link issues when applicable and attach screenshots or output snippets for UI-heavy content. Confirm CI notebook validation passes locally before requesting review. - -## Metadata & Publication Workflow -New or relocated content must have an entry in `registry.yaml` with an accurate path, date, and tag set so the static site generator includes it. When collaborating, coordinate author slugs in `authors.yaml` to avoid duplicates, and run `python -m yaml lint registry.yaml` (or your preferred YAML linter) to catch syntax errors before submitting. diff --git a/authors.yaml b/authors.yaml index 80c9aebe8e..0faf6d884b 100644 --- a/authors.yaml +++ b/authors.yaml @@ -3,11 +3,6 @@ # You can optionally customize how your information shows up cookbook.openai.com over here. # If your information is not present here, it will be pulled from your GitHub profile. -daveleo-openai: - name: "Dave Leo" - website: "https://www.linkedin.com/in/davidanthonyleo/" - avatar: "https://media.licdn.com/dms/image/v2/C5603AQF2Kg-D7XJKNw/profile-displayphoto-shrink_800_800/profile-displayphoto-shrink_800_800/0/1612654752234?e=1761782400&v=beta&t=RkO9jCbJrY6Ox9YRbMA6HAAZhxfYJV1OsZeIT3YatBM" - jonlim-openai: name: "Jonathan Lim" website: "https://www.linkedin.com/in/jonlmr" @@ -493,12 +488,7 @@ heejingithub: website: "https://www.linkedin.com/in/heejc/" avatar: "https://avatars.githubusercontent.com/u/169293861" -himadri: - name: "Himadri Acharya" - website: "https://www.linkedin.com/in/himadri-acharya-086ba261/" - avatar: "https://avatars.githubusercontent.com/u/14100684?v=4" - - paytonison: +paytonison: name: "Payton Ison" website: "https://linkedin.com/in/paytonison" avatar: "https://avatars.githubusercontent.com/u/148833579" diff --git a/examples/codex/Autofix-github-actions.ipynb b/examples/codex/Autofix-github-actions.ipynb deleted file mode 100644 index b492626bb1..0000000000 --- a/examples/codex/Autofix-github-actions.ipynb +++ /dev/null @@ -1,223 +0,0 @@ -{ - "cells": [ - { - "cell_type": "markdown", - "id": "e2884696", - "metadata": {}, - "source": [ - "# Autofix CI failures on GitHub with Codex-cli\n", - "\n", - "## Purpose of this cookbook\n", - "\n", - "This cookbook shows you how to embed the OpenAI Codex CLI into your CI/CD pipeline so that when your builds or tests fail, codex automatically generates & proposes fixes. The following is an example in a node project with CI running in GitHub Actions. \n", - "\n", - "## End to End Flow\n", - "\n", - "Below is the pipeline flow we’ll implement:\n", - "\n", - "![](images/ci-codex-workflow.png)" - ] - }, - { - "cell_type": "markdown", - "id": "f83ce964", - "metadata": {}, - "source": [ - "## Prerequisites\n", - "\n", - "- A GitHub Repo with Actions workflows\n", - "\n", - "- You’ll need to create `OPENAI_API_KEY` as an environment variable in GitHub settings under https://github.com/{org-name}/{repo-name}/settings/secrets/actions. You can also set this at org level(for sharing secrets across multiple repos) \n", - "\n", - "- Codex requires python as a prerequisite to use `codex login`\n", - "\n", - "- You’ll need to check the setting to enable actions to create PRs on your repo, and also in your organization:\n", - "\n", - "![](images/github-pr-settings.png)" - ] - }, - { - "cell_type": "markdown", - "id": "99f5bed1", - "metadata": {}, - "source": [ - "\n", - "## Step 3: Insert Codex in your CI pipeline\n", - "\n", - "The following YAML shows a GitHub action that auto triggers when CI fails, installs Codex, uses codex exec and then makes a PR on the failing branch with the fix. Replace \"CI\" with the name of the workflow you want to monitor. " - ] - }, - { - "cell_type": "markdown", - "id": "a9f9b368", - "metadata": {}, - "source": [ - "```yaml\n", - "\n", - "name: Codex Auto-Fix on Failure\n", - "\n", - "on:\n", - " workflow_run:\n", - " # Trigger this job after any run of the primary CI workflow completes\n", - " workflows: [\"CI\"]\n", - " types: [completed]\n", - "\n", - "permissions:\n", - " contents: write\n", - " pull-requests: write\n", - "\n", - "jobs:\n", - " auto-fix:\n", - " # Only run when the referenced workflow concluded with a failure\n", - " if: ${{ github.event.workflow_run.conclusion == 'failure' }}\n", - " runs-on: ubuntu-latest\n", - " env:\n", - " OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}\n", - " FAILED_WORKFLOW_NAME: ${{ github.event.workflow_run.name }}\n", - " FAILED_RUN_URL: ${{ github.event.workflow_run.html_url }}\n", - " FAILED_HEAD_BRANCH: ${{ github.event.workflow_run.head_branch }}\n", - " FAILED_HEAD_SHA: ${{ github.event.workflow_run.head_sha }}\n", - " steps:\n", - " - name: Check prerequisites\n", - " run: |\n", - " if [ -z \"$OPENAI_API_KEY\" ]; then\n", - " echo \"OPENAI_API_KEY secret is not set. Skipping auto-fix.\" >&2\n", - " exit 1\n", - " fi\n", - "\n", - " - name: Checkout failing ref\n", - " uses: actions/checkout@v4\n", - " with:\n", - " ref: ${{ env.FAILED_HEAD_SHA }}\n", - " fetch-depth: 0\n", - "\n", - " - name: Setup Node.js\n", - " uses: actions/setup-node@v4\n", - " with:\n", - " node-version: '20'\n", - " cache: 'npm'\n", - "\n", - " - name: Install dependencies\n", - " run: |\n", - " if [ -f package-lock.json ]; then npm ci; else npm i; fi\n", - "\n", - " - name: Prepare Codex prerequisites\n", - " shell: bash\n", - " run: |\n", - " # Ensure python3 exists for Codex' login helper\n", - " if ! command -v python3 >/dev/null 2>&1; then\n", - " sudo apt-get update\n", - " sudo apt-get install -y python3\n", - " fi\n", - "\n", - " # Ensure Codex config dir exists and is writable\n", - " mkdir -p \"$HOME/.codex\"\n", - " # (Optional) pin an explicit home for Codex config/logs\n", - " echo \"CODEX_HOME=$HOME/.codex\" >> $GITHUB_ENV\n", - "\n", - " - name: Install Codex CLI\n", - " run: npm i -g @openai/codex\n", - "\n", - " - name: Authenticate Codex (non-interactive)\n", - " env:\n", - " # if you set CODEX_HOME above, export it here too\n", - " CODEX_HOME: ${{ env.CODEX_HOME }}\n", - " OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}\n", - " run: codex login --api-key \"$OPENAI_API_KEY\"\n", - "\n", - " - name: Run Codex to fix CI failure\n", - " run: |\n", - " codex exec --full-auto --sandbox workspace-write \"You are working in a Node.js monorepo with Jest tests and GitHub Actions. Read the repository, run the test suite, identify the minimal change needed to make all tests pass, implement only that change, and stop. Do not refactor unrelated code or files. Keep changes small and surgical.\"\n", - "\n", - " - name: Verify tests\n", - " run: npm test --silent\n", - "\n", - " - name: Create pull request with fixes\n", - " if: success()\n", - " uses: peter-evans/create-pull-request@v6\n", - " with:\n", - " commit-message: \"fix(ci): auto-fix failing tests via Codex\"\n", - " branch: codex/auto-fix-${{ github.event.workflow_run.run_id }}\n", - " base: ${{ env.FAILED_HEAD_BRANCH }}\n", - " title: \"Auto-fix failing CI via Codex\"\n", - " body: |\n", - " Codex automatically generated this PR in response to a CI failure on workflow `${{ env.FAILED_WORKFLOW_NAME }}`.\n", - "\n", - " Failed run: ${{ env.FAILED_RUN_URL }}\n", - " Head branch: `${{ env.FAILED_HEAD_BRANCH }}`\n", - "\n", - " This PR contains minimal changes intended solely to make the CI pass.\n", - "```\n" - ] - }, - { - "cell_type": "markdown", - "id": "8148024b", - "metadata": {}, - "source": [ - "## Step 4: Actions Workflow kicked off\n", - "\n", - "You can navigate to the Actions tab under Repo to view the failing jobs in your Actions workflow. \n", - "\n", - "\n", - "![](images/failing-workflow.png)\n" - ] - }, - { - "cell_type": "markdown", - "id": "64671aae", - "metadata": {}, - "source": [ - "The Codex workflow should be triggered upon completion of the failed workflow. \n", - "\n", - "\n", - "![](images/codex-workflow.png)\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "id": "d08a3ecc", - "metadata": {}, - "source": [ - "## Step 5: Codex generated PR for review\n", - "And after the Codex workflow completes execution, it should open a pull request from the feature branch codex/auto-fix. Check to see if everything looks good and then merge it.\n", - "\n", - "![](images/codex-pr.png)" - ] - }, - { - "cell_type": "markdown", - "id": "f4c1f3a0", - "metadata": {}, - "source": [ - "## Conclusion\n", - "\n", - "This automation seamlessly integrates OpenAI Codex CLI with GitHub Actions to automatically propose fixes for failing CI runs.\n", - "\n", - "By leveraging Codex, you can reduce manual intervention, accelerate code reviews, and keep your main branch healthy. The workflow ensures that test failures are addressed quickly and efficiently, letting developers focus on higher-value tasks. Explore more about codex-cli and its capabilities [here](https://github.com/openai/codex/)." - ] - } - ], - "metadata": { - "kernelspec": { - "display_name": "Python 3", - "language": "python", - "name": "python3" - }, - "language_info": { - "codemirror_mode": { - "name": "ipython", - "version": 3 - }, - "file_extension": ".py", - "mimetype": "text/x-python", - "name": "python", - "nbconvert_exporter": "python", - "pygments_lexer": "ipython3", - "version": "3.13.7" - } - }, - "nbformat": 4, - "nbformat_minor": 5 -} diff --git a/examples/codex/images/ci-codex-workflow.png b/examples/codex/images/ci-codex-workflow.png deleted file mode 100644 index 4be3dc4fb48ebf1a0b8ab476eaaa94751d0891ad..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 186532 zcmdpd1zQ};wl)@mO9<`~2u>ildvFNu4k5^(!6Ag;!7bSSG{{ z5F;aF#L$Z7{jAAbYluIqoBjGmT9 zVF;i&MLx6s=~(zIzwo$QJndSCtt%Ris_^WlLJ{uuE!8L!j)erdz!w5OxzA>(U4A%s z8=J#85=|&(r0)>H-bI74sigYovl7ImJ4(NUhKrG;n*U6Tr*qEYJSF5`J5V+$$L_U3fz$nPu~^4 zQSa15Euz(jjS@J6vxXt(U~wIzq!!Hg#s={$L@V^n^8{y4+<-RjX%FTsS@LiwdFGW~ z@Rvd2HQ=dbpvcP$QD*07d`!NIf(l)vmaVZ@l-aLiBWoQ zO2F$pBa;`Z*<%J@u1JS^sl8*qNnDXNLw(mEbmO=`bHeh9-5JGrU^EX~v!S=1X5ejh%9Z!f7-@dcF$nRI=$!$I*El&unX%H8HFU_& z5JRcEsIpldKHjYudGIyew(owabc$Z(Ynj2Lz!N~lM$tw!{G2w?EnT5zB=eE&2j%Mo z;Cbm;puV`0_5$Tq@UAMc^%`;;ynDczqoXljC9_4nq#tu-2TATjElP`*gye_k3~2Ji zk|#McUnmMp^H&A3vxhD~H)=PjB5{DjH>2A(^~RRQDD!1wNi(%G;`)8_4Y7Q8Od%@0>NV&3(4n&&0*({?mQe>GsF&AmQ+cwu8(e#tvBu z9~U#QJHQwNH-Z6|FUp)>kiQWj>;CbF5unrzxs@5E@&UbegBPwT8WRKK{yYd>7`e^Q z{aT*yKB&^PQze1~&Kev2c?AM~2s1m<^;TBrVm|hl0RFOo6Fx-Vl_&RNR47l>s3_zUn?d2!4cM68eWztEIF0^2ThA;}~v%RZ9t&t`)sY3^b> z`%pK?#={n*4rzt^fPLQ2C8yNLCsC>;dUNa&0)xMuY-_y6%LpUL?%qbLLMWDT&xzbd zJs0xAKac?CJP@@_a?o3a^PAXxiI6sB)Ip~2BrwU<;n#}fHeo&HH|}XDGukIt?jq(x zcmK%DhosW>h{bPxfNV`JWkX~loy|!JCr-9 zJ8bHb8&FHbq-f&bDY=Gtze@S4l&)Z)yr8(CAf&ucCqj1_|Ax*tPPjkqx%686soab2 zk>3};n}1jM?l7)jK{ES!UlEHYKjy^-jWd6Jluv?B*!}1GfNZ()?;dZnUrG#+4ag5* z$)j%~Uy? z0^7p*0+%WEDg0gLN%sOu?bd?FaG4z4vHN$7RX%fQdaMU@2NVaz$da9Ef%(*vIzQBM z$9_1;tv|_-sQvD;J+U2h?tCkL`^2l1UV1`4zOWx;kfodJsdLs~)k+{~`$;Q7D|6bp zSfJQNPo-LJPHQe;PU67q;9!oSnv#W9t`oqn|DIVQ9PV=_7MGR-}T=A!F%>ux==k=}|#^dj_T@ox0JikaoEx@1~x zg`eEQr>x(G>A;oMqw-%lQ|lS|Y1S=Z z_)*a&vL@%I-KI?CicGrHD)Vtk02iR;isy)pGD#(M^h(6AC9lP2`Tg=!zb!wB)(k%t zzM6Eq5r+|Ue+K^)2sMP)K;_Wq@C1&9iZAPR7Okov2Xz_eu~Ce!s+Fp*s&Sd^@xsbF z_2-zfm?>(Ir;HGMX%dKVmqI}FT;r>kj)9KlOrT?fmRqk8IZdqZt{L^-^_G5V%`l12dh<@kjhpyU22#3+_%!u5jA4v|>bLnw%DSStf>YepbQg0Pdj*yGnQ2xW ztnPEMJLfz1J5MK2Y~*a*Y$$Kk?%(_Q3$XjDdjUT(&ro@=o#U0@U4}ME+e)c$HgQc` z2o0Z)7^Ka{Z>DKxPBCwVGk3FBN>o&tLA6z*MO>-y7%k714RRcnva^^yu=QAH9Nyrt6K{C_cgEWxaah0 zE*`GdjsSM(OVc{`nuaYIb*QgABlm5O-#+BGk5ps;KgfM>o9dcQ|LIm*T+CIZ zuNythyzBiT|FETw!9>L+P_nb8d8pa)u1%!7#rkgI(pV7ZL)HP+ok?eOlBk0~7oWyl z=<5-iecOTeyYH194u#0P>WaGyH-D)_i5|S?wl5aQ_O{qm+Z|iWLF`3~zMxE|I1-Na zTr-_NubF3H05&-@ZE5uGlucOsW8L0t5RasoTQ`r^WS$ARUlU%kD}fhBcxxJKC~v1O zK$7GRt@FJ3(mG*EGsauJvDf|K|30_DU8(?EPY|n=p=mM1;Sj7%;$9h zv@~ft?sXnWyQ8?vAH82zWh+{MqW?p*!m#85IB$h1RE$CCwe<$c>ndX z?mnzPelCARMc@0A7n?}hY0+K%`Ax`D+;X+w@_ENrceMZY!`KpUed%59{$?9+(0HeH z?jm{6*#2Sc;mQcy{CJ#qUi>(p>b~!8y~NP$e`-1{JKh_vQUlZz3Gr>=is9v;+8n$N zY(*earLgCOyLKUjtFnM2I$n5_ku5-pMIhytM=rG@YC1*iTviAF3l6el?zOr+>;`vv z0M~}LJE0CAng@?_1pg}xPj>fM*Wq}ffJMrK=i2W!|Gdfd7!bazc$>JKqo1YW2A?zU z4_{jn_`)-eO&ywGgj?#!St%*OF+tB!;E>=+;GRIw;Gtg;c+$VmW#C`JA^yIO00$Rh z1BdkIGs@5->=Oh1!rJ_CM2riDdkX#a4EpuRLip=xq_Ql;zt6j%_u$0UCFSIxM|E>& zOG^ircK}y0H%AEc0;=O1T^Bewf)}tayxd#5V`%>~HX1svI!cNH<^X#(QwxBZC7Xx6 zBdi}dArArQslBDEDV2x4or8;jhcM0WCj_A9u*>W;RKFi_wH2n(QBtLn1UOq#@vw2Q zanOjMQBhF|Ia^o>yp@vqvpMvaFwHwxS4RPMc6WDoHg_&IfU`CGD}H`{b`DNbt~LM%DpJC2-=y`mOOMJ$!w%lG@wVQn!!E@X3#$`|2RU=;k7>CQ|5%j!HL1iNr`KC!0#`j z-nY!^h#u>!>3uIU|7yk38G}2Z^VVD;H``E3Pm^i9Wg<`Wj_sSYH=}$oJudAa9cFMl z2DOwNZW!3Vd34X;H{MJ^!M zvzAvyjV&yhJ|)XnJBOu|pmk=$9eKgqD$OhQJHYO3y6uDTLpjnF*mL2C9Gr9ozM1rs zrJ5Z6mq5#KQJx1>8kn2{tk9J z{~;4};Iemg<^%WSqe9LqTe0S#v(!CL zjKQJarliq}s|ygBHBPC?GG%GQr8woexc=ydZ;4!pFD2-y|M637?UvKTr+1dUnf;K1 z3^13c9kACk`s}e@=&MjQRCT5jMyMADL$;m$x5?v}tc-4=Q zGRnY(d2#rc(XG6sRz;dt8z`2y{X-YqV#VMBPBO2WtUaz8mlyk}K1aLg$BD|NZ%_op z_0}KBCEhO^-QO5f%tzPKY35yibZz$cBeTu~iyRHRc!M}IeN^&tL`S`QuCBoA>yHbK zU9m^!z+#mM^O^e6jz^l0d?a{sa`Z38Hpm#`@0uKryfRhD!PN`UsfbixW%`{d`e*if z3SX57)}XG9G_Nn*1CNI9T!2p22A-Kj{G;Y~CrOaT%#FDazwHnnnvoV!nGVf=DEPZd z$$nkYI1GdF+`!b{2NBd-x~X)A@}%CSj1Aqx?hKoX z(en+rQLVqn%mWXzV`@Ss%-?F4PJ=?%9^a*=LX^v`6em-Er5Gn`BER;M+9k8SoDWx| zFWYa>Rt{m)e`NV)#GS|NWi+l7Y{2~YP}RFT4(4s5&MEc%J$qRMc#Anq2gkrwbCE4` zlJ?70%E#4CS)VRhjLXML*_zq(`4~%22Ou~qR@a;c-IMcZ$_k+ZaPk5&^PY= z2=m{5vJeOA`tSuVjDL5XPlX_d!#W>GCzo>~vO2Km=`h-`ZFuCRT}dWEwAmB)k{1*s zd*sn8SaTWry|($lzLRwMx@VOY;Hf}qNewKQ~#k3n_q18fq?d?Pv$my{baPgZIM{k%@W~v=eqEv!`S>c}7u^ z#_m16o(`fSBewOFz53ukOR#A?v{LO%ix4zM!Ik+gr z`G`oj`yvuOomKFA^Je0n%P1GcRRptOV-&L;8^VrOcLVKWtn*v@X#!7HRu1Q*)<=)o z`lu2}lKo9<-6`KuhQR1uMP*usDGJ!>@NDi{sB|9van#!x2%5({iKkOKG*sbah&~9) zScrGpFlz2+-5aYoYgw`~bOVSaLmH8#puWwm>a6swF)W(3 zX~|;mBz%ANtlz1je+OxoXs)6&HDhdlVfX8@M$6_}J zW`#@209%7~l3{!l>ICxI7B0|eVorSCy7iHlLti#sblznqvGLWHs11$n{;wwx!5e_j z(a*ETR<~Yt2)NqKdpZ883{c3w+Usz;`+wzqV2l#jNJX#vwP1Y1(i#7?l&L;(!yi0y$^mq`$Dx&`GXft zR-#Myr7nq}{3yf0y$6xfQkfr*tVmH<2pgL^Nn4(J2`|0@!Iz%tr#z(>clV{sf+NGt z5}EVRFtx;8(F&l9;~1==c}jHai^V~gl7xu9(b^t_c7{WB1aQqE#RRonR@W|2;aZt% zzX8u7b+vLN^)t#&ge#K9q+L?aqZ}ad8-Du(dznx`fx>WS!6DP1v(D5Rz#6HOoN(7V z<)d{hRYvaUzr3t%ZMKgZzB;FboKV`I9@n;96nCaA-*2tPDX`gOr1(5)%%=J9_0#4o z^D~v;iQSBjv{f`mNj`%4`m_9&@r2~o>OkezE=Q^ER_U|wPAU{C2f4are?Y()Muay- zbj5MMNRB4rSe}YHmu5KIE!XM_JLR*Ei)jvm$+o?VX(Y&Q7&7cZL=HEoqn&t`7Y)(Rlv~!Ym!KH6dUw0?G{OGOWMjXL#l4;Z z*HR(%Z1vc~s|TkgAm~)6pBhP3k2SYYrUCH%iN-VC<|i5=6UGV=Kbq(d(f)KNd$?jI z)FW|UDrDbmR##o_LavPU{OdChH!n-TTylLE-&1S2Sy#M1*&lL`GN6KL!F#Q)Tt_Ak zwvs*<8`^rGX2D5GrpP{uG0S+S>!k{2ETv%2!!s|~6rh;a0)}Zf77D7Mn6@o|TC6~M zkKwa=oT|Z5-NvTR+XQ=krg@pvGrhw|Hn3CauYB|nNR)Zq7I!Isx>FBvV6)nWMhW8> zOTKe;fnN|~uV<8p{9C&(Dom5=caJGThevy*7rpE5lb(Nu($@%~S#D_zciyK$SHdZi zlwio^oCZ?z2j2N1*9Z`npmOc?0(b=nfVI?g?vlRR`+zYYF~#E?Y|>#)gW@%Eh}sJtuLTpGi7Oaz|$e@hxHD^oHt#HuY-97 z%hU3u%bHruQooDBG;X&Zm&vOI;^Ccc`N7~i&b9=LBy%L4WyVw9;Fjvinh0{7zi&Lp04L0{hKZl;m0nO0R%Skz4Bn(lU0>3}ptd@__kRHS3s?Ladom(? zwXk{%YJGHE+%u}x)dz>y)#s=E;@T#Y`#45!SKv=aJdI}-^+&ibBZVmz=KBjp=*kyK zeax2aCuR2I7kCd0=y)?GDU`yFc4Luqy?F`k(9h%X3#n;KBGal_2o~Qpy{%8S#(r_n<7Ky%b+MZt~Iop!%=!UliL(3;4)2}x4 zp1tSC7uSmyREc;s3h**Haa$){uET6g>B_ty>=<-Kga!ASP!0iQtOV*-B(U7)S=#ou zqI2g*CFd|hSfzq_@D&39Ba$*B(n}{4MlDqyspoIe#1p0P$HOIL#1(R*%#wQRM{P&q zpXc1$>xpq>cy6WWTix=PBDwhX+K5l7H0_c+Ln}`l(1#eRgYJPKmu2HUVCni_uo-H6 z{z%SLY*$OaD8Z!pB{$I`f!EpS7iW%J3CQ(SU(V9cmb z{RHKJx}~iDO0VNT!#ShJN(1Bzi~NrK!I$f-@%x^E0j@<=CMvcc5JTx3WIdX%yV*x% zc|NFI?GgsPYO|JfB&PfBGM0~HQdFz+y;7w1f{uuM7y|R#D;}_Lc`^jL{V8F>itvVh zLp`;Q%U$K{f-d{5le3DlgR;1pbWrHpTy4KRZp06mm$r{jCT*_VjGMGUmF^d(N9Fh0 zZlYw|H-uXR6vnnyOBXr(ZK!Agl42jZ1%;y=lG!56GaT%?p-gJG&;#33V)FcIq0_w= z?4JGKEoZ5H#^y{ufhJ0NO%HX`Rw4DkoOkWYi~hGhevhPl?y zv67){{O!;Rz2Sr}&**Iarf-AvIAiR)v}YrNM@>Au+RF0L=K=>+6D06fs$9ijZ}i(q z##qzyfu&PF^_IeMHI(%Cuu=cq^zc23mC9lOp3a2ujflyK$v+IX-Yd}N;Jjt|Zge*X z;yF8sU~j2Q`3Zgqsh4H3Igj+APIN|MDdivK$KB7!xA91g=0Y8f%#J0{ybLY=D@^?v zt)521Aw*Os?YwYy$(%NL!$SHNy*>y^ZW^|F^RuC z{MiW^>iT7Jj?HC4sGXe(0f&3TLd-gtj*?9>1QMIA zDSl!Bkm&bG1>@WCRqj&HVY$1W%2AQV6eUt&&MOUN9|z!PjS=o3*NRO8^+#TL%`=6J zs{KMj!hg_Tj&?X%q7%uUc)tql`AwT4+f#Br)IAutr>cGZ+b%0tr-SSO#2yCpuYyY_ zQHbR8YZA!5_UCp|$pK4bD+d?vwTx@P(>r1(p5Uz|%jBzH}u-6rqFRFVn z3ZS6iU;g$b5*7wLfj&x6`uk2#QA%wu{$EqLM#yP*k-SLHFyQGkJ2BiF=NE(17Wb_> zBOP3GWL$GY0s~A<2XX=^qVIpIqNFqZwxJDIUvLXe*0=Ri$gXy41FpJ_a8Q|qfzAE+ z9MjbU3NFHaRv_}Pm(e}ZHe8ik=cCq!n?(HFy|TuEx_<}XKONMJE7I2P?75rE;P|2o zu+RmJw%9~-@Bb^nUDAthW5qd69XSJ&ZenrgVk^EH7GgSetuv+;a2NHz;qc)@^XWZ} zzRlYnL+Y&Z3M5CKcQR(LU564Qmxwa@`3t_B)`4D*ng-1;Z1HrmIygm4JHC>f&dPux zw=lZcxBT+)m}ZBSuZA8>b8DB>|Ba8dLqph6iwzvw34eDaiZ_WqQqMusly^j0w`CZx z(6F5yp$AzhYw2HkeS~)Fr+g8xvKIc1t5++|P)pghbPwv<*FV9G_yxnB;D;}8oM@|` zL!B4Wo>xL8l@c#6**(7NIqJO>wOpR|^fhO_9;I1$bwW>wASJJ8m&R>>+Wo?GBP20x z*+I3``rkQB8zdaP+o$~4po$xNT*F%&oAz!&Nzt`#cE1zXt~;MZ%jsc&-=w!8Di}f*eC5*+x=$Oty zqsxlb(#PFWETMYd`hKN3q#ebNkx@DUUBOl39#lXwTMcK5pJZ|ca%9*--n&o!^nQOG z+DdRusuSb;=C95G7rxIxo{q;`Vji<4b!*{>z;};$q9=GEE~CQkzZmLIA^{7bY`)JB zH+DKgw|5!h-J2z-%NU`0R?Wz-wlnhz!DcSas%6B5>*)EXa2xhdyA^|Mq-@RSM9M3*B>(eD|Y`jhA0t)JbF3i zV&3K*rN}l3E9z>G;O{|%5v5i46)M=z5YO(&ar9~%7Y~A6M)>n%_4z>r3am)UF;d#f zN^du&{lF64wOCMyI!&zs0k9wx{m@aR&ab<*K^oHUe(?+CX~5_gUS5RdcTWe49b)AeXL1y+5mK5Jv@tH+7_S4T;Y zQISE@%tj~StDn%>_;5asBX%vVG>PB%nJkGlNq?)OuUM(j%WhE)l33w@a_f5>VG2N9F+J7;pir* z3wg7}Xa5$@ey`|05D*1=D-&JY!d{>Ae5DYf!+URFeG1SlTpyKJL-tLby~Ol z)O4gY(dC?`iHx>Rkz#cHJMy+W20)kt5h^XKj~caPSO_x2b|}o#nxCsMLnJ-d}t9_T^OrS1<$RcRVB`@|>uX~h& zvHNurqaB-|?c>k?BqlO$Bt@)I$?o7G&MN`{f$uA%WDz24LcEbi!m;i_OVQ57BfqQm zjT2bzFUmn=LKoxmE>c+M@~jyjbRq?ZsY2cNmWdYum%lrK8~k0u?3MQ>!aIw$pQPM3D-JmS+e=8S{Gj>Rw)Uxyb1;F$ z0N4Hv3k3&#q^xVM_7~yx!1P92_GRA){ROkSsbUB2h0~;3vbq;rLLFbwdVKqwr5Di0 zdyi(JiI{PwLl>bge9SS|Dph;v0k}0oq32mR&(%KY`|ej?QY%VvOG#Dga-^cpae|K3DD>%4f&vl16!U{6ePbrh7x(L=0)$2HfJjUZQT@l)==IPG9tX_27 z1wl3Dt1Gwb!TE#=AowW;jzOs^iWpQnOi#TfKK_!_%9UfHr_=|wOqcp>GKW};oTc2! zyp`LBGN+)P8UKY|Ujw}N3y(PJETu7;S`V}vL)x2qFxdJJNc`uo4C*Z7-pyb4UHmCR z*@OJFPr3+3Q_O+|mPue=8`!&KN{ZEHJU_tEPAsQCZhUa@e zABVC&NTJ&&=1#YWH<#n$2UM5qx;Wzn>&oiPa#3&Y4gEw@ySN97`q!;|LZAD*=PmaF z-rSGL;yE*uRSuNNrd;|wYLlNaz4fDE6%#P?IlljK0OHC14pjM)(vfi2BEt$*F~WS@HhEOJGB=$zb$@weM7_UaSt)p*6zgEbYsz&wz4=7Xz~ z7Eef;`0Udb@Nowi?EeV%cj#aMm(#eG?eg@l%Uymr#oa1wxF`lJvJ@+>^-3k=YAXNC zbkUjx*Ti4%E&azO?3AR6q}+FzNFL;`t|M!KWg9wH!Z-aYOX?ju#;-p0MdBwVZe1|D z#9h->U!G@sd6>gUc=L+2jTfIuv!)mne;j^|$gvD4ZonoZOw39; z57CaeWowF}^~kA>Kj}4>(Z?_B{$O(cA(rO|X<;8dtL*s6G1)06#>;xfk9_(Sj5vjk zJV!}0ANH+F@8LerFiE1}weESJZjBLc-UQBY+BCKN=x+JBT>oq>@}Ux1%N$M(lp`8s z+?bqSe!S$lp%As~!nZAi3bwuz-RSsJF1r!KOCgUV@8!lDI(@Xn7>|dZCYIp(y?cLe6`s1 z-n_wns}c$f|1NR-Q*HK<;nio$M!X*OjF(xeaSWP*JTEd1ZS+n+XQrcu!S#|z(l5Q{V$R+RQzuze#e?&GM9W*PD=be z#-S?tA+K=Dsab2_;#SGqF(}|ktA%_ZabNsR#s7~5`aP;4={5=Tz9Lw1cj0DyS&>}b zzou*ZYJmfkWbP0yh?!UA`74lOQcIv}jr!=D!AWL7NO4EMG}QCNM?hLyA7H70w}oZ<^UG`NTsG_=I0tV zf`23ne~W9M7MkBWgj{~|tUmWv3OU1u9AkHyQ=EM@Y(0Qx&?hEXtVS{@Ea2MuVf(qB zvYfw1lez4$qeh>|Uy83TkGCv*V55@u#hY9=*I6(497-`iB#}~m3KygjT}^F$|7^NJ z=Rd6vv=(RLWSI$07kOIHJ;ebYy$uw-4oqa!pH`_sA+}}lR^Kw%(*u!=2<#P2Yi~K| ztUAKbh2zQ6+5Gi=>cYr^7Snd2(Z$czQpjy-2!qH;SiBQ^VW>mP;(-Hbt9%W?W6mbu zz8)=gd0-cV2>hQSCah$lCfSxCKp$Y8k&B&i7}S(hN0Dk}t*bK~4W-5b!al2nex_-> z=qcU55X&u7$%85Ju?V1O0*hc-xZ6R5IWLx{7lk>#jJAm4VOv&(jl7{KyFg=zQmLdQ zdlp)@BdxN9a=cpq+1E;})X}uPRRGQ<2*P;E5Evk|-|XaVTE#pdtFMkf-hq0mPG6|a ztzx;?N$#nqy?+|pEl@}wu#6gacF2lgOQFoLSyE)6h5ot}|4bHdd{)Qoz`{%bf6oEx zrjKoamo|}A9nV`K4@UqOZR%#b|F&0SJshqxX5l^fO5NNl%LZ*2Mp5zv;Uu9Qz#ZNuHI? zhef|V5D;1(#U?axY)3t&DiPwscOG{?vq(vTrQbHwy4sC%Q+W=hwt_GaLyqu4wf1b# zvf~=os1x1zx?>fl*lEUx%1iE$yJi0}vlW=pqA)8SUzGYkqYQ>7t=M-8WQ1eKs1GiutINS zgvo!=R4D7FgTzvD36tb7$zCqIuZ{lBFYaK*KGq&4->=U6_XWV)f_9-P-p1MCxd)hE zvNhVJsCv;WZNdt_mBq%qO=#p{T3L5n>wj0<&h#|ta=-Jrxpxn-H3qidabQT}!Y14B zz<`gI7Psmz`~Rk=5-TRS9FQsKk}Oe2s@IWnD&Iiavq2vGUk$Z;N6WifwIt^!0O$BL z4B&MbaE3s?ak{#s`fyDB5dyW@ZzzuVK1*kryiJ$YrS@G@#t&1*)$5N_>P7p#z6-sW z#uu8#+&-ra{?1&lEP}cxk7-5LXm4fr>iz?E`@KZU2nbr776G}-M_(IP<1^2{ubxEv z-`7lXh{=r?LIPh*Gx&}(RHH7BLaBSJ#3_hzFC`2~Vw;p7$qqfW*TVPdz{~taH0vX$ zr@JXqY7C|SC|XXLTb=7}h5<|TUV>*sT=PJ_ud<1gky3jD&~( zZUXptz6$m`1kZ&(AwEgSP2A)LxTMWvY*jS0V>)mwT$H%mBWcwm)4#7~|F?AZPoy11 zfRKj?qa}g~o42JSf0<>#fFI;@G>0@!Z5IZ%|8-}`Q=t|3PN?k(`}BqQyusClP=2K; ztN6S>@%Xz0`$yh`W^Ig;+}yFC1yy3-`Oo`FlZreaW*^*||Jy1=AtM-|5Bomk``_kw z8sB#tzb-7|XRLKzCG=@0{EZ018Bp57+fvovTP<{+9|0hj0DaT@UjO@E`zOx(^e+YC z%U$p{_TC#YcO~+dzRV+=*i*rOgjT*w3kup(;+3{?%^x{_IE0N4-)=DC13 zLjhq;&KJKzL|lJ_$@22H^1QGlsCd+F0^{K z^pQ%BAVv7x+1r`mG>R&aj=GK|0v)JFF-cm$iG-BHJc1FRNA? zTkjgvisHv%adzs}!amD<=%nx_qYAOeU8SfuGwT-et5(?{2@@=Yxee%koL0)m!=uNN zYVWAt)wGtKG~^8N627@qZ>S?VAUGI;ZBfAz`1FUznn6N>f-!f@rrK>)XeO@^Yz*Ck z1h^Bizr_|#K2{gKP*-Z@F}~z3af*DLR^O^aDH68eE_~|n>3PNPY$)8OsR6EHmrG=w zYpzKvsjsSG1-3`Ye$n`8?^F1JM74GpfmKf16+Q2S1bpu4bRmD&V}5ZB@Y zhqk7VNMZF<6Nm<5uS>aAC z*Uqo|L$3S*@=a6nxr_In8TAH;-_}=dok7W6=5h8NFA{N3RIt0wB4qU z-lS_$K(+vnR?xtFIClOM%BDGV(o){^t;+B=jP*Yzc=`WMq!_47#6sL`E+rpN*&)t* znXkewQh+xpoyoql$<-}6?dO4HSxfR=$mCEyFk&dHZHJGljz?)_l-fm0qWR zsrRDob(P9p6_1+tqxRi;7~GRRC{Ip&=nuIADAmZ8%Q~D}(k?bWLN|9c;$nX$coOTm zjs5To8?(qtncLL0LjLpZQ26PU-_cdD@m;ksDXef^ohYi?z4Ns^MnOWYIk~rQ@Apm( zX5*UynI@vyhw#$t@I0cc48Catfm;&d66d1Q>vCB5n@b)|bdZX1C2o%mR=G(~v0?sN zarx{!rJCz6wgK4MF9_hn0&%?eZX3)20>$#6Gh^3XrV-Z@0rQU%i|kd36lTyOuSAsz zZMU7azZV!FH#>*f|6ulYmiY3=YusEdEtxJ@U-yoWH(>1L>&WF9S2AZSLug`4F2Mlt zN5nPIK#IU1{xypPfInb#Z)<(rYo)0BNm~%}mOxx$E?)AS-hzN;xjx_r;BgD!U|%X7 zR~6e*@J9hAExb*C=`<;TZMbqToVnzk?{gz+;&Zv}hmJss^Q-^{#N>_^NbD5_$QM_q zu}?QAYQg;LaWgccFNTYorWKt5rrl1{l;x?;7TF3YEd)*-RGLv;!r+s3!vLk$uG>b-x?oGr&f-QU zYE0}EKi^C(H)BNK%LeU7>nsk4V38}x`*&=fK}E1gtsz)$)7jGYztGnAH)q1))Dztp zi}rIpLrG}>>lj5K|LnlR@qkt7uVEvUEJqZZ((4DCa zhzU%W8bdej=Uq(Ot3%b|?ScCuC$aeAsx{u7LLT$NjlybqX0G+! z=C`p9_=&nvAT%r;1mX>>(0}ysZ!qH*A2DZ&mBz4O`&1@sg`(&mg`0%Ng&R5sls5j@(O_aa*zq0jgiG{{ zD>6n*Lxb-EhX_yS`2~aMO+Rx0N>VdwM%k@y7nx15P>|d58ni5-e`D}a?~T+k{xg;y z+p|xSz&g{@s#7Q_-%xeIeCg*bZpf3Nwd@Esz56xx$2A1v;N#s+aXg(-ujxv?Ik(ql zUH5MiQsTu;sPCf8#eOomuky?dh$~LFqJUD@uL~}eNI!q9*@10Q)la#LW50=$W}zr| zE-NYZhwkrJm0^>U*WuwlWh!bTeC59<@$@|&0=a>=JqHT>dpK*&?ZL&Lce|~>VKWao ze4w!Zb%=jmXrj0Ne1vysfkaJIMhFNb(Ym6+k;zEd(Rae%KA=04NO`~Vw03?1iJzqJ zg98>>-Vpuzmb1p{q*RtWHHPWTR&~X$QtR&cb56zeVFt67VC~l1sbS2Q0?77$u?2T` zwnOVVwDUt7Dd}FNZ5B&Ck%K;O_7((%Y#kVuS7x}jKCYqle3?pOlU$;9xcJDv(KdeME78O3aN?w(mwb4+g5^JC-2H?erI5#zrdG+t@A$>?utl}F2Djai`?N|{ z^D^6K4z45IdfL%prBi9+b7^}&Y?lBeRFGn5y#(5E0v%^1H--0cPw;eBGpdvwGRgpA2faKe0F1^GSyO>pDZ+6)VTS!XsSY!<(2+&0 zi~;w4ZI=^E;fsfbua(LwzO4|XWzI*xvnx4Czx+Cvk~$yoB!zNK*R=3>Ub+WG%hHSG z+|lyki}W9!Oh%yZyl3rKa%pv3exvhE`ovDH=9ZdEUj}z4y}vW>DYta`-cT%yeJD}b zBk~`Oq5K5+%H+RQ9zTN7qLQMYL~$iYc+Y8gIpR*zH0V8<0&3dgws(jkDziQNsQ0=3 zKJpYLlUQh754cs1{>^i;NRNp?&a^U^I>2z<}O|D3r8?V!-9J=RA z%o{^r5+UGC&ZC^{^r)p%r$+Ojkh`c)uCc-3Cvbx2rZh=iMPv-?7HF!qUdk9{a)5lrFKHoQ zlPZUj5%pGBAK4$q*)y8-Lssuvj_&5jWClr2`=9Qul}n1zCdAj_z5LuuhJX(LNzro+ z56EZWce5E;$LUTWUd7}Av>sl!IHYGj-#)J|VWCkUjwbIiU9-TLAD&JqZ{dA6ih};d@Pa`Cv1Q~qqG+1X zbgWcx>S&imE%9r|0iuP@xNLHLaW*;1>p=xi(vore$C}LRK93&n&CzhP0dJigEYoje zThB=2WnEBNmPwMx?d44I63>o!TPH999-SwJwHckib3)Q@ISac~{!3urr5f*aRjXn0 z3t9hjSj!qvea8tgQpj$*m6$@p5*AF$DT8VUxS6IvQNNL4%m_5m{ zD>xz$j+JqKj%(dv=`x-YAEKy(Z}z!jvmKVv^Qc5|(%El>@jvkmoZ@p6@L%Wk%gL2A zvR$5v&3u>Q6)nUaQ17F)R2mxa)ETcuz;>L;k5fTUn=yLrt!xqJ_o1#yq)r%lH$C>2 zcR1q5{fkE}&c}h$`vDnMt2v*LrnBbp7s6o$fO2ZfQ7BIkSWJ^cV(v5(S5{g8`=wybD|WP0F$d?U{5sc2W2a~78Epq6RsQD5`xlzn}} zVC-2qN}E7*XPJev>b9PROXhXsTF^*r^+3~A8rPJ29QHVzF1^r@Jk%r>3n^A?_1!X6eVbsrZUy zD9qpl8sid8tEb70CnK&svmmOcr`$+(mrMQt&=q@KkOLzvLcVflui~nQyf(Ded}NIn zbz9D9KJ~lJnj$c@(=^@uTJ$j=HM5yAileIG#`n+#NDrQ#ie?a6MJ*<;+^q&YZ*?}m zusbpK+Q}`>{9VPNxNdt|BX`ta)g|nP=@sw%u2fMW`t+MnJnyc$S;J2Q5-4pBP>R-6c(+lMmT|dQw<3KHPjX4zbC!r?EMgr)w^z7gbV2w<(R!diYajXod z)GVzxE7JExwZjC_O~%3ZK%-3q(k!8NV|{t$c7aVkVR=$zu{Wyj80u?Wt{PER^TGMQ)0QeD)=bo`dBj@&>a-R z|Jcwr#X`TM|M~6MmKmCD))Var%>aIzm$VU7DFkB$4)hn)J~fFG@5BhlUb5GD`ix%r zsGQ|sJ@Me2QxzkCWe_bWxjEgqdE#R@S@{UYGClmM>NQ7Z$$>uZFkLCw zEY~99M+t6Q^CQeiWG(9Z<|@1?!#Cm&R*d43Vz9u?2n+JO{q*vkC&|Dj>Szi!OW$Yd z7%%%M4wu{F!Fg4SG}0XRV6!HaPv*Gc<}F^fqM#sEe$+UtLE#Yt-ETt<-Ie=rTi>p0 z&lMv0xAP=iee^OoKoNJJhHMUVTA_Sj#ftLJ!-h9!f|GiqjJsN=-d%vV@ z90%d`S=C^DiW>-N%PJ{>cu)1sao+fAuwD#Ogz7fotB&da2Ma;;zTOiGZmWl(Brmx9 zWO~PYc^ahg)+?eTx%kfS{LZcKj713A_y5u_{Zjk>#3kbi@@JbY%W~xVqEpf4h1-gJ zbfDakSKY(;r`Scta z=~@P5F9?hFpE@l~rZgD0`UVwDcP(Pjq0MzR{`Mdb+uVKH7;;D( zGY;xx3^25ElJAu!X|n|xZ~~q|!Y!inzm<<%tUYVqEcJ4#{0Dir2$HPbcg7iKG~FXe zzHxH4oy%M8sJN9Q&x>a-Aqdw+3}QC_oX=mwTLFR6YU)}^k)8_vaCmk$E$u?Dr_h}Qy?Syox+vHUci7oCR)>kEI z>pwaMoi@nTmL7BDP01EzhkoL_;k7DgH zsce&J^l37c3^bQ#=10HR#`)hotL6!mzvE!8WZ$fhscvR_(8uFA2&`5Y)u$jG7qR;9 zII227g%D5|m4O(L*eZMEc9t8*bUJO?aSdgG6}i1EMC0B@7L3eW0YFIh+(@I*Yc4xR zK=ym&Zdsb@`eq|w#CJCqSVnVxg(X1DY z+8EU_BI!OOMS{^s@kS%r&ifxDQbuQ`ffW|(DMjq1d+3Xx1P59gJ$jC^|81q)_^ecp5?5O@KR-VK^Ge_AS1{$euhks>%w4?21OK zjq|S}Dj^oAQt{u>H~f7~wP`WHmHqx8xi zw^yU*(r84Hr3T9`d7wN#GF*gM$z;9p{&7&I;Sl6TE88qmh3Jg5=e*0esw=)t`cp

R!fg6kI;DPGDW4ZnAQLr}7*fm!)M^U5+`+ntXP@G^vx( z*=p7)ERMwA<1cTGtg8D~CStY4Vn&c(@{eUp)@AP?gS1KTQ6I_GDf!fS6$R#@^>JzN&xNtVKs zykKp~@;EK9SdXy62@;{z33X9BB2C&a7>Xlf#&}6#!lp9uL@bBef%1`Bk%u|-NchYQgKiMEHf_qq3krOPD z$USWhIiqbPzpY{@AI8BuV(_A2OaE{m_k49qNI7b+OwqmMFTFvU)peGhdnq@*SDB@M zk(b&aR-^J-o^hWfH|2$LuANq&q(`tKsJtTQ^ZVN$-4AOvb%D;M&eG@PmwK%&PMJf_ z`BvWtV`YnsmKJSZzSGnx%B(Wre)lM!tSJTcD(kF!mf@p~Finf^XAvhzpL!!trBhIL zlqpuK+8}Kh{ctLMwHnL29!$p8EU3Y3a80CvqDVKfiChIWz6?r61 z`9dzq8}UkK;-J38K=3xJDedk9b5=(vd|va2ncP@dslzW)z4RO>M4E6m9}a1IlEX(yQT<7VcWyF*MK`dnALo zwv^v-qRUc?urOPPVa_{B^4x+2_>lmq!oxkdKqY>vxhi#UFc zDuJbAkO5JlBciioU_zOM9U_2IpyR3hX$O#N>M4?p63-0@;ltenp6-RWjsU`0dE;UN zPb4Nn5SH12%rwm1D+frctQ$HF2x@K{l(=?GD}V7Q9FQpa?}VG9eLC7mXD zA+5eyeNfkRIOMkugL}v@a%&lZ1w#rVq5#>Xb0ghkp-!;+BaJX3x5TH;kaNfp@{z12 z%f||25?BTxVDCGTok;+sU5%A2Cdpx-M_Ibt|9p$UBrL`HmQ%XV5Ry zCI7ou*(VG8G^z6$<`P)yhCD#ZBXrakZk9TR%9?NVebSA{R&U80WFuEQq=Pi=CG-TA zHaFi3V>%cDS~>1z8rM{$zrAwIr3gWVcvJtzAqgy`Gu@7?@||!@f>mB*o2MQakb#G` z0{%%>(>)RD)>>X(7|d zOJzm;eITc8ybHNTR?*|+3BA}i$s6@+EXT=2c|+&mPFbN4 z(2hLjF^_3F0UcHu%yu}eI9uJDYDT&kiWKH+{Bn0#}g;UtUnZm=H?MfWJbHIeJEAU zoj@937@gMxGsg~PcVAry`wtGKy}6S3hEOP*8wNM0RY5+;vuN;E-PZntGuO>83m2GB z`WpUR4VHx<1{XnKHcYuR?S~5?*NK?Fq{YeQD-^zEuiv^J%4l=N;dTTYb!#d_C#sdl z=oB=w{U9ZpD{hSJE-jPdfng8`%`HLvTrbp{=-5xaX6ssv@V7h)5~$TdE@RTLmmn{r z1JYb^q~(#1eB>fJVY8wnOSs&H!rZ=JeTWXk$~xIcEPve}otH(B=Sy7(qm$RlPIRc6 zI|vy7zhxm5u&r|BnILAGtPsyQ`2N2QS08z&j0VxwWQ1>97NQf@TuFkgXl_VwBLm34 zad0gM8bAtA1s$H4$Y~GV{A!Ru^&Yfx%=J~8W7ee{OT}i&tr{J}!{zPZ8xpf7J(b}3I zI47rV5J!wSH(a8#+T7Hq%1xghbx<3s3@dLj0MK0R`+IhKtPP9K=B&q~)8AYfO(P&* zwX0J>{&VX}SMng}TA7izt3q_@7cuYKR#^^`X`2gS|7>*9x2;0xJLUxLP(k*#^phua z4fSnVn@yG~%Ryc?m){_|7C|I#b9L_D%l%J17q|K{Rj>N{9m5KqaOLs{=Vd`U`?^K1 z;dV)P5rV*2gau=hXF-587eX$M`xUh5T!Mlm+WJl|lQDR=RKCjhAm!UM5oexxW|PO< z+m^L0z7HPCIvimiWZBwJkdN|{uAz=_`wT*H>2pC0H#a-FTM$OA-Ug{OS?1(XkT*@| zw$F3>mVjuTy!V;)|?=3)tw5300fjPvN2Z%RyO;MesdfHcDKS31at=bLf}{@ z-P{2&a<}BEx_l}lFhyWZHF5+_jHE0m;u=?-ZHfVciDJ2|A)`p?^n}TJE1P`h`$lmY z_8Lo#vpHsz<0+Ds>#SDR&26=eyvld!H1Nqs1c*CK@0+sFca(H)bP{rojTmie zitaihB}T8>+z-+t?yyp83d@{t&h(A<=6r||uGV=Hr*x%)I=*+W=Lbn(QDTq?zJbyQ z_ZYojn`GF{#ZSH=Fe&ohV;-ty;V2VWIz_$_Nw%)M>j)w>V#K;FVXPxRr;`eEot-2e z%xSW!)4mMn{(4w%w80b|zpNiIlHA|2^jO{|f#v(n=VJaPan1>I2rO7-84U}CHfK#( z0IrRm%tUS=I#v-_lr|j|B#`enkJ-A8k4`88-~Bpq_R9z?@<)fod|KxEixK*@X?Fuz zYW^Cehz=-%a~ygCOD9kK<_MeWkoC{%wDofp`sTS|ZoOX>pFAaJlB|Z0c^=ov@{BVH zEO|?AiD#YMLU@J+T+1u-?wI>a-i(8}pb$G@#nC$Q%EFwE9$gQPIS7LkT}KFw-#vA{ z6In|ZAOexX%A~TOtd7HB5Lifmc|#`&>y=5Y$LVq&R!L2cAcvJ{WQjSGRs{nNNqQj$ z41TEBQz0z67DyVhfn3m*uM0aRu*h=Q%LDI_9_5N$G~kB3QK$WHfMs2MU6;Q8|DM1y zFyK3thjkcw&VWb`fST(|opOB~lECt9NMz*kR5|OPQ|8D>-<-tPG99&l4iRe3!?4b3 z15s1;maNP)X(Pymuww0_S$aHGzsM)$#Qa#&i?mdq=Y+88Y6EtjGcXlarrRDv%Fz?# zzkw+8S$Q5HX~|b{=P72 zBnuNt8FC4krIDmCOa+995i<&biPaHcbA!eGYlb`};Wa7@j?WBK>=U?t^QZ3%0sy|g9!MPNyf3FdY9m&u;-_sk)%SO#V1ROjU34}W;;T#VxypCc@k zwQ!+XNnq*d7-i26rO9=B;(Q!$p8HO5OOtP(BCyB=GBe!C+CAj8Avs_S`*4|R&v|~C)!ac7Sef4lO^Yx{3!(?nPlq#Dfy{P2=F^+~ zeHrh%y|yPUu)<}rkuV00l$BLM8G#|PM%s{G21}-%%X;LySfwo^3-f*AhCBUD+~gtB zdCnldI%=@Tb6F=@Vf8_qo7S9kTe5Oa+#TeG=j*r-7$ZPmH&~{=*^;-=&qThj(=EpFZTfi; zSW9wV)Yj#ZYQsXc2T#7@j7(iDZU!kX}i8%bb< zD&_@fCgaTQnbGS6AZDhvm|LKk@f)?*;g$v_+GAI#p1^XC4lRXxU5A!XL`GTZ#dC-_ zojvhoph&SeVY2QX|F0vjJe$foHwn*C?kOa5J4^Zgeh({gQL^oIsQR?__wMz0Qwc2C zGpRSSto(3UrSws#D7Y;9?2-67QfYxj$@1J(LwM8)EQAT!zO4Q!3(BMKk9mq(*+C+x zw>p%o>Y)E@T%ocovqDt%raBpO%iVquSZO)Q0tJb=>ihe2%^|S3E+^SX7AjM7Iwv|I zI$-LjjutnysZOmj=bj{@yl+m>iHbQ^wp>A`2rTKO45qa!(p|@%JYAL7DQ|W1tlTJl zI%r6eK8y)09Ypu6!(ZLi(cwBMZtu|Pk{<{#-!|8Zu20Y0pj8P1Z__cNy>* z%i70^$7sA~TSkjf`2x5dA+?=`&KkX3R1nK45xx`Ib zHXRm$rHmUKL@tm&F(2KQH7Gg8T9T|;H`uYRY>fZ!2`o4~C{{e>y4Mc|j-PR^jHvfX zy3r?DEV3vXhs+}+4C_e)R;z-zW4^;i2ul--A?~enWejUt&uM26L-Mc>thC5?_h?U* z`Mv04m&Y^%bQS{};)IFxuaenh0c$=apn8|&mnkgzpgbFp$#P1MCUcZu?uP!KJ0K;m z4Lq0zL^{>`M}&ugbQ;()z`PgTpig%lpE?LA$I4T`A%fNK!)TsB3?U=yQp*K25nOhYoIG#=Ai94hb%f9iY>`(^N70?LxoZ#I?L*Xag zh!pORecW^6>F*)nxtJuG2Gey)T>tz|jyI9O(s4HG$qF!CO4pSeq^Wy!nsw}^8p5MW zU?s`S1!G-+9R-u(e`Z)ss^jah8~+~@$uY?4z`1eOjVxuY{qj;VsX!pJQt{R!jzIvOFLi*5%LmfAeO!#!_%UtRWEMNJZ^k4g!qd zIX;Jp0V*UpB4?a$n@C_uqjH7>^gXO}=sI%=EFF0|gFG|OnaluZ|!%r|?e#i=L*OlS`|Dyfv7R zgP({=)*S0dwTx4HDNX6FO`q!nYf>F-_h?ry4+3hL4%|Pt>==C zEp>i85YTRo%R2c*-=8B#juHyB3Ss2*R0tu|mOqYBn61C9v|;ret9<0DD_eioK_L(b zgn-yuh0qExR_%%rowk}uG*)d@82jjWwpF&G1J;&+4^!*5T9cIot#A{h)(Loqg4kAp zvGPpsA4SS0_I(=U2XEAU0w{6-nPn5OoZN6E{jfETY)YblwHR7 zfyY8IYO4;#Xlq-&%PMctfo&@Uec-O`N6JlfZX15ePOg@x3|JjUJ}bAm!9Nt$wvmaU zP`6dmVg$0SkkiK=4QE@$$FkP&45GHJnl#QQs|Ur1XIo`U{RqpJwgQjUhx!o8Ut66@ z-Uso~mRz@*TaW_%Z6_Ru(%x3kic!(FLSPV-ZFQsZjqnSrm$qbgkZUEH_Ym*Zlp7xyb;QPs)=+LZv zE^m-0LAo~_l%F7++BfC@_F+BHR>uh|mnPFfDQ~OgSpjNY2r{azq@}K8y$P$yt;fjG zAfwyLW#JCoR$+}!M2mmxrX>gDZCDq!)w{$I1Xe49%43G;NDV=jH+u>1->LDaZZlZ_D1Hyjd9uj#d`4nxR#<)&!xPHo<*)$HW0RiovtbUok!4`{31(p6J;-1dUe ziFPRnr6&7g;H<558-uS~cgb_I*y_kua||MXWqP(ATyhWuS<7Sc5uWsqowm=SrmMyP zM#HCnmHMa6Rkzi{thZz@xl2b-=9M-0T3IuQ&nEZP^Kr0BVGvafgBToWx_!!VvdH`7 zWejvQ{Wk`^w^mXe$5rQrKzNTexl59$N$$QRFg{bE_DAGJ@Az4UE&&pR^_rbco$$pVXukr8HPhK1c!Fk;QM zE8WlLSRk=v;mVXN2eG15XN3 zj+v)Td*i!P){-mQW93&__YE-^(I$28%M4t^&607GTnMF+#Uht0X*bcnc6w@}{i(cV zS>=AZK@bkj`ee$r^e`>3)KBj9RzFr&$r#r1bO&WkJ=pI+rqStGIl<9_KHf!loD;|~ z&(XomNveGrhqS;VE4VqYvp$!%;?2p%#Y1M6V z@@+3c+Dzb{maSa=cB&(j1r{3_r&YS(LC>cnxks5-XVr7xU^BO$c9mP8yrj3wiz!{& zHM)rQM&9$#SSH8m=B$E)l%3s2OTEsxBQP1jYmwvvVs8tqMqvH2v0DL#*g1jTKkQpau?s6(hM#L8o9Tk2-}r*`#ya zn2rf7{`i-NIrln6O*(a?%mt*5n4`peF=GPDoHG;|_xRp9!MiD~Imu_vuhqvrC!YQu z--K*4_ZXKEt1Q9sAi^dQSg_Q=i&4Pl3PD-a$uiL@Jg4NEGQy1|=L0+n1QyK0O170l zWoNXkrQ11bOzmHF$tY?vU4Z->|W1|^LeiXR(_j0 ziu{l^a}gpzxNhpWg|$S(Uiwcz{q)xPHowYVf-*Z^g)r;DoOlsY+Qt1Qu(+^s!}qP{ z$3*YQBYCLvztjBEIUl!=bG|L_@jh)#4uT>=m1ksxIxGHhdGJ2@M=tw@RRos&&AIe; zG6$A9Gu0FEBS+-rR3J0yA#+zD^esB*RxuU#PA2;He&tCUk8tHmWl&)3BI>@wB(RXq z76Kq!5b#I|bqIbsgUHAI4r2n#97_n9skV9iZqMezQ{N%N4j;HP{UX%4B!RM5VRq3z)jvoT4Hz{|y zA>rvSWW;d_MPMz-O+rZ{ngu|o&oMTg^~!N82uon7%R5A1iojyAu`Vc6ER=FI-l&d7 zu@5;Sh)t(VQ6zK-_bZGEEb|>1VOrNn>L3_BGBTu&XsjuC{YXE8#GD>Fqwb;T&gmeC zTPGE9;yFsxRA+N17*IMWRTSl%2MaN{?mKt#j=XnM2rSAGg&VP$c|vIhIt}R_q3U*1L#1gqi0J`b}BD9TtJbI!>NR3rl6zd^Uq{tm|DjIr&}&(x&oC z8SX8x43H=jtB4QrAP}XQ6|uCb?~jN9$fx(mQ66=;xCO22Jg!UgZvU4S?w_n4rAOPuWok}13<8TAP)_VDKakw=XI(IwjWCKi z+Ex-+h<^lE693#9=LB_uyHY+kMPMa4$T}eAyJ^+|IwkuRbkbOXWg2$sLO3@aIzWhP z1QtA5CXsc!4G1jxlUA_?QZ~T?%lD8M>H%vHZKQObj6iOt8<;k4PCKphp$w$FTSj2% zgvdYM0KOGb=erHetP1M8`j@Uws|YOZHw(G6P@YQwa5tAGWzF;BLAYh8NKm7XgDq>FTSl z=Eh)vY#}F(Qz!y!MQ$3=Kv?ywLu7#<5KmU8Wkr(fuak_xZV4>OHsXKpd*8b$N6MKw zS`aP0m5k9Hnd`yGBPDl^V%(D)o?)rbk2HxxW3IE~+l(?( zq}J6T#20t4Tyc>CP662K_nK#cg2yHC%y4ToM+v+siaQnZ*`f2zV})^32`nUIOrE@w zob+7?h%fr0FKV5A`F5BDmJSySn_Rt1hi)B#Apgv}#2P0pPWGF?(t)(%X1YYMEIe6O zx$1|GOU~t_e#|8&4uim=H02090*m#+95+8il#V}waFqoXrJj3Mx{I(t7?1FeWg+>Y zBcVLA2(X_bu*d;Az-N8dXKj^#D?{~iZ5;Pb&JT;g(iyZm7D6E=ziuQp%dFi7av>(u zZUYjqV?9BZXg}tJgR;Q#Zq^$HI}k4{pLArktA`zuz+!o*Y#J3`7qsKt(^#sIU&{zA z_`BDD5d4uGh!wJ6S)iAYQ7n2`Z+mVPfrXHiZX_VWVlIKD-ku-so9ZncLb+IH?WaDb z)qrpJ-ettsK@(VHdrZ#Vti|OWl8Q{td7{jX#*!xG4gCulzF(myu&hK$W^$`rhap*o z_(`i{Wye6qlzz2S0?W6FC-Vs5ZV;7jur46}w2!bP&yjIsw?;Zs5{7WqmMzQwb?@`P zvd#)w-qGz4qjT!toVa&-5AT5->_TKAC$lW=G%ocd31YIG#p2w-s-3>8?mt=x=ns+6 z2K~tP@y0mfIEEsyR^%$OfXTIvyaGW8XTe}bdr%WUuSin;wc!Ztg}|bKQbcu_lrwHy zl#Sl(6A`2Xz$->8GYCk0t?2zH%B^IROtXsDY%(_)_#e>%-$S(EP^?~xrWUelGuV8dKD({WtR zy`S96IS&oGm~U$?CI?Aik#XFPWGdB)p zNLg4Gt=kT+yM;z3>U+$sVqEnks<+UmKkN=OYv ziumcx^yGE&h_#mitys~nSMbSID;iQ7RESTH4f5Dq!`LB=DoQ%4~)wJ~JAZ{&_i zC!rt7m%WCbz(N3$eMlWvOv~!KbaN?H-dL5!N}#PhQU<5OZU`*#q?}cL=z8kEa=R>; z%ZjCzds(ouu3F|ohv?Eqh@0L?H(56j;k{(hl-yB&4YnZwk@g0C4O(PAE=zM>r+2VA z8t4q|cK9bamgEa)7L%YtALKv>`BOP(ik_wwu!9=cKFdIbH}T*W_ll z@pN_a$;RqYn|2&S5m+m7RVQ47nJZ2y4Czzq_z~f*ak5O*S0uasI*ADEg}|bSPym@c zD_OX}mjy~hx&ul-t0yi{8fOZcxGCZ~Xhy4y{-&i1 z^XXL`-(~Myb=`Y%B%9DT*|hf{qqs3hSKn=upz-mrgKz`miYk zmNaSPbdsc%B@9zzTg*LWeiWpDv{D9P z?wlJ+5~P&H93|gx0!v3ly8h%({$y(sa26AkOA0DH)mh4}@~pn$s&husob^peNL{3pNh8^6&8Gpw4Y{P$%}-GGY@jyYg- z9#gp``=xg;K{-NN!G!y&IX1|B@70bdZ&~)ngBb@&VEHEPk~S0pAbnr`)nDCQ_~4}d zK)#Lh3kgh~BfrSFy;y)qgSw@R=W1d|b#8>-qfJoWVWlmB8(AxVw71$t11d-_kdsIeOW0pHQD!DKfV9!d#y#`OYX&OuvkMfXm+;=-x4h#SxKgslQAalszT&Yu8 zqO!sNZd0PlsdfdFx3xlSEXSyy(r`66=Zmh@h2 zT#{(s4NvkJVeET6M~i_^_YhWCLp0c1cZcQ05DncB79vA5_FLEM;M7DQ6p$f0 z4MQkm$l+Ed!kS=+hI}?4p{NYeSPju=4WXD0p_~k1SukAv>Q~>2Bgmv7Iz@xB8OmY1 z9z@m<9oQiRS-Urs!XcEZAsW{ql+PgutRXsG?OyRmM{g@W2ceV>p{Neg!5%^(8)9ya zA(Yo4I-f%{u0triLy!V3?odRAXuR9Kp)?P%O3x4^#}LZ@5Q@|gqp92C&XZ6+hUh2^ z(Qvotqw_b!DESbI^$;U)!&R?()gcs$A(ZYRl%pY5aT#LXmLc58hOpQgLYd2`_H7>{ zYC|Z2LntII??TCFamUIpLy%`fbXte70BZROBY4W2P&|fE*atWV$uk6b($eSN451he zL23*^yfiFBnHi!3Jw&HGZ`t);-pbQZZig7TY-KDuJwqs$OXKv7K}HT?$uxwO(h!Q^ z5F_(Ln4~wXmEY*}4B$Loh4#yB9fG5a1E5AJ+ z5+=`;ffz9!!Xl@=Kdd;0us%2;-S8D}bS#HZDqGoeOPLeU*!wbE8T z#UF}IE6=V8VSU%$9~PAD`@@oR2iV&CqI zj`R>@R;yng3PpVgGP{+buo7yx$nzLMAEI;L^2UA~qO&qYN2lepyo^ppYp3Dse$NM? zFhmEXl_B{O9qblAdFWX(P8m#W#_u4Hf=n1-r9297CiB9|rj^4W8ir652QrR)R$s_6 z@-0Tohaly)MN5XpyfRIWD;Lq>YxQf}O%o@+H|C!jf-r6WD%;9@5H^Evlh5+jyM3QD zN9VKE7wrQXt?utNXeX3AvLVQaAqM)I90?+Ihz?te9wy#JKgu9I^2R;zj*i$6mOlCS z(!J+`JZ$ZY?{8l0NQa_)3n>A-*?CZ@5>OK`1UM3 z$wjhRJ@hR>B(%ChKEo$m@Y-X_uDqouga_f$BakDjg0_Z!5oBL$ zceT@GV;|I8Z7JEUy!tNfVvyV?$Y*+;`b1X>VzB7|LAnera`*jcy2sUJaj%HYk3^5ng37z_$z|!bTH>HyzMOmP{#E4IOhj=KhNHv`Sacl4? ziIl5Vfuceg({Z5yNvkxATj#^`88@ZHF=d;=fDkx{Kv|+B>V)gWz!DzPmV58@tu}bt zBTy*th(;aMaR=VF$|GbA(n{w=$7{a{EFA+1(mDc5r-M=_Zz*odt1_b@RECrnc-vFX zC~v+KF_0bXIRsV`dT<#NSUPfHE!aq2!5RWTE|3ZL%R18bYv*F zzSH+=o1_a-C{HuU3+aTFkOY>_Em9_|4_ap??_am~-8ztDgKvQMxcp}v@{GI+OQ%-% zb=1^-`DjzGVd8n6$FPcQ*deUd8RZZOw5s0tHgb#nHOK^iIM^#^;#aO?{hGC4h8^4u zV8Bit78g0CE<4vT^L`y6dva9!5flHm4*FipoOJ0BDm!5b)^LS`GKZ|~gU%xoFN1TP z<|MF`+no?tI^8<9h=?fy%X{Fivq^5jH!*^bbSe|_3pRb-bPOBBj!yGlgEASGSFJ;v zWzzfQ6*-`+E1NpO%Ad|D@>VBWM?0*0mJ(RfrmT}&I?7olvyIj^sq4N`=NLhRtR;Ju z`&9&%xWlq$2+PAplt{0-q>~TJ3zJ$a|k6w zh#&TQvRUQ$V1_v`32-apJmr~}c8_KlJ_md&gC!OT5bNvBz(7AgA??(n| zyVWb-BCf2*(vZ3Xo!Iwkzw`;zIe6}e9M5m7{n0^y>{H&tx@ue5SGJOuK=d;dSk^`! zov&Sy()id4GZ2V|qxqNyl+7$J1nHKUUNpqbbt4qBV{5861ERVm5Qs*nnb2oBv1jVL z(+~>gR{0V~bl{swaUV}86U`)B*)&H2)A1OY-u9elqoHf_2r=cC7G}`VSoTr1;}D2j zX3C!9Y8Yfga~FutYMUD(6zy&Cn`4B_57Xypc$d06^e+qH+v*9^aPBqc`U-_-TfEHE zL$PY6vC%17Ij2n@_im428Olv_2jbRX#hp+%+8idZ=(ix!+QjBe+?kPw;&y_2R;GjN z7<2S|K9t7h=8^yJWiP$bnsHl|CPpsTMyZTO$EUeKa}Uht?YZ3qY1!sp;|diDRWomg zh5wo1Gj$ioID{g$wF;7XIHWChMxrj@;(V3*i4PZ;AWXKtsZRs6uoP?aENKh-cRLPJV_{JXSjJ#OYxAb^Yt_4!{U@)2NNa5**V+EvXT}_C z+|aav+D|ey2%07*+($NZ|5%k@(#ge0ndg2N9p0^XaM!hH3b_;mG%bGb;;JWpb0)!P zSqRdp$z^(gbdu?P`WD<`Xs79`{%}fO;?|xKBOsE%YY3 z$hh3l71bsBA-$LkS{0<%H_{JG4`I%}Ao#X0$aDuW-pb>nAN}Yy-3GmLoNoCbZ9#~& zybzbRB}lMs@x?$(lPO$xO}#NCchx=h(i9q|b|A-LIc`&eENg9g5UlN9ZO~MFt1zhl zju!$kDA06adS#Yh`;`CY^NqoSCO3}Pmv8@?vcQU5pA(`{Xd_A70+{a$*l{CE0cktO zvNLBk(!&(A!l{RVa)KFW(UZYwR9ZdcwRL-=Sw^6YIEhmnMq#Eb1fr2`qhHaHZdMEK z?}Ipu&KWf_Dr>Z9%96?G4{NKmWC+EznV)*MZ!%(}y!Y{o$EcD~H}UTzL<7E+?ocwe z>XA64+31o{r*#%xJ9+1vdyVvkqSC%~oUTk8Jk@_#O5?H+%GlOupKnS_lN_B~_S~H7 z@?D;=XiB`~!>aUTyzUZabv0xZ=Fqq4p|gS=2L#G!nLa@R&Ql`osRH#H(Hj!=G^wE!9J*}aTb z`5yJeJJk_!+7j2SgX#->_j~j-N2`_H{{4xQ{EE)NmJIRzM$_eo`=x1)B}~SbqweDI zjp9}()`cu5@+sTSEO*KPcMP(KRRpUPuA;1_xS_1GiV>&y<%4u({Jx!h%8_v+-du!O zbQtYtCCO@VPEZb&Uwd_@mxrGBE@eJP_LZw~U2&ghmA}N_NIkcfIl(*SGn_o{Tl+lE zGUvU@nY{Jw`yIR&-mDrDi^SZ!q)FXlZIglAQFa9BxHNi3`H_}&x&w^isXUN9iNE;!GVh#~ zsQcs<%+CrJB-fV(@xeXIy0|lbdvR!=mz6EgXMOa}eAfHmE`3=x$;Wj;+R3a;v$C~L zKY%?)V>XeUk=dOFf(;pOt%Y3-ZCY%xTM(eW!fyT?koVv6NHRl?~q_FZz6t zX8E99o|9(j)mHgV_ltWFM9_+V#%J_-EX@2Y(vu-^z2fwkX=R(yp(Apij#C!>UOM#?E;=6o@!K8qWp<>s0(7nU}5Dx?C3 z6kbOR8^ts?kNN$~BVrUQr$Ql62owS*3xR_suuj%@)q4wpLSSbI(3!{sR*sBi z%$H{Vx(tj8Qtx3gXwJTv5~IxziP$<7LJ?TI%w;B|CKk4G%&K6uE*4oPDP{^@mMp8# zs{c+V0!3h*O!VvBg~0Jez$l@CAMSz6+$nrLrOkxqtgoza#~q)GnMEwPx=aO195*o= zllfIHD@Zit7|Slim`TU29>%(B%FT4@eRZu6CJib{NI>_u+ zguQt@)!X+!ZmNzU4si^bWj^Ad=*W;+WXN<388d~DA%w_u%=1`=3=t9$k|}fMSqPaj zC3A@M-KTqxd*9dR_xSyB9}oA{KCjo_Yk01;*IsMwF6K7)qb{SCBGay#0MTJMv#jaM z7CSe?U{y9RkC}&r2x7l5FCJeNnJGv3vtLGm)5$@aa}}2D_uTBaFEQg1x`cucdObH~@tN{iwCLou>_7jwGbIg49>tO%?wO*?_&;_i^0{&0L;d8**4YSK zN901hCdX%+VY&W~oEgvVP_BV~VPk4b0ci?%L2JWN;)& zQHq-pHCr9LWkLg%Y6^>_Lgy)%9<;M~Ts3}HRY^obJbxI*=EmdG^{wqnYT&OQ6Z{$P z5xX7|A|w54`wREnF8{@ZSR6dqI$|73;_7oaF`Yj3`*_dFR@-~YEw+3yR_ZsQ32jf~ zx2z#OiQ~8LgN8f?%@a8-Z!U=MgR4ic z8pK6Dka>G~Va{98q=LX~L@J-M^K#S!Rrv)kc|$ z8sQc;<)Z7)^K@&Clv*BuPO`!?M8ixR5@j3br1`g1#n$3);T`zR);BTx8Bu%*RFa9C z79ghLXx8#6J4*;;rHrLP?9B6!Jr(%y_GYo^%%srd*^7&j52+XVczui79+4o@oq0QC zT7X^*g-8iIdOY6jc_Z^)HFvI7(Zr2F04h2cF7!t6R|7uT-7u%1^ND`dgsv%&b4*^2 zKbI6FQ@w5G+-?+Z=CyN~wD9LNA*694M`=c@t^2cfz#Ye?ja2zknlr^gbCF_~3yTaK zYF)=Y7So&^Hnnioxq)OSTY1O( zf>0g1N83-QpBWu==za}xD523M<_cME_GXQ^{OWnJ8EN=7Dqi;mgkNTdME6lRxW`!J z>+?X-_ZHS5zfIpVVO~Y~KcDg?_^Irxf!hDj+DO#DQczfYvxj2tC4B>gAqL|3lZu41-J}E*J3Ks z;C>3{s}B*-v%HwDxMl7O>K|eU=F9O$?sDAeHMLuPRXLSvWq~~WRyJsnbSIHvT;hnk zx8``xJaqQDz43o&K?ui)Gxe-hZ#2yIE|eBV_hf=@cC-<#EEeVL`tT zdL)V3U7uAS9fKXzn1v5C++F*_$`r1;oeW>zmE?Wx!dzr zkK7vN?!=s?i^(&b?3Xl3Eqzkq8XFI5&00cT25*Th^lQ(F9UEt*3=fkl(8s8X&IDW^ z#1F#WRQil@J2FuToW z%=F2Sbqmo6YK#7ccHKwYWc`NT_=x&n`2&;cOU22~h8Enf|MkP=_pd#xdYv$8e%12f zO(*U1ns0|kjUWHo0aa0=Yi~!5#`Xi;kY~AqQpnXCQfVuT8Fs(;lfAHgU3-O1ju8-e`uHOr3U4)wp@I1zsLu?}0WJN#}lVPDzZM3(2wxR<6Pg=;Gq zH$HC?(^92Rz!z^EZ*&}!+zq|Zde55Ox^_ElU8UEf$N9~LeP}7oz4m9&19|TWd`_n$XY$|RA^~NQX)k9a=yE=|oR^XQ>dkm9qUiP2&yG#kiyYHj zM?N0EQTNx|&l?BajP)rS&t3NUA3Ojw5y&Sfc~kwpdUfmWy-`EwTly7tdTxvU#=InQ zG*rWIVIi6iqVuk=AKmc;jkca{DBXJWwBpKfkse7Jd#3NOUiuyJi)9P~FzRjjq1lV~ zt5|;RbHFAqJYM1q?l$g~g)qb;BPBxGH_+%K<0sI@<`_n8?|m$f5BA$+*yV>BzmaF?VOqjly4cIU!YHhPMR$r2XNx4;CZKNOt13$u zGqTu!$;DrPy2Rj^QN?BVOIT(Jk2p2>{MsGOlM91cocs3bvSfmZ>yU;*ZWE~({P0cV zlK&4xy=SM&B68*-(8yjVsSNn-q$MSkL!WJbuHD-hC*x|F$t2|2a{;4!n9Q1u+sypYJ~>5T?h~ zeT2`i_vR9rM%bKJh#tv*rG(jU z3<|05+Yfq?7Gg0RO2rFK)Q!_8-dkmOyEP^at2B<=zE0ySaC}(?B+2q||D(0Hx>vt^ z0s#MDN~LENG06gig`xPZD6h{O$79A&;htq0GxT(krkc2G>mK?sJdv;B6Dt{SGmq_D zgb3q9(q1@;=xd?>+Am)M%bPfxD-{#>JMpO4A{2#Wp$gwfsA~?tzb{J;JTsz&+fTe% zsMo+ecdge$mC-f8B(R0bG(lTOsqj@rY7lLA<}N}Fw=insk++QsytKybx&TKKT(r!f z#@z`CI$*!a9kKf!^4QDr>SqA2Tj9Zluj@QvZ<29i5RGUfuFrq0{{PTM3M!Q^9c@Nq z0@2#{u%r+&ocq!isv)Oq?CnY~xJCmCKfn!Xgd*24J!0RGpOwM0sWkHobl?q!s9v;H zBU5i!;r{;fPB=?o%_a3>jZ%$bj~kiy&0{^lJvH4cFH-uU!4J76ax}_EQRWdxW?$s| znD~q+QFhn9IqKw#{|8;AUJ>?9-4zrJ;10qw)ADnrTpKXB|BgLOtBb__Yl>CR$~A$4 zPmh290=KiT#aB$aO~BPnRs5XJ1W^Ie>F5wZ5m1(#Ev)f;07Co^jdiMC1GDA=s?w|q z`ChT^xks5`=56G>bgVC4N@ z=OgnW0W`1Xbhi-Y?LEuVa82IDR+o*0W_tj!bd+|20KYQG_OU0{ijLlJA>Zx{zH6XE6dV zh^8Wjp@h;5!`sZbn1jes1~W}WABdC-tvd&Pxa6osJ;u3H^9)aidWLlHrA809YOax` zn|LPu(f5xW>5CbUecXtm+J{3L)MWzTUC7O}7L-L0T~>||0r?dMA$2p1^ZGw@*-6K# zOE3>0y>CtP)_)poRgO?Bs8Ru8=x?1gNE^dLL*!OF+u+or^)!XefJb!n#le2EoW*n3 zR+IqPr7s9V9|lIZge@N*uJ=aHATF@idL3*Py)g#b=h51vD$%tKu1NW&Oxy_vn;^a* zI#$OS1>zW6|MOQJs|Qu3_YD6>lp7Q&8l*$7G9Je^J*1L9KR*a-9tdUe(>3z2$oe(< z(l;!{pA^Xz)nN)@Zat6I0=|l+n|N8j{KbVI#(Bky9%uc1@kUJI2OB=K&pj<|miBA2 z(m@-b95$acnH~Yi*it6_t)E5|%;e)f4&3>+Hc|5rfp}5B;a>LwC10s8gq;7M>G%@J zr{Fa0bzwX4xO#gWXPxy8=6Drl@2Va{wXTvWVaY!Tg{1vKjgtGQ1bVdnhyEQy??QF?f= z{E!((D4-(89mfx6#-s}`h%^w#0U%(eHESo16XpkvUAF`zEkw=sCo$f@n{__9BGq`` z#eX~Wz{{4yk<_N45EvqRti6wT@hi}=e3FB>z~lC6)8A>b2jz zOr+$tqjw(OGO&NE?M4xm?JmpTsw-r`ny3#kR{0o8%Uj0!=KlXq>r$T@9?5}uGShbh z$E+S}SO8Z5_qaA%v;4W^CdGF{0mdbH$l}WTwF-kq2dXs=0`AZ60)F$1>#!KP&M{tJ zbiAP6K=*x%VLxx~Ocyb&v3?Dj>Cv~+Bi64A;ry>KeufxvvyF)ee(hOe7`)x`Rd~Op zn=HI)iCt=*HR4O41vRnFf3yu)u*@2 z`F4+3-of^wRZQr|qK48%H~guWM=CKBD( z(uR(@{Th#>w(Fap5ves0tciz0u%Q%NPx2{DF$%U*wFl#n?3udQFc7;ZCX zQI=UqKmAM+kuDn?qWf^#`yu0$!mo@99HF0g+-CyCy_-Q)hWDZ8;7o+5oEQ#*NSQl; z8(Bqr;77(=foKH54^!q$-zy}!OF}#GI9af{K|PwMkKkIkwJrgzWiXuwTbS`^PCvr6No!_6AG0(fD_%8;@9 z(G4XSocKZU;{d6edBoh>79GaCvwJhqJ1^}baA9KRb6D&qohxv8wUc5u zB!Yr)UbK20LCf};-^71y+rcwUqpB6_x=w*N3Y_REYErbe7fsX0>rY_B;!-GE7B|z_ zE(&QHaVfj7Kt{I6ha}#{p?zT>K=>DE_&sO$eXalOS@3UeGXk!aA;>Mb#-gS1mQ22x zaim;Lqnsi?GX5zQfy@Ck-|drWu#*$fbmjaYE<-j`qZ$=MyT&@lRp3{j->fjbu@%pd zD+~f^=@ehgJNg>8=E62w6sTjugqaK^%5!`9TF1;MhkJb&U^0};v}TH|Oh$>0NG6|( zedsp%jJg?T6~xKvCFgKK$7oX{FNyobw1uR^2F59sba;F;4q`THKiUHMFF)%MQe`S7 z&M-vT2W}q>&?8l{3TrhWW?FdYfBP7}*nIg2Tr~*t&`s-F%`*?!zHy+Jo6oa^Mvj9m zH)oV;19D|HjKAlJ_V4tSm%N$^76EwDls57gn?W~=l`OQHOFOG<3na*~?S_i@2#_6@ zND%y9M0s)$?gZi_;L8>#lb^}k%E=aNxGMWv;8mn)gq9#HgVWf=a8z)L*u(EtRa826 zW}b$1;42qtDrHjh^2j*VLyU((HSh9@Pw`1e%};|J>RNzSCe>VEk(F!XPRq-oEl@xr zO_enHbIlX%e?cHhGo<*dj|ULa%p)JO?}rRi?y4XSN^^*;mUGO1INY(^$)>D*K|n9P z5kvmLoYpg#=@Z}Vl(>a5Bzq88WDJ4uOr1|ts{QRe;c+XwJ4vr?Ll~doU&Xu%0OiIcFNokQn^41&z}T;HH@edDC2!i6Onz z(GZRxY9Q_RNqRBLq>ZO<5^?Y2;b%4_9CBVBn3jau2fDCa;oVjq;>Y+9qhxnBH`B4e zKx!f0%X)8e5=m%Zpx=ok>YH{j;ORj0A$K&$St5-@dHCRH^KTcFRZPRxEwPi8!gFPZ zJ|Rc2-X1z^tk(kE^pvlSLNdEbAdz~~4Md}qL-?=jkALb*2)bT5QbX7pSLJg}YN_YR z*oAr;S7o7l)R3y?rRlnztUq2{67T~ChJd~pXcjvzkH}7qMoI`{RX&NF99O_=XpI0v z6cL%_Y;yQCh-fT>{VT89olfD%<>eUON3+lmCF5DTq_{;X^~L#=dZGzY%gq&^c$uw1#v8oZqAr1CO@W}1dTID+L70tC z1v*P8_9ykOz>2wxBCDdDIIk?%R!PI+C&$R1IHf&u-NSZEB8yfW?Mk-2*xTEvo177x-CKgg*80v$bU!CH@DmkF zS6^q9N*VtuJSI}K#}tkYG_MoJn^7!76ABOy<&6RqFKF_IlFF>8z)jCkxxUOd=tsCW zofny3b1Fu*qLqgpLE{^<(r8M#2{>0|Tu25!2A}b&^7~TdcnD?H*}oxIBM9i4tqdGh zm(D59B&jw+S&N7ZNmOMkf%#uF^vvB^sSt_z$#QC@BnEfeXF)(zOI&Vxlsp=huNWAp z-OSs``-t>qmB#bJAM|Ex+IRB%MA`#%Ne_*SR2@jx@P9OO07_hy8P zJ0#jQ^RdeJj)pX+>7-!^*PCAsgm{lR)7B<6qm8wGHVX0vaSyD1Y;^Oy)@h!#SQb8K zs5zZt@V8_%-SDlJhbRtlvoiFwH|amKpAYOLSmDjJmK6{5RC&e*Z#Pfdjz7*X$bowWUN)Ox5>*a1%c37%oYo;{8`pW$8R_-L%)} z!3za?1|!o5N+wV%B8c!*836=T*jC%|H=Z++SZsQQ*gYN0}S z)4{UGc+DoaQF$SS=zj8XQo^KSSp*S(6HD+l^#eIdV&zhr5yf8Hck1AfIP5xsAkPQ# ztSp084O0$MmWZnA<{y>^PCn*;0(MKJe$bEt#B}F9#?^VXw~oPfVN$8O4U**nei^-)JT-XE7i@>qMTV|;!d({1Ywu`JZR7)H7Hv*A7qiFq|Z$UKj{D@h-9hmOi zl6GYBT8>_p!aEpqwc~{y0wMb6%crK#Sh57Xv7BIa zDDdt$66|(b@Jxwxro-5rC1JmUFBcXwcqA(>Bpt>d#8^J6=`=GQQQ8)teGh}FG+Q9y zsW$YqEUnTDF`yfq|Zc^R%GVjv1rwyd-9nuaE#vdr_3RJUF= zucWSX;ZY<0a`Q>pz&WcaTsuPve31*6=R3T=jDFPubQP-gblM`%hQ80*QVoZ~s62uM zgB$pJs9HxY{uRKR|TjctF9Kwn2a^i456(hY=pdM z-%=ZYT1eV!uEWJ^y;1PmkM48(3P6Dxpy1$wE~{}z1Nc~;7cz07R0r1Z_y`<29F|x) za~3)>^0LQcaS;$$BWnMf6iJiAIf>d58X;j$DSW?H0`W0K4r_BhG_T-^4!29G!#l<- zlp!A-ExAGytJxOYE&A2Z)T`tQsk`lAE>}qXqK*_5=slkG0%=#_fUeJ>l{JD~-o+no z)LpJ@@uW?%nOX#SnRen+2jtF`-+EnTAor+}X0^F#0uv|`8~Z~9Qv$Btz*$@XF+&SC z{m6}g2Cd7cgn|3b580+bSLCQY@*^7#9*Oa0n>X3(GkeMhV_tl9G8z zU!ieTHLG8m?ObVia;_bhoIiTI51KsJ53$FKc{?;-S`w>>9=Tmn5$KrL1{NlC>%Y*K%PxW1k;tihjR}2=D z+oR^@lF<_%Uw_vu7Ex9(YIcg!e5JcxxCfgEAzU~}@9^M>vWnQVa1m&JFEm0Ps(7KT zj)k#BqIvT;;@8|{m)O;*vKnQNC_yLmb0Gqn%a!0P!^)H!dC9E(PK_>k((L5TE+p}~ z!c^dqT5F@Hw&Cm~E%ExCsD#8=+&L z)1OJaicjZX8h^gldR{J@pU%D6sSf=P7I7UW<4qXwnu-oBJY@}GjSY;JH@u2xF22=~ z?DEd&D)UGFQKhzbtdfz&W1eD%PLbB_iPD-mY%lJb)pO;3f;qWn+>@A4HqA|q+(mc( z`e>UdIV^Extn*uK?s4T&M#QgAumgJC;-t2gYiyCf^k0=lIJbGA*}5wph!51ruy#N4 z-=l=dlDzv$BO^}8{F7NAL0S#6RU`C?aXkt}_MB^qdz19bW1-`=IdM(KrYoY<%7*;1 zydz4%nN+Mg;x9dP|JYaoDxmK#+mrbL7@IUh;v;5sqNaL$gjJPUcnG*>2gbaM6pV_@ zpF?HXZE^%J#a@?jAQZk}2=FOctmwXNqV8m#=4x6MY>`Xmvq8!zkND&-gS0UFwWHoE zVhX!esMoCvR|+O9s~H!==koN8^&}<;ozl^4{3C@Q1Ohk5`V>?>-W zParNSI6r_Qi$>iw3G`dgL4^kHqTWRY(yrQt%|x5J25|*R1-&jYO(*9~rUhu4eaqyj zeZ|MR-!#=-j$+MhvRSA`vyk1A+eEAEBTd!ZKbrK33iBbk0+cqhI?5=WQnj3>0iGW6 zMUh0zag<*_2k;@r7uE+c$V_96ht@~T=Z7mLtwJn|k`T)%QL<%0(huRH1Gr({IW2Q?a9>Y<586q>EAgzIxi=mdW?-UJMi&X&WCVy|9Kr#fN4!iwp_CMa9 zj2w;z2@5fT2R6dSKhV6t9wJIm0Ud@mcerSyh@JSF%kPm4LC07logYkA8SO*wCFo8D z?{3wx-Rh#)V;JR{t0J{iJeU3aC;7er1BX?RK4WA#;qTML}3*d|MrWrWr4pKe{;er z;SV8WSbfpjM5+))mg|86RIB8=+nufh!@JNDX&2%d5|&q8g)chSVj`2h0k9Gu1UxV( zYLjS=(QUOr{s0*{H49fOM2MXtRsgJ6FtVOIqN(uQ?iR7KeebdTH5YRNZ!j6ZXg+<9 zIci(GQhW3%g978>t+&kxr(&H=twmemxjTfUEU#{E-A8U%upAaxWB<^yVxM-qh$s#_rdRk>eiTjCNrQPyeVb#He#-DLNBTYJ%?;P2`899G20Yr zZjUJJBa$nchJ;)cU zZ;izKD#X;6yQn`<5Usw7Q$=#uf;*>ft-h|W6MFATR&Vu3?=X7bKnOnUj_Y}N^5L_7 z1-?`dBYa(FLpWdTju@!n@@CRt6!Q<;{D{B@iJ-bC!9ghzFm+e|g2nj151CE0z=MdJ zN{7#|uiQxhAp_&hM^4yt-?PuaM$(r$vl4@oe{8G?z#_Wv2V0l1@6G1~-|9aq~u-^&i0S`3zJTeThlU2n)9D!W#KS4PiX`nFdzMrw1 zI9~#NFyfF`DrMBs5>K#0yxG7BvdE4R_RP~ygB1Yd z&D^yzY-Fs@1{0j;SUJEFAzJDJcycQ<8A~T$g7Hh>EgW1A)_^WswTAai)C~gCRKnoGF|TPm*5hXW0*z}brCzaXenjzBfiZljguiR zL4W~PqMm;oQOuO(eEMoKG$9|{bG$}E^=@Hru$9_lAgaSNKQtZNly%e53V933T31V1MGW(@~ zC#l!OqOi~*1(tk#7E88&Wy4~j#Tp}&_WeIiu%8(MY}ZM<+4Sf{DElxg)(`Qh#||O6 zgkdWDs)Ih3W2qRSRJ!4Q@nnby7&3F&I71aX$u9&Mv+&0d>@O5ez5@NW z?CD|qhhdn__qhB*7~k|nbZ4DlrS+=)%y6G#OQ z?!~*tgK!2Sb|f)JddahfHFU(eBix-#I~qyI{yM6#I+lhAyBU z-Oth#{$W@Jc=9%)g8}>D|APcbHNfR<74Z|MW&=U6b?F%A{Nq!CHDpEsw*Q1oD`UL? zferBBY#qcK2P-^hF`;4R_}D!x+DhF4UmkvYw}u}(BnV{9kEm1Sk+J?Fm;;lX4qS4%s^-^bd=az!MG@gv<$7%Q2_T_FCVuqy!0J zz}>L@`tADdlQCQ-2PY>IQh4IPN)6+;8%bl~>p2jWE9G*l{#bQa=L9ry`VlrF-(2)sm2CN$$P+8D&#mbV~>@7fzMz_(0ea$Bz8zOh*uZB_Y!}P z6;g_JFyx2@`-}ez<4@axQzURn2V7oE@jPLwd@vB6rTDjMJXqmj0&L$UuS+`dWWRt1 z=e@{(;$Ve`1P|!p4KDWe6SS=cs#;+C(cSoD2xhgin3azX$>fQZd?(ibWATr=h(#U~W(-sBmU;|gDLE7XCUT3YaRTdCm{ehgJUsa6f*s-l@*s2`9(cN8r*bJ8 z3^|a&{^I|__|taa6bY98>*9oy0=D;a+h%BC)$Wv^)BtQB-9GCM?grRI7fFUy4&m_XJlsko1{lNBJNnQA^i3Q|SkoSMMkEcCME5K#fplmCw zlVEZRe|M4rDWm)zUBi;{G>D(V`BSE#;S?24Io_v9`Csqdmw*Ep(bRxu66{ni^<4$@ zuh4jRLQ0C{Yv4&r*|y@zm`vXVa2Sl0j2&OF95UudgxG%%N(4gO2YW?^-P=p00h$cI z6t1Dh4mleEhGZrr?_sk&8>BFZ&~CQ<0*eUwgy6HOlmV3|KEdCqUAxOzrY>CuL%whA ze#1^hkr{}_R`5UiavI}L+kw+0_^*pY!vKUQTXRI>C02M&`AL2l2WLZW3I7C&u!%V* zyQvjS##J%oL?;zm=51#YsGpS%<1qmd-K-0#Z|6~Xq7;=Lbv9609;)g+m z=$5ayu!x|J2@71|hB#pT3cndlkoEJz>L8Y>eRsf+$%WN1tQ<#6fguZIm}Aku=64$7 zPuqdhB>1n3BY*_LBcE5~eH*J&r~D**;Dg$J3F|wD#pu(!($gE)Q+wM}Xmxsn{9nkE z!aMCgp7JpNyIc1=jlZY1{HH+tG|rzg1*fQR+VMsbq!5F4DN{oq^qs z_!X18&&MbX#Ih?7RM!7mD+g5)FQ`Gq(eg|%lPbuF-Vd-f^WLqftmRAd{xN3LpCz{p ziUL&>n7m!aLAl=GVET`8!ozB;=7abKh*gZ_C)S*VhNN@}4ZSv7^uklc9^P;KJ&b6h zk9)lbx`$abx|X&hmooj|xVwVdiWlb?-I>X_y`~SRl5~u{cmIFhItsbI|HXf~P4d{u zbAi|Rnoor5h#seb*(696n)A3sqmTYwh^YX${#%aUQ{(46r(_DIvY;GP97QZX#fmIA z1_7{XtED`(cmjd81T;ME<2HG@n3|pgy8W`BO-#pqg59m!I(qDfjh+Xc#_1B9A!<3- zoWJI1>)yG-w8(VAE`bh=-n`rKK5=BK<-dT)?z8!l+R3om!^w43rr-N*Z|m++Cm&*! zT3`o*)OBcYDeQZG-Y~%w>u6^D)cW~zv|i#>CzDe%`EtMo)%GT`l-j<3Owgt1SiHn~ zJnRoI^_{_sl00_T2ZbYU^F7O*rUBiTgHgNlacVEr9{(KFYZ!NII-eF;Ld*51A{S#% zXfVmcY~O%~vHGk9K2|YA$tWNY-cX&3wcf*yx5#jIXsr zv2%=tv-~T5K(>Uv=h>J(%HZc!_tO|vfwI!_X&-QC@$>)-a*8H*fucm%^&fu0KH7QG zz;wJ3yz}<8H=p;ivhmF%BTt(_<@+Tc&Mt#`pIe+6UJp`59E=9NyZ-T+voV!(@ygRy z*mBKGPJ)d3;7=(bcUpy`*(t09GLtqOS4F)tBF0iS>Mri162L%~{iB=@%H{Q~pOBb{aU$h`D|+Iq|*)=AYa zi3Oih3GJ@3_=y+#4(m#vPeS2tKo;LBcl`Q8d6T=4cr%&nF7u^EFP^><5#!hzk0958 z-tpOcKb?yy_v0^$yPrsX*jxe5>pN>cNzP6m|EN6p+`O;_bSxL6y5gW>ei>8Bvh0@e z;n$@;WjstBzrIzP>j03Bgb_pM?wX>~Ryv!v!#bUvD}}Gq&GcM{wcEDMA6?-eVY94P zypXi~N#|>f{5N7JyK49gWO1r-ab;ZkckinDzxbbd8A!{@ZHlSjKmR}U1ZQ?=E=n?8 z?ipI9&y?W<=SYY$E_(O&CD!gI8w^k$Qk4*9Py06?Icz}1 zF=HhcILB@-fEE`t>OhRYwOt(vk%@=gx6Y(0;3X~ z{fR$_M9o{Prc46D`>&b$z z^-^mr&D;uWRuC-c;I|f#uNFQj3e|=;d6gp(JHZV{2it=$p59L#0R`gaQ=Y2}XZlGV zTcSG?}TtBeN>o7_bb9-IkgT6TqBTYM~h zYI@BsI+2iZkn*ZcWz-cS4>iMn&)n3PT%T-@pwh9h)Mz3*t#4wEt6m$APJuKFY*JDX5nfE-Bu3~cCxtwT{XQX)#cH#doR_z;4pqQ=JI+A zaOL_QOY)9Ij@(^Q@4I)Juv zG1_=FBj)cOv~Ri3<_D@=EjbS%cS#OG*kZ zKQVn*KrwTL3#cazCvoE0zi3x-_^eh}A?t0S8zCKH3Et9gS=2)%ST^F%`wKEe z1ibkmWi&hx;tKomhrBehcwX6Ab2;V5q3Z_k`08Bk!i3G}Y-Ulj7u!9(&UHF4PmJjh zS>}+pp_NO+uLlUiWQCA6`&U~lU7;yzd+DGKezbfc(ZI6sj+gId1c|W;ODIe141Tx+ zyPtzm^QL%s547%Kcbx~332n+_44`AqmI(rv*noo(M<^Rut=Ba*pX6j@bM$;Gaw#Bt z^KiXhl5A4Sv3hUnk+V@Vrm9K*&E1dcTIDalC97vt`U~o}q{KmL z%0o~HeGM(Uc_U1?b~Hw<>m5jl!hkgmfm!||lyi$GM5joB(Fx@*7ygM#+5l~x>keZz zxN`Z7Xhp&*%O~16jX#T`!e`3xHGj?h#fWTbA)6%E?>DzL_2_-E@7nZ+)kPb}kc%T(< zEzuRmQw=XQ>NcMc*nw%{T0Y-PngZ>QB4(i7mcF4g38Xk#(@+EQN*<4EQ*yW>~h^Mxh7LX zqwUxm3$;7e4ZpX#`I2bd6pgOjGIEiKc_9Fp zK;!a3=byZ#+VeL;es!>r&TEG7{gQ9-Fr%3Ub-t-a&e4LKyRu|<`(9Z|R|xX+!-Hr7 z`vP3G(Kjqqstp@sLu(X+1GQ;2St56D%zmx?<8L>ZP(WCC>ki1Nt9JmV%AC{EB{M{fXGqP$nB-D^qj%LmZxm+QhXY(2{ zIkV^`N@6O-pe+qQL$IwAODdj}(4qqkHn#0Xn2s-JH7XeDUr&%(DtK}mqKNqRdi>(W z1Z7Y2m+o{}Z#p>@hrwi~EX>6vUM?-LT4~HC$QQ;+yGxz8u`b6D)Km6G=Dcb6nV9hE zu5OX(>$}}U?wvm-95O1#ET3Lii2+UAVNJo*yL7PT*Gu%gVomw))g7MSr`U}zAu8Nt zdt?jp@Xupnu$j=BGlV2OxIA`R^QaESzI$?!9rTCH*C#<)HQ{d!cuvaX5@>XzL~ZQl zw1!Zh!4$#2sgmV4-*@EUHJ+CXBtdW&#bu4g5?wqno$4*6_gAD+9`jd9T|HR}4L#mM zpZowC19vHV=#Fu-~+oR=2^uxs6(Y3_4P7u%gGUvhce0@ z-Hg6Gx(>itzq_28>iIxuNGNpnHhckT2HYr0nvUjoGe+ zqL>0r`=kBI0yV>vp;i6B1bYY9q6r`yo|tf=u7|q_D2c)45#ITNSY<>cJkN({C}k~1 zcR**U!~=7hgwySFfaBHyTg*4l6cm1Ze01IQ3f1#`%D@9LD%NHmcosxslLaJE8pS9e zVF87r#gQa6oY?)#iW)OP{%t$fh)h2{##v`tec(96PMJV=r~0|1dXUk|S`(1ccko*7o#&OD+Jbrxw2 zr{|5*?7^Up>}GC5K)Y-%XV7k`G0A$Ex#uqxBWyYnrn?(XTHMe9w2U<}QgODVgUBy@ zH8yxxciBI_nuX!12ESpQmq#^7j1zNx${^N@X}09;92f}Om0k&Gh8PU3=abz&mvc;g zpQ-Vj?mml!@v9{<9eBDT9s_3Eiiih))v`;m2Xx3mG+5;2R*M4J6Mo0;3fPMtyMJf zB*<4VyVEOOz4Kmw`?Xi-0ytlc#IqK_RN+^zeZ9M1&G{o`^2%*jQ&+q2ys?6M3 z8M23D$lqn*3(LMP*pQnX0>PLaOs0{8=p!n^@>{-)CD7*uFJ_(*n1wwpxt1SFG9H;}~Q zI$C)9$-%Uu(mJj){=GkZyyt--UAL>Q#hA6Wr_DQ-frH)WMLr7<>#{~TVDF^DZw@bBa^a&! zTA1&%Oz2tdNbOH#ObM1(SOdnQ1i zL(`W0vVyr)ikg-=%wWp(c;d=HY8}2i5#0!lLNHl+5pCQc-3D$*Nh*=Et>=rztpelT z8CPt02pIPj1z*rIE%uWg%8VPWVf8|jI)+!4Kj5-F$CvnH&_(Cl1pe)H3q@8unCo5S z{U(0iU5n<+D@`o!r4ioHI?*66e}FJoz~o+zRmp)YtuNWiz2r}{lEt^3ES~FfAK?bL zQZ*g#w_i4>fZuuiYuDa5FHS6I+gX;Y^`LA6*J1`g>`TBIN~T$8Mpj-3WB!KFLU2V~ z?(93P$pCZIk@rr!^@jU0$0g8Vja> zpg0cdCf~TicAbLQSglzc`)TQS9abiXy7iW{EGL;kr6_Qm3fR}3)xxB_jyj@PvnXqC znEnVS7M@dg- zXzpJg95zfVWs8TP0C_N%teT^OqqJgc2j_5uc(~;Ng z%8-=oek#Jh2K3cqiB+09o5z~X(S95eAQ-vRRlK4>vBVGfe!R7}a)rg0N?SB!UWTe? zaM@!VK#xZ$oSEPXTs8q#j$1yE+eL#CQlE%MnZsMF0B88YKSw2;^;(7;<Ehw(D(_yRoy8Yy*jAh*$KUYU(6p`xvgaSj+a z7)wMD22{hbk-}&XaGf@P&-FN^+_}FU>O1z_ITD^P>6kp#|0C(5H=eMd)+m zf-NYCNe5)Y%_Kn=9p0{o=IzEJ-#eYLQR2Vo=C6Q*ODw`M({V{Y@{2m-hp?;ASzYd+ zQZi}X%Eo>LrZhLuNA8;y=$=ZL-T0)Co|__Vd$O9Wm)8&Mbf_+c+p6J|f-VGzcjse> zY*dlH_y5@X%dn{4@BJSZ zq;n9dp+UMENo5F8BouLwkPhjRZWuSz!Be2J#Of3FNnhsQdTMhjYbZL7ceo$k9ANNJ? zeBJYZwqYQ$ns^j|G=$OjjYDR-J7wrE>JqxlhbD_&Okq>$$NJj41yi!B0q{?vnBj+X zeIg^Vahw-ZETqeAb$Ot6q6dGlfVHTuc|s?`(m_ZS!L9bbTQpfFggb->hQXqVu9;o3 z$reP^Ny0x{Qt?7#vA^trDYaY7XmF^z9<~x6K48yL3jBhtDe>=G(EmL(U)B!1hri#r zUaiSwyTKUx%0E%#tSDxOlJNmLs?OVex60PDAyyreI)blny3%>{D^_d>agP9I4%%|% zYSn~{#2cp#;8fDU5YU93kcs@Djr%Yd&xWqw>6WHXU>8erS}YiSBOcNBR6Xj7OyLN6 zDDWDFRYNL7`R2bN?EhU$w{n0MpCr5;IEQPQG1NE6zFTE5f1Hprl`b%@+WfvV4*{3)`V7dRyzk%f#QyziZ1fM_o+ZE>n^(1reT0TJYsm>PNmwWJNLikW0CVoWF zkXpZGv8tA2v8eN}08AGKxAeYVYFcThq-9?KIg^K!cpWFG#?|oX#YbIJIX81T8&1=aI2 zP_+EWfGx00BIi4rjaOAEm>MI%pEoY!L2H6gW@OIC64-ffUAr4T^H z@nwYnxWJ6MuQhb26w1Y^{*&ypAnArucZ6jG-?xFOY8Ios@_)Bn{pX}J62xE8h{y{Q z^1=-yEWk4Xvh+4bq2TU|B$aUYmCwgFw_ju%)QyF$g6Xy#$V;zvDkQeyH69f8QCMWW&yG*<8EmwBIn0MG|+ zY)2uY;b`xxT@jf-|Nrbh82Ch#fi?81SN{?Z3>ZFA+gn*huvm?XuBfvsenF?M4HDMD zCZFp8nBqD!{Cj2^*P<6)37S0j=h8q>nn@rs)f{tcM^bM~Dd>5d?OaW=Kupr-Wppc% zh-yf%n;}yRuV%+5({yb$S&9lTLja%r|JLAu&|dI2E-SwBT`Woe;fU=sy@&qqsI`62 z*xjG9Fj8}63`8_6ZTFK-7}$P7Jv1+6htWdKWN zB%o{6y5$OUeQ^wwZ}46ViGp(8L_f_YHjeq(sbZEtG8dxlJTWBB0GF4}^0@nX7b>`R zWr~2~ubR78AB|<5cGcUnm>=|vUVrkems&h^tNZ+a0_ju6%{xRh0Y%1V_5ZEK{@<$+ zs7}_QKq&w2Vya7$6^mb)RH&1>mT^_k?@Lw9H~txDrRS*4BnB>BtgpDpL$aKaiw}L0 z-I}C;cyr1brn`)7k)mcic< zKS}~a$?wzzm8GTi5i6Aa03w0_dV^S+D9cwCdA*Xa6j5^&$$bL=X)O>tUz_)bic_Xq zR|0VvQlq7IO3{V%BH8l=k%%`4kN^sL86$1*W&htDMgNJmvX}x&74;;E9z*@ndq=4o zfPE|?DF7jzf~GPSH%xy>MWE#=Y+hwnW-Ekxm`rBRRJQoPGT^Z*Zb^fF6j?R! zLP)-sKVNW}0(>zys$2G_G5tR(J$iX20sl;rug>9EUzgxh7?FM3<>n6p$xo&KwS@ia zfdC9zP~AR|(aNOhN`{sAzHA2D@by$!7=QUD7QJ$JqH!lGC3(@{+5yOAJ`vb&q*yO-|)e08xKUwO&dvj3g!5uD}`3&my(XK&<&*#DHO z|MjJE)V)6D5nDJTUk_hqw@BA^y3N-?x%%|O)aT>>0LgRRddO5B{azfgrLLtFmyYoi zK^wzG>Y#e5T41*~1?4xn&(`POF&q!zc%}e=v}L ztGa7qd!6cziEGi9Anivln{FJ0sx|6^6ZilWXLFZ63#zo|mx}=CcM`x$Z-yL26j9vC zTvWdEJ^lGw(%9jn+<%NH|Bdzwp5F^Q0Q^V}_!IoCCzB_wNMC^)fIj{q-2BVO z#}J~!l=@Ne`A0;ml>GXsAAu!Ryd_wR?quJq-5JVGnXuGG3yWj=L_%)^m$l zxx3T%cVc`wx)%dpKg$f0uzB^bnCzKtt>b>C`_}?d#b~MP;}~5s0LsvI0sz*`d3b#E z(2kUJ$2YL6ynAo&lOP0@7QB(IwZW1h+_%gh7o7TK$Q9iDXa9fB2K4KYN{@lk7Rg}7 z;e_i_7w7}2{B{r_|FgIHPD*+l{iLVO%QyRnXDBpu7+oH#cPp0`gBgqa@JHe2Ut@LA zkx{qt6tuebwDX9BH2#p;J#8I5ewd`>*{xDHdtg`a**DeLC; zygla!kn{YdCfVibn`W`9CxB}ck(q_7`@B0WU^mJuf}H8BFZ|oa`=8C{!53xqFPsF z4)N=O8W;esv+WU)u}oLu%<{LP<>VSNLg-L#4ndobJ`SYiygLiY|EI6$KOqZR>$14n zlxPPq)j+hC{rBE&#Cw#<_rEp6I$jzsn>s2l zG+2P$=e@OHYa&zJvjs^36i1V^khWZs%0|MkZFJ35)pxu>HW z+4z0$+kDO4E`a?j@xLnZf8GG>*&!JY7HOvdT%Y;6weLOUFb=J)AR<-EP^)9dXxmij z^q`y6vfrMFO6GDsp?^#+$tu(+G=>L;D?GXhKv;7u$>`+A(XReL@Pj9%TfXOjiEo`o zJJkr2BN5F@B}UHZeMq_2Mn}D>KH_eJW>-^ol$F;!Z!U3cXJ+rXlg)CGtS;(m?E~99 zVoyNxiL#gvddM}*cB51nJRda^ATowHiQ1~s@?I4_!1*TL27)>z%Bp7q@9H=a5@kfw z4buzDB%7v zO2?|2mOPz)Jv9%wVNgpqJsfgz*>|X|Et>ukPi}YMl>hKNz(G`DhQ=~8PF?=y#5OCm zh|+CWSg|#r1Ue^yIB({ z21xmVSudDwOzj@2?Yt81BM zo20k(h1@4yIc45NgmfU$$ol$4dNwj+ltAI$GE-6qkV+R>oX}QAI256XfHGp~7 z@$`%2f=w)efgBy@iC*1khll|QvpPaLh0fDk-yTCqcwR5$iw(6=nb;xFD4*|82q*>i0ZR^nAywcQ3^z<5KPgTv>i%D zHzaVmM|Zn~?=QMS3VNTC(OF#lp?^$-5rNN5jT2R%s+Nhyx|37{ghJi6i&Baycy!Eo z*szGQ88OPh7dUi4K-!c-#2m!D3it_bE(+;>-t5*eV9ubug(rceK1_#q2Sab$gi{Vn z38AUSudC4$SEkUiWBZNTkFN7}T_YUhh5J=ZW_?m3#1+7{e==^~QjmCLqwfvUf;;vyz8*s30mB;v_)-G>o<^f|* z;2hbL4ynDmsT!qYzw5n}3L~;y3P7U&S-=LUW)loeK<+;U_#P7+At%VYTjW~9i>!EG zg>eOlQBSu^_Hkz2tf6<5SxGbI%*B|g7rq5mBbsiB{tqYBF}}7aBkp26J+f%YRk;2o za8fV2bVtJ?fhx!!c?>xm=#&qd|8R%AyJ>d?B2^3KP2}x9@??8b`7km(p!b?CA|nTw z2%@_XscQyev(gttD33gN(cg=96QsJqkoPwcSUrR&Dim-qyH)Ujfg{-e0-gw6FgAZl zYK>)q-=z=)V{6Bds;KH&mFRH5=}cI|L+OCxKt9__eD@VNgE+Rj!MkV)+T?Z9nCm25 z?funciZg3!9zJ>+&g3p|?xFlLfae#s>3iI4MzzV)!x6)^J-Oij9zXIIt1j^peH%p(vQ>B z1BQo9W>3d; zPfBXucW7By719~R+0M6gjYl&g2q6XE!l~_b>f1hK?-f|?*(Q*sE8uj0AB7zyRwm$C7yB1{V!u;Z~WZaKhIctqqMj7 zly0jCfhQ^1)tC2}1*^N}7I>V(*LTLsZC@A*EJi;2tf1v*cM&pv zF&A#rxHGSNI|UzmiH`x9o4%7`qe%_s$k|~+*pvF2j;dGe5gn?#6mRw8vMTLo4}x`Z zu*>87j0*KyoAvByBTJ6?0i*f9JRU&V=}KH@VVc5G1>p16zr>}zy5IJKRQhk#FLNtX z>~islGeDJlv#-=!!;RVY$SPCu0@#n@{Y6iPzShsYi?ag|9MiG)q9Q+H^V&0VEI}tg z)ZE{IaU`3%eIPDs7RCD%gfD0C;j=BcRnwghT)qTUK7yqmGOwlrYqfFj9JlSFQ9PQF zW+%i{1>9dz?&Cp%I(DoS{krDYB!?6YIR`40;-sH4cCrU^J_Uq2Fwq%_Bu4ki=YzT{ zjC(*WA388$#!x(b{g~^hKH~$WXMbRGqS1Be5l=EH*nXe>+;{=D^Gtf@x5(Nox6Jmn z4sY|?4lQxlUrF9mHY=rJl<1b7EZ?;T2q+Bg<2kgZpdrSAim2P z!TJ?FF#Wd$r6aT)c7LPuPB_^em?d9!bf;Tus*~K0%+7=T++uu$5wUERNM@U6O(t(N ziaBG0+P{qILXTG7#s#|N2L{ z6Cz*lx9hH)N4E=OZmszzEK8`p!~ROsc5*R#?7Jcv?Lhhaa(UH1)jNw{V(GICE{il0 zD%m3u7_E1klKLjWfpD3^ue}yO@iCIwj|!#Xq7Iumx?W*`npnL3N{%GCIJhmpm3NUF zio1IYO>So!%FA-lJ@vfsFPRE*Ns(DK^r$=_$>z7&k2#la8TQ_2aqLZD=#KN$K%M*F zxqn=eGvL^7Sw;+BXWvzDzG&1+M)^-Dr|mT8m?ORerCCzUL;4Q--r2UZdfT6)S%>YT zUCrSEm2V=gt4^|a=9Ap%PUMN1U`Ze=6lWipF`*h(@ao+IPir5?zNvmXml3GxW=(Zp zm?@j3b1m@as4%0#7oiYpnrpz1nsApwtG>{02xKTRVm-OV4Zo~LqVL9z8LUKC`@jkv zP{;!H^?duyJU%kaL*u~S%Ezh?oZkU0==y!q5;fPtp4l+*hjuM%O5F-BZ5oyKtS{yc z=o)!EQ`({~m1t=<(kMBTDrWchJ&^wOgH1R3V?KGa*Vv3R9~g4p z2bK%&14mc($8aGVgz7JTq0fwJ*ssLl8UuTB{u#3(V_&f|#=Aqt%k-$Zl|FESL!?+{ zqr_aTh_0i(Xh$E|WE0wbc?RvCEZIIr4@)yecRXxPTA6JuFAweKPQV9wo9_A!>Z_H5 z?I)Wq7bPrOX>8ycO+=UUolO%75;IjmaF0jw+7Py^Q}9%7CiSspx$&svhJ#Ix?-`Aa zR%Jyvx6)s0T2 zfq?@we`EQB>GvI>wsT_Kzox{;H#_zs@4|Fxz;fS?S|u_Yw;Gwl6EjM(?D5DCprhiB`$y-4%?@7?q=$H?!L7W9F? zvb}3-r|s&3c8B|;sK5R^C`q5w&PiFuKYiYYkIr>DP<_EPPLOqKZy@xOo;ipU-e}bs z8@heCC0Gqx|3XfzvNB)R78I51rSK!+cw@Z zhtb8;h=J|bkTUzh4vMDnLDj&8(wh<=T6Q@_smE25`u>4j`hUCfwf^WYLylzJFF+vq z?~o~{s1@Q}_tss5OWD5DlB=&NvjJio&$0rSwN2q;A>JDcC=IPXAUa@NTZFU*`h1-| z=uFM2m$4KZJ-;tUT!uik{{$&<7zDSVWe?LRnyKUFHgaU}j&CD?Npa12V!F9n4bv1u z__r`?Zk;NZJh^4i%%R=P#nxFPoZuFflkpS}Q0|qHJ}fE;X~FAf^!M3MvP|E}(~(*W zjA9dfgasBjv(_~-4xK7RZ5T$ws1xa>j^&p@zI`Rty$Ucx8(#e?mOR`PE~n+Ab7rLYUTwb8s-*Be6VpyOn%YHIZ0{qJdiWp;{~hrNe(+0(0>H9 zm{1sn__ujvQ8*K@)Bxt72Lcp&rIa5&V~6vFCL1=Mu;eo_<;EU+=TI%LUHwRFJ~*p(=6%9 zjOpX9@(;{LZt^D(`Dp{sjU=GDE>9Uj=z{~}oK{*BTYJ`02}J63a(~XhECba}nlTWI zl!bk-Fnr0r+`~xSr%C>>_bO5>dWPn=*zi>zV?*9DT3Vb2rbTObRd{{_ElHIJQj!g5 zK(7T~(9bZPM4M^}CCU}Db))={=Y9gm_&QWmn8_q3n=1E3FOA-9Kp|A7~YY3pQP?@Z;-OoNltf+c915^AE-Xt=zqb~GfI$k5Mq!QnAg+YGp=8G^dJCtA1q(9fc z3Um&1kHE*0jM3#bp7C!m>hr!Y!K=j8P`v~G?!Or9ei&0=IPALDlhN`+0^i2P%E0FRWg3F-@z4$Iz^)T?)RcMzERzx}Q#1haVO%_pj8gSImV+V}BV% z2du`mlm&?}-+Q8g_6)&}p^{I`TknMO0)Nz@|StPP9=o~(t;4Cmv_nBjC)ufLy1HboaM5o#x z1w1+lKg!k>+}Vm!`ft4TW21EtH6lB1+?VO8zHPPgiH#NwLZ~m{^L&2z!YVcHy%Dhl zm}^84;+H=G@d`74(O1^?a_{tTO=&5JFo=SMm$-=$)yX&_1GyiL>Wmf4Oq_zUI*bpj z%W%t2Vex`5@bU|(qAbB1a1#BTI-u3%Kut8jx+gUBX))Dh49>%`*%)aP-l*!TzM=C- zc-!<~B>$cS7ho7pyeC9rU-% z_G!lxk6b#XNEOlIqnFt&CzCH;(=qv!X1Ly17hPTx}=6xk)r8D_we zeRGSv^9w)ObFw`qnY@QsG-+hpy6w?GDfR{CS)AGnO#`7OM>|}~hp;q(g_}i`xO%M> z|HT-|1YTYU>gCdAYbS=_tnw2DnN?^H>re-fH2l0|`0=76NV=twlt}gd&QBs{C2hp~ zQ_p^hWDlmJI2^b@A*%u@{j)^to(vCWOd-@Y&}C`!5$C|CYQ?+z@&-vh6!BM7#=b@8S zm%acul^#(6S_3UpVFEvI1}hB2CY0Jb2#s)Xr|MD~MKhYHkj$IH^+kJ}jitda^njqV zy_JIEkWRQmQxt|HbdtwNUHpJzg%76N8)h-BSybw{<=AfZ&|XNC$>RBZxU$CGTnS1iCeoe9szY#g{$IB zfPRHPcjYG1hbd6@H=daV2mKJY1TXs`N2;0r0(f zB-k?v1)1wQ=m6n-`(lfK=l)pV?vL(c0jvZ*m}>4e#fe-SvDdUMh) ze+I=KWsJ8q6q0W6BO!3cH2VXD=NULZE(sEaP;6V#8O#GxTOjj z*1^-%{UIomyi8O=@>I0l+fH}zapB?^YGmw@4U9@eCdP&&6ol)7=L3H=$fkHf_iFU~ zAw%ZdOO}u$-iQx>xVn{TxTUv$C(aqiHBe;f(NMuMg%xvLAN|P0we&kSJRlAxGz zDtRi1e1ftH824B=6xBJ<4e`K^$m@74`t88?p39GdPRRYjm)_9@2nvxu8$zPC2Ye9L zI@zzSxK5u8QR^}CG(D_fzXTtniC zzG`ZDnHC4pi^2Dzn1_qKH{D!L;w)P~K?OoFNU#?G%ix0;%?P+W zSL|2gxVpV&LPk&c3S8qFf8FlLhNxl~r62=(Z~18wul2%;U^hiZ0G3kpOHlipCP}Yj zv`n-NghA!urZ4LltXK0Yo~50|PqUBo(_0yiNWV1+Kh{d6Ta{A!2L**L###cqLcA4khPZ&INVi;@Up$igsgN(6lHdOi}@mk$r| znB#0%pG`b=gt#)qF>+vMbCth5rkjjA1{JN$kS1ms(=<>}BIc}=a_ObM%EpLB<{e%d zl&E)zj0d7;rs$;Hvm@RKN))TmA*0OHmC#${$e?qVDA4UJvj)wtKiF&a`vaoetnru0V@7;DG(Qe^Ra#MW$&VGpp(;eZFSN`miTS?1wbr3PQNI#q=WdpQ zT{9(yRDO<>z@&?sQC)rG?F3eW;vcl&1&cQR>8y)(!fv0 zg93#&zQ(E|l}|DURD=?_mucx5$&vrOUM7uh;agCO`pdd8av_X~lJb*Mo8Ixpr-(fN z{kr8uK0Fe0U!}RhRtxMQ&}1tXc8KZG`}d`#yG$}#u)}?$r!Z4DUNcUr>VP8Np6^5v zS=JkZ3jVEnqR%}@`>+(Hh7xvM+2n5A%lPI!+&$Wac>5OoM}z=D-RL3YPyBYB%jH?m z!FMx-G`d^R>*u#I2snkh(Jo8tMK-|^Ge72wM3rBI7~ETzY04wl+UT_Zw5LA zc;zvq1m7KoyJv((C%FNdF!`$_0@7B)B#aL)A@_#EbcqA;Q`{P`9#uWUDy`a6=}=%w zWs5)X_mbw4|MRj4L+&%PT2UGT?<9Q6Cg^lflW7S$sZA=I%ylyU%Hs5XH|L*D%hG$Z zMM9y^gm@fp4y2P5a3GA->OYaANJ3~&7!LX7P|v3!A!6d5G9d~x-rB0Ou5=heEK`J? zX=WVn>BB$~VmYjewiWTscXW`sB@$nS3%!s4B`GG%2|Sjiks)PM-{b%#Q}RA@HkE=* zyK*INN4v>LB*m<9Pfw5)CHdEv_o)!!IMcBKO1Eo`CJ%4@zIDAF*NzAcr?LU>k9=2d zp)L+Th0va#K_@BW8Yh#DA$aak#R(%qM#Dj3n{ri%$M4;!#bSi&ZAmvl1ZfDKV(mXK zN!15q=pt{K+Ytp)ILhV-_cEpr1;wy@b;ZPMH1|n~4Jk6CX?kDhRb_NuWDf_5z6d10 zCz~HhjJwapZS;x4Fa1^Q+EPg<1^)(lS9jsPU~t>e_3-#y_1IW*Z$m&Mi-}fP1rig< zw@l6Z@xuS2WynLwb&m;R!8AcscagOd4s{J3zN`^{KbyA;KQt=@C#?V@?}_g$8VQkoAW9yPQ-#QJ60w4 zV-EVHr|y)`EA`qv<=SNN95OYsX-r=CqY)R$IrrSyq&|3*%wzAqxs@nutwCvcm=U?H zGJKYKm|v3~J&|v@3HKMtfSSd+Du0>kW*xtPVDcz2g?9nXJk%76t{32_#LP5auUJbt z7pT441SCDW>@h>wx@6{F*dO{IHP|H)lDfSUo}$xJgHC!gAP0-0@TFwe+$fiKnPbXp zoLTkXX3fvOQA;JCcLxe@tyuZ#z{4G>D2)xYho&C$)tVtnzq#=;5;yPZ;uUD6k_0$O z(nd{Z3i0%myQ(onjr)AG_>ulFezIcy{oTVGSF;|v!0>9)zAM?HpL^P-w6-cgmPKzIIdjK0iH^NCxWvCUP_pIL4v9Iw zfls~4ncXrd9RRuV3Yxat-3~f zQ`8YA$Fh)5Lj<#-0p8K}LWU0`{e1&htHOjmkl4qCbnj3%HTBYtLXwvGbtenaLf(v1 z`%t&}v*zBL)Rw@(2aU?IxXV};zk~UZ66PCzzxF-gG}2v|*4~r~ZM2D(g$_`Zx@>}Q zfatp3P%}f9h2L>;)OU+GzvNjUZ=5DClCRn#xo z*XqP{{>=i=PjO&RU{Slac!i_Fp(^E5oCRrPUs}gq!Ut>vRo}7lkHtCQRyhIg8CSds zFJG4FecLMU`KV1zbRUbFT%FY6(JI*swFz0rx@N#U@`e>eF>at|Wc;S53U?^9TF!|& zy65mj+|({OX;QNz2#ZmXnR@+D$mJCYYY#yY@yV?zSwIq{>=lt#s#*Q(TZw~+pkPFG z>s|#m_ibAhxP!ZlxB{E9b@jq6?=52)PkEc)c8VVtU)$B8L#5Wdj5H*y3fiHNkIFn5 zUKVPn`4Z?h{)pXe%-Tk+x`|4MTr%f<2q`bGvDAPkDO!8O9!_Ux$U(caQ8sv59!!5yo24nHw4naL%&7593UpM=kAjX)fscRKM#m z2{6`Oh>_{i=q}GQfy{n57b;9s6&{dy-^-8ET& z$7`T^g*yUrbY-}zyJckO_8wYC=HWmi`$YH0UG+g)|G`Mh1^;awC6q|3|MML+;oG*T zo(u;^(E;VGU^f2l!^fnAPs%y3yX8&4&F|skR84k2z}GQV&{dCbzf?ct8?4G+la6Ae z=r%4P&ap#=xzV)IeL8)pV#|j_fAcM5!+WiMoj<)l&7Hvm|CLyzcC}YX1DA6 zF?m8h*YTz+-Lr|l2v}Db^Cwan`aE~NHDLoTzL}$aPWpH%eahv*t^4;H!>$^b;f>Wv zNvM@BtOQ_l-$ti5Rgoz z9-UK`TSG@a%Bb@<31OUSpeym#ytTozCMrywca26x-QKM!PyBj+3t*kqfgbU3F%Hx= za{&Xa2$jqA^Icp%%yl)>ch^ih3&N7j=OWG7HpD5V5{B6JpFSRF4y^|xb%^fN%kmc6 z^z_S?Dh>iu$HFPdXOhsqpvfbyo^qdakFUW571%m>Py?be4W4EYb%`=Y#0n@+nDNjT zFVM314X?Q}Q7^KpaD4%EGC5SrXIlM(Adjo*n4b83V8JnXW)eE6`tJIB>`B9#l z{Kj06N@Srh7go7Uw4uI_(@EBddbs>KZw{*@W1p%-GKQTMUsct^4ih5Fe(2soc@OD} zaORTJJ-1&LeVm!A&J3+Tp?N}O^oPBOzrU~<1)=LJ{U0KGwd5I2X_r}K$SE-=pwHVS zOeRFC@jLYX-G1Obo(`p4)s1Nh11@S$(_md%<$V4Y(g6MM7J| zKm~=(8JOF$#T!QlG3xWv?GWe1|$z#CIrXqyeJN@g*WGl z4chPICoya&vMcvPb1IZVYt~D>U)D_0dHWwxQqo;4B511o@-35ns+X&%2ziGXQFry` zbDA>G`rEK6DW$HfJ6IeL3o;>NC)n^e4`(7;#+paV)*X&^>rk*z(#xcyiu^YV2fADTS(xvxn-Kn8V30Zh+T{Hli&3FQ&g?ViWG(SiM7F?{U=O;`BQsE&izL z-(b_g)8tewI>uKAecX1FyiE{m9pK%Et_vLL(q8EFYfhc)4oKLQW9zs|6`F&cRu?-} zC>DMPBw1aCcdw=$#W|eVx2IKY-R$TU*PlBB^C_Ujn30JcTYTm!eAA}YbcPayyjhJJ z0}0w^r83$$rDDa4Q``dN{9E?GJ2i-Rz#vc%5W05yhi|ZkgIE2Vm83{WcmTOsrtg|) zvM^Vda zxPk>kKp2D^`*?U*Jxz=J*L($JR*r3IO811wgy1;cGWW-bXz9+o2~{#2bJ#Yu>AG-r zvQ-Ro5xq?mAR5Onq#9puzPxSyFztru>y7rdsbRS+0jGc)@{5=I416+=DuNfFq9`|J}|=>qVmlr*Gua1*8B>L)cim4aOu zQTkqI)eBE~(5Ys!PXKg6j3l67R|}5?3@9SyiD@U((=f?m zEOZBB9R^8-dN{|Lfq9pc99kP>tIw1|?-KkQ&v*i^ud!e7GdB0kg1Cb z>dE98q2oqeqWRo_Q$BD~!#hUkb$!R`5S$(QSraEd=3w_5TyvG^_8aJ+I$%u}A>01| z*WU|TYho_;(Bt+jqo+AERc6@P?$5g$-^A97mtA0ToC}YQ`B$sZy9=coe&rqviuVegC)K|hI; zIZ-ivDiAfp1bb zzcH2rZxp+(W!<^46%lc!6luw6MLh$lqMa8~N9!F%b^=gl-Af8E0yG}r-jjH=K7ps%WO;;8DEPp+?6taka zp|VhAOmNkTHALI9oyV+409|JJ&WS;>BVV`+ghr z#3+{|(>)me7tHYX)kNk$ZEUPV zkLhTqf=k54?WG9e zQ})qF`PbJO%0#yVR@1R0K7>{mugIu*615A+Y_i+f@RvzL)XEsOg-EFh=) z<$w%VfwS&dQN|%cRF1Md- zF51d%yApkKn?PAeBdAJBTs=5dZ!Ybm*YyHNf{T7C*go!#(A{- zt3&N`!Vu*VW{@&*t2z60YOKk;1KGxdRZ8Rw$vBNeJR!)a!dILlJJ23%91Z-ZN=m}8G0E(>kV2o19v#TA>|y+{ieJFI9}lB)O9&>e0F zcVIPPTZW^tiYm8Eqg8^eW6aLRcG9Em^tLnvhTXWA>}upy2XCLzz;mAOYsqo+7n7Nj zBbDc?Uu~}gqgSBxn68ls@!51B?9&%qqQ%j9fR)YMgz}`0ju}?#D5LQG@9bEiDw1<^2)82Xm za!X2ZDs)sB%giv);T}m9nU>WvlR$4Y;+R@1=~kS7PuwXt@VSqbr0M9-tIuB~C+F3J zYrN*kN`|=e{bfTyne(i?Xri(wNt*`yrzaQg%zKD25vJTkx2Jc%3xy!W)fWW_QQ&HH zcdn8e8vS%H>SrAqJngY4v*VZrAyJmR7LIi<7d}ZKpZMc(jE%@G#eKh%UOJ3!y86r# zW!l&Ka1yjrO30+nWSb0~mD6i}nf++oXe+DhIWZL$Q|B_nrc%N}+JV~tVO4lf z!%S5{p1N=JIZo$szdx$q1@EJ7GRa{@4fa!bO%dv-{_PfZ8|y`gH)@SBXU6Md3^GzF z9JDrANwlj(6qWX;My;d9ZJ-he;V&Cm9auS!za}c}XzI_YFzu1$KlU5=jcdbS9t!7~ zDLQzfUm%n7(O}4EKgQ$r`4H{?(O3;WR59l~<~-dwrUvE!`VouINAA~6MF>S@YRPrR z{z!gK8*L4_9OP!sMtWU-*?FIVWb0ywv~$^5;6r^#9a#e$oBb9o6DB zbMV>#=e){oCnM^q`W-Drz@D!?x@`J1AYTA1U;w3cGhUnsg~v{*h9R5z!DZ46Eakb* z?{-V6KiY=EF;H?F5ZP>Zk*XP{*o0Y43RTF=M&o!=b-kj};n&?) z!fqps$mK7OcK&|!amQ(&oZ@mH;C$@l8)q+cIe%#etoJVI7vc=Y8{aPqzX~Bc=%*M+X^9n6*&Lq3#D1p z5cl(gi|JqgZU{8({_ILs8|*eQFlasQY5yyBpLyKXC!))Xdx`_Al^@KIrfz%~_nJMS zvP-|E)z&@?8gI?WVrD1}>)NP8PH^rkF#TmXPVIM{2-t_~5y`8bt%q5Su)jXCgTvGM zWK889Z$NLGK~~=jqNP4ax5JE*r@k`+tptPE(aeJER7bj^EAiONQ`Btfc#^74fdsyTrlpQBko?x_ zR|#^C>+X1%^Oy0?p+2e%Ou+nWsEZF?G`-L74)2FL&IY0|k;)K{xc&pf2Vc@>iiNH( z4DJeea~Dj~Z5x##O%cqo^t_OeJ)jJ`Jay1Jz2r4JzupT9`fgL%DxM<^J~Duw6B}}r zXG$&gRRMLAdA@T1@a_b4+IsnrVqV69G9j7A z!`J9`@?ppl<3-##m)@@d|L&{s2iS<0TlVja#d*wIB=Jk$KqxR&0F=> z@6HxY{Gmn`V0tIS#nZ<$B7K|E5QbIRt*eQrHKTzSnj|53$w;0pMdgC(7N<6 z(GiWg^z!?mV30_H$Z|L}X5wX}H`#@_rt<6JxE$DTaH0qr$yyuvXe?x`Vq{7NciFv7 zLO<9>$4f^Jp+6anyzkiokQvG;G1~f#h0-^SUEi6l&wF7kzZ#mo@sm%93&)&^)XwXO zVK|Bo|K(j^th50O2PMU6N;f-l&rhTlGNZ?g>V(}x9@G#(H47DJA*D4BXj_+c&km~d zg&@IMIJim6x`BPv3Qvf(hMjro20w}}bUk3=nZ|o0`q8Semq?dBA1hGV`oeDT`h*9C z@-GkQt!6Wsh=n7-b9i|RkxO|ka#8yPGCWaitF+y@&Jr6uheZgOW&lVfQ&qkv^8h=U z(TRj)H|XJ`U9W2c7siW;4$rBKr9_qovxSFGfx_)WCvFVFr|`Z8pc*oOUj~s1@tmiug?k?N?;od~L+g&Bb%{R}--6zATWV+z)~xcL9r zd&{`0n(hr)5S5aWmPSS3C?L`)2&f?4(g+AhcOF1m1f)emkdOwEPN}2R0i>i;q`Tw1 zvsG@npXdMn-f!=x=Zh%ry=TpuHEY(anRQ(YmY|*jue%U^m2UT5{ggILKU%Q<@P0GV z4d24CwyjKWij(FAq6*6kuqGzTa!lf0F5?U2^7`o!kKDiL0y4ASo5f#-IT8wSq!EHY z_ol;;{8SB+LDMs8tuV^nG=X_>(RQnEA~0jccF{QT1cp4qM1rE5nE~kgR|#4D-O|fN zDQLxUHnpFH&5erf-9Hkgo35GEW=M1?BT!iwWwD8yUK|zElekGqf^R+RHgM{agO_CN$n?g3fghrN=nIK~t7(BHq;| zl`#tp_pwIL32Mpe8y$qvUzws7qdvQfHw^K%Ns=>SGYWR6NK%MtwD1_mdr9czS@G&>l3Cx-LS*>G(pd43G+NdyJAHR)4v-d?Q(8(MKX0+LFq73Jbrk{9j< zIN)A3WpKs4Zr0{SKqT25KfNI6|HF@$bD{d#QYH%5KfiB3_dS#2a9QgIoaTFitg@vP zQ+`&Y{ufIJa+g0eSxTe}2E^S@cQkuL>T@3tKiAGCq8%n3DR%cuY3jWa658Hnx9P#S z6h3+aDjUkWb@2vkoFh68L)FpDI%D20sP0ZUiJbD=!9)J3)Wf6CPF@rTN%yAz=0|n3Tir-0L*BNh_{$;4P@)t#5;C;qI`<3sCIt`a6xh z?|G#kx3@f1V;&&!dEz84=wre+JHbYKXZRO^Nc1_s`tVx41sdSZOEZ3qKOA`bN#uSL z75}4z{U6hA{`yi5$dh>ba^4AJUZVZD6yfeJs;y*yk)o6Rs$*(Z@?mk0_Y7~vUz_y{ z^(1bXE1taFY8OAOki4AC5L#^)i7uYJ$k#As%uZV}7m-OGtg?N}pW5#DVPo--9m4Vo z++UY-#^)nJvKzhpb42k}8~^>L1c}f)FLz(07?X(ce))K@&cQ)eU5R?~QemjYK$Jz) z3wmzLWGSs19!?*g+_@X$v}nZU_8s6>4bSy?n1fs^?qeEPE7sfexoO|5HZF??WX>A- z3r)daJ0)zSdF(E_8V*#S=Wx|Oq*&`~sp4mq_wTn;yse62?RqC-Mz7q@ub1>uIS7D_jTPpLfTF zZ}7|7mg{O;J-dwGN)8Ua1K|0Uj_?jQi8rE2Q*W=3YKX_$Cxc_3G>p9)iZ`Z^MPsW3 z1*K)o#cz7OT@NZRf_(oh0>kwtxvbNIZuXKbzFV4dM=9iMoM!|7*7LITGLR=_Um<1 zEUq7(>?u;=<%ah}X>N1`i3q)z(dY>ll&KWqf)%n@3-&YO>R%SoG}RYNXi!t2zT@y} zUr^O{MmiU}NLfD)Z+H`ry+QHyT=P(S-;qmv4Tm$2qf*@<&~+e$o_uH{a={JssSJI&R(4 zFHmq@b6KQZ#0Y($MJz(cy_FMvU^X#*B*onrQ@`z9P3NG+%4ch%o>+6vyKzR(oVDB@ z`>z{(zJKyVgTnQTe1oa?>WEt-16F4mVN8O0+GP#mvVqAbVd*6OiSCPEH&sgQ;Db7p zF)7yHAJdH%w)%4IU+SFUsx^!XQg_pT_&vq7 zj|H7P=3Fkey5;T%zJSiDJOM=XUVqcmJEktO8Y+!rtUFPUL-1a9{&XKVM^)yL6gs|6 zb+`XOr#kUM((fe~tuhN<_xo-?E>X|$1>S*Q*W4zK({0sWiH;eu*ySF2x1b|bCwM7; z6gTJ~rz?-`zVqA)J%X!F#KX4Rm)Eu4`pr=qbkVeN3zn0nbiQp)6Ef*{X)2oHCm!do z>o=V4Sa-W@kx-VtGz@N8Y!c1Ao*UZI(-qko z&ncQ$$_;St6++C+db%GV$4=^ca9fV22YF(I<{E<8iQzkT1)AmgzlR}88ds|2@kQ#- z<((hu54BbWJS8X=WTl4qFI~vhPP;9nv(^8aqag>L6eNLh$U5;fBI&Mc==%35M9P=) z{{~BVpofbFV=3B`C~x2ph$;51-6lC9dh^V%n~|`k$)y`{Q*Swf9s8#7Tzha6>>-A$ z!JzH}(_x$qhBT~w{EdUTCXy2IU+_=>aAWW}dU>v1&xIyd<1Qmi+YA2AoYimY?-Swb zn~)#GrL;J_Qqf`6Tjtn~E5WMcYnT5LqCrA#sB?0D5RuC@8O~$E+>ne;(fJ3V48*hl zz)E;w(0Sd%B~{&f)C8v$M_>Oc^-O}*bNVeYTPa^gwP#5}Tu-VucLx zmV38DZJ7MxJ2$c>aft(7g{b`>Xff9c-(Zk8`#Jl?SOYYJxiltqJ(Xw2T7ANw{Nk{9 zn3TTXjg`PwkH*<4yd6zU&zd3Esex85qf2i@%jv=Xvh9sY(wjGzP1b_d!=Wq9r&tt! z{>mM8j`PzUfm@crW~FN}V&mf^oKe$uS95Y6Q;nv3_5BDHvgEuNon`*!r z0Ofl4Xsb@}pkTgak8r3pKZ}|8rvcTldI$l5>c$uOdsHQ$Km_ID z1o$SX*6BQ`Cn4C3-Bhxg4*~!Nq4Is0@Ijy<&wK#LF)JO8pDd3nav9~a)D{iN1eTEC zs6HV9<+IiHJD#5}kl6#W5`#`kYXGg%SEmimxBsJG*9xaFY5{Ob1{{E7i~*!Cs8nLy z0fLVRbjnp9VQx*@ zi`-6Q63RNvUp>g*igoCD$W#(g zaRwv!baK|C5!I(I0PGwY#B(1AqNx(V__F2Yoh%_wmIBprFO$ri8#=(f;@+y11jt96 z;7(Ek$%dC3A(Yq(*mE{?l4)JSA{qeeumV+XnBmD&gcCA&6{1 z2=5dzKc}`qgP&^uR07~0*m=;QQp0a51dMsL`yEz}zc>TbC%He+`r|6|`tJz7&n3uJ ze>QCR!5|KT)aHKgM}5y91!$VOjsdomhZ8{Aa7YGtBX*P>NRJgeffc@}{qIg7d+M~; z-cWQ^jI@F5W#eThI{V*zIcH_iyNh(dzull8U56N4K02d<-+!%Qs{fD$L3ND+jvV6yF_$q4vTO9ui0`wKMEN9}ysbY!+)b!FJTZW-qeAC=zjz2`Wu4#d$|fW`lSJ zAqyx_FL^_N6?TFaHFs~;!Z(=UJ^Do@Z@o+z9`qjg7Zb@rG~0yWz$>#bxcovy^qPC$MNkK^eeJsaK0 zNAl1>ac7CV(XRhP2uJh~=L4wI#OYAkW7<_(~@n4z17S*qH{(1{kYvq*c*)1j zK8?mUDnX_lF3e#k;5WvbvF~IXC`+-)<9NHGa#;cD^cBu-neGj!b*D1_Bimd z(kqv4`9Ou2P(`-;deWLNlRX5F5wHqNXcq$HeoMMNpeX1+?7};pQPgc8K}+N0J0f{asp!Wuqm4S(Nub^Ax06vK0ld@`I{ z_w%e!J65Yu0(l1Kcn~4%SP;7-_@CAqUVk55U6q44{KBP0+_}ce;L_E{JfpqGphSGV z$qsRJ{?Zb;4Lf$wS0s6g?-fT$R*IMx0QtisH4zP^y2aoBBZZLaw-+ zPqUO)Y!bQ`SK}qZytvP65M@PVD}*0;8)+GA9V-_uyjHzr&tELG({j>nhEGDYli&eS}A0J(&VB;+z(`MIyfLP#6n#rh5a zf+o*$i26ah$mScQee%%^lhPNlkp`LK3|b=;wbKYkth1KmccW1Vj%a1X4!H~kgH-Wf z2qfBh6%f)h2Fd>whr|S&d+#2HMxY=32CXfCKs?; zEKSKPHeZLiUrWqe-R;EPHT;BUK#W7GEa%dy#4ouexNIud_OgUt1j zWP(UJQf!tZlcJ6#e@soDF=s<3Kp)q+oJ1@Oe96&{?N+yx37Y>mXJx01yB-Tl$?-TB z-u{PsG+8&E9>gmxhPK=(J|By>)KJezcx*5c- z+0pP^M{XE}%7~DNz!tY!Lg9zJ#j?};6M|CyyklfEa zB$4jjwc-f|w95iR?z~!vB7PiI)uSY}a_9n2YVd1b-1AQ#&@BsKu)qq~4fJfbI;sN; z#vFhlvvsJlv=F=A+-zt0%-x8?ytkI_$Aa{k7S1jEK>_B~Tsl9p{3mB8FQbXjQ~#iN z(rB^phC*X0d-pBDdB6#Ac2?<8V6nUUb8D2et!enxvG`x;krl6QkifMxe3l8t(`}R9 zo35l(S-RGrFU-Hk7RQk6Q?A1Oya2HMrO#u4y9EVf`~;pSe|77A3zc$3LH@yjLN&-`jv&g2C3$a=a;Xp#YrK zpOLHm?YY}Jg)w4L25-~dRq0i`OH(hm_-X;RE49$5&zIW;&odgx3h&PlzbD6P3>T4X zV;8&Vjm^XHsga#q^uwJ8dN2SzRg@-ZnUB%BubWrFJB%xk#%?r7--#l3bq3+)tonk> z{d4r#Ijl1B8fC-k`W!RiI_I#pj?9Y^Y+IQ%z9=iVB|ik6UNNasaDvWNr}3Z^hG z-E1}hi}`)5;h9%@ZsDNh2uq529dZZka(DvJr=C(@T5P`y!OvPgQI7sBnz!pFf*hv; z@zvYNe%8%2;&+%D#Hn%RJ?5hL=+ik6@5oB1z{*dEQf%d5u|W3=XE!Ifd|rdLhfFP$ z7!;dmsE$qvJj&Re08zT_?b!#`^n@F!|FN9t4*KIYe<}&$ z&V~`8x8t`qot=*fd?%o_Hg4jJ=;W6L7Zk!CjjMc0W~b2uF6I!7W{Iv(5?l{J$KDKS zd!x5aEU!KdOL{{vZz?)-0Kj6ZyPSnk`$)M+wH6D*Jil;1Kh-DlO3sXq`nTJhEj|BJ zakzxgIW2@nlxUehk{(8Ktt@h-so!t0klwE8&RIPfP+hSfYc#-BoVb)J?3T+1F8 z!3%<%*J8e?C{<7C=D0ssnM7XllF)tfQih`7?9R#i1(SU7C(Z`n3mn-;${hCi>k9x5 zPb^sJx_mbuQU-3{#l|ubmOaE`-^wVE*Z3<1kp42Md{H!Z9+v=qd9chf#Zx-e{;qXO&#HyJm87pexJ z@QaFF@#gpAU+U|$DHhxGQm$jO!u%QgyTsdV=CMB0lF?pq6&~zY%t2eoIJ>FA^FBvC zo+rmLL>h9U_d(YohEcCup|eXg)I?;sTym0kh~*_VF3Jt!^X}{g0QBD2pR-5ac4fTV zOyKg&Hq%gcug=HV5eNG{ku5=|TbO#4Px{qn1#2^sj5?(JbY8oyRAh?JK=crH{!K5O zots#xTEB7nSOa@%bl%zZu@w=U$%zPgsC3UiMmPs4a#0a<2XntD&Y@Lqbxw8xyv^}E z3I%13807CemuO%i(gpZrhlJI8&x8QxEuYY>8=q%o*5BQOsK1#VIf!3V5EIrE_BP0r zyQ==wW|rD?|I#}2GK>PT9dMgwi_=-;?1;jG&pGVIuUGTuXMIzf<6`7C>%EvbxHsUj z_W4rTvEn4mztS*!a=t9%9@bosRwa!K3#DOOgrrN2CE`} zS}M{Sf^jCuCn$>REEQ2=pheuFkX*`6qQc-14!%TDIMNWZ_vxFg^y`;n6u!4H*o1xE zw_1JUoTJxN%um#en{&Vai-)9MpWvp{&sp3Wj1RSK1XGfU)Huuww9(8}$Y9MB0SN~AT_;}VrX>Yj9xO6J`=T+7`TiDb9692A3rckUhl$MQ)m*buu9dUeIT61 zYkPkpTjSwRtcJOM5Z>EATW3ptqHf&H>dSYBR++SPo|=U|q-ZN7s{j>E!CCWYi=00A zUpxTpLFX8*q+YQUATOzpC&cZIT-Rt@6LD9H#`IGm!Rua(8#3Ob-8Ng82q28JI0CHc zGH_)~Wmuq+J?3=7dBR{nSNq}5GA4IkGW6-E#qVle0EMf-Ds zIa`MwgI)N?)}83_TNrE^6TfyXKNI#QHj47;gN&N{_>&64?$SOFmD&@-G|aL%veCcO zfjYaP$dU4@w0IolD{#3fQ(%$~E z$_r`gQYx9OjF-qI>M7B(`E_z4qmN7EO^jx|foo6~1S8(exq}Buxupko@o7Y~=dK$U zNw$kkWSe>sHW%&~WF-}qX6C{g@MiDDzh!tN1j6tUN71mK66q;1M78RMpCydmvpwyz z>OWp4Lc~9*Wd6rXt0svCNgs3Js~!#oJf-z^rSC^!%BjRYtvZs9rfYt7PVxekkbGZq2RP*tNI;2i2$Y(QBhxlRqko5{ibjiR} z=uHO#HI(nGo2im!;2M91*DsfiC-pzE3!Ayk_Ju&4g^{Z_oJ;!6v$DFPhSY}7@z{^H z<0h+K(iCUs7Cy#*vi)&=`0pmAKj>Uyl)av+)Qwc-MZ@4qy2NDl$Ml-r6&n4jecc+1 z3tPEd2sH$g1F{hhVtwO(fXEJ>^uSc$+T=E0aL3g4E#)lsEDn?HSQyMT+`vGAp$d2; zinb`f?q-Jt!$)JGj|))ZW9GDRCydhd=@(4Uf4Nmo@7U)v=m5~g9chmyF2&BJxN>}E zS)7TK>!jfZ{vg*eg)HCQN~JmdejyPen5?>r?C`h9OT(XvajGrzs&Cr1DrbMFqoT~b0Z zzohhNLLN(gKGjwQ=f;w9WvenZS}u#N^3b5MZiiMQWlAn7i}e;cgMXeP@8MGMH}qT< zvp~J&9z65stWWdiU4smkD`iGc0-#9%zc4)-Zd^ zHK&lwES*VNs%E|*F*}mbDa606nq~Q7o3C-bHJkcg7OZE!i&feIBV(aHr<4CgO8Zh# zHfg?JV6zpJ3-zW4enMP%1Zaec1r4Jho~x)|d95`#XD6CY)j%CfUzsjTtCXZO`?>0c zV%@KTy_b?~?O-3x{3V<6RcErXX&dRj>((vCWy`OWm=L|jmCf_wzMg&ZhvmKSF2;gD zT`Z;o{M}|qXerTK_1IaZWmcRI2ak_^Ef!dF=l~I3%6GCps9qrP&SA+?zVn15#BiP~9bNOrF{b`=Mt2X8KGUJ7 zBdU^c21f}-NahCenzX>Du)GKT*y^)zpWx@)A9h#^ugB3xDY)GF9Kt13UJ~V{x=1wY zc!*2$n_Qz!+G~{!z|-B14Q$$dV{wAyUep{~v;(UN~}?TB5o%7buU5V696GRLJ8}MpbE#FB}LL{}|R9H~PI(pyCE|v{4pq6|@7p zjK=a=*;?c0<2)Meonra*44+Ze>^7DVl%0E`+OWf|eq65lJGgV(Z;9e+eD(OxYpFcSkIGl>|dl! z%^%snow)c?>Ug~CsczMe15D-Z$1vTP?YCBpVv2eDvOP9iAf;H)evNIv{_v=g=VNYy znd3|;XBK~SwjmZrkQM2;V845t=I1Q_c;4F1+2PiyGRx$k*?Dl>o&q*4{K`jTTEN7} z(3QKge9K7Iv%_~;m-+;>gu45NWu{Ydo3ouN>+7%aFniC=G zm)I(>n7yQCRO_$tF+;5a*0v{tb4bv6!;c6#gep7qFEyuKbgatn9v`4lLqI3Xs;A08g-S9uPUVd!U#BDfFD;8=3*i;teW0(|NP?yJR;?Acb2 zFW0e|rxQiAek|`_Dv*u*D4B1!_(VCCWSpy1Blew&YR}RylIlTaXww?MAq*fT*wxVc zto4LGb#$1)N6S%H(Qt}hN}`EVsTW7tsk^3s*TpkfUzT9o2CP-uv0B4isdzOlec!ZR zo$EjL5J&WJMD{x*^2c)R<{owpe(<}9kUyX9udy|NY1sFG`E|&qdDh~3l|^r%1xOOB zz7-wjP@pD0c5qNuM%>T)t~;EoVWw*_P%@}HtS;2eq1dnDZmzkQJ%p?r{SlWk9=X%E ze~hwAbil{$6PS%sDm}j6oz<`=2+R7Zp7l`0OuHZ9PJ58T&tV2Xwj2;2cBuY!p!@bd za0!u$!zO^-Cno9KE3ck+;JL>^(fJ{F&j|%(I0POP^CSvzg#oyY#Vma0fM2X+#%ypvI4U&{HnDu|8T`6{2@;HAE9r0@T4gCFaLSf3tkNFTReiWzS=#0kbmybLVeDGi z2P-31D%(r1zR(Bgfgh6oT8<252qBP4RiUDH$u((;-`N{ z(q|$yprB}aQN{v`wP<;t4$>Zl!)!O)0j(4y)H*Lk_S9b)P|SX?qzCk%6_Ud ze|wDKS-b_`taGFthE)T?fWUTjlYR~=tMGalI_8-AwdizHtcmN-V!w1$o~SJjA>#e2 zbyvxTJ?;e=K6+r#;ZSbnW_3SjNG(+mJ?i&vx&JL#CR{TMS1mVVIEan+vAT{X9*^&=QeJpT^J}Wo;ktj=d*8OXq1CFdW`-GdWVdytasFLSfoXENV;Scq$ z0XxJY94;hSsel^&2t+X`iM-52cxF-Ji;lW*0l8-=Zc(Kn#t3V*$?sv|BuyMhfc&t-0fL)pN0oSK|Sfp8?_Hq97Bj+g|u|C+n ztYf&vKzx89@OqAU($3RP7RU>MWWO7Jbqz#QufMK{|JrAY5{gpg=)YR7+$m1_6h+Wd zSz2LvY^+^pHw{Or9hTF^=r4DDNwI)^9n@`veYebOGpou)4mq&4>o>QoaU4Z}O`*3H zVLhU!2tniKYn4vqQV2843paakb@8L3*+5C%sG90yuDhYAqOc1J0rKcwWd?U;iCOLi zZg=@ekQ@#>j7EAKyY<(c-s^~gC`UhLgslHMd+~ln@71L;rE%FIcl^3-(QSpioxWAg z0lvHm)w!aA`=tl)V`Cvxdcr;3s{CdCx_lla*SclFxUiI4^2rH**mf6cVdMY~xy$9v zkE18~CIt@11)jSD+wlIhTcmekm_Z}>IF!B^#%Q-#N%HUA^!9eB(nc+=<{Xz;B3bFr zPo~BIc2K%y)R2b*4BX%a-;yTQ*FGez$u}{2{Orvqr0#H2W}UXpQ@F)$a?0$TPuhn= zTm}lq`O8G%wIokarv#pbSHZebl`>m&nxn)e5`A|soSFWEZ|aWIaoq{e^6n>8-@b#z ziuUj_%CT5`U?Kr~$#6cvQMXDPQF1tB(M0I~R(P%cJaoe%;bf%nh!4<|R9yk(CxLGC z8K3lrR!>m81v@2+L?(l3!&MrLExi@bcahXX2#%7veuuG7^y9)%2&95*k4C11r$Qk}5zC!X=Sgj1;aqEYUqIS%Jv% z(5w}9lLj2535Hj<9o-T~Y3j!JTxx%R4**d}f#qbsr-}TO#;`C0L-?RO%#jQP{e*H( zfW^vQP1nOg^Dgx|{@ME!&p}RVNOJNohm$%Lnm^MR*a44jg0y}$7O4hqbQAi^3no*V zl($0b)K2e|>6JidOO6K=Zl_5oRo`>PQnje#)m4ajiqYUG2OR#5vH=1HC&1e9NmX4c zeEsiTiail9PAEjf-=l#-cJyx;xoWi=(W6(A)-W#Q`gQL%oQ`5HKM2@V@iPwzyWJ_2I~k6eFD{{A$rU;pRGY0LEV+61pUX!KIsaMeF*N&#DA?z@v7&rf^t zLkwv0HhE3pscMPhWWjIvX;G~IY|S&kQC7Sp0(JToI&VJcH&-68fg0{#X@+`9=t{wt zr@JYqHoyyAa=7$(1kein*$&Bqru&(>JV;NQj)?&?vggJt!pJ{|)p{2oa$8Xb3W?K4 zRG~Agh0GMW)6IANBxspmaA!>8_gCq>FEaqIb1!K5{ppNghq5VlQVXDH{8Tzw@Ed_Q z<$uqJLO=qZD<%qYDjc1+2joL93Cf+459&>W@wRX$CHUtc{n9>CMkw~+@KUHSE8I*1Bu;=VQeFSz1JTm*>k|F)6wDG`3 zH8(MK)=!5~{U&(CZ)$t$^qmJ2Pk=JpvHtnD-YWR_PG=|Q-#-7oWa9GfpK%=o@|#~G zayrcm(RuGf30e`Qf}5u;|A3T`@M-7a>1TK$x~u@q@h7LL!@>>Fkg712$GKB2s6q); zmb3_)Ur`SDlAZ?8+%5A~a<TrPGGk zA;!YZ$)75xN(irqB4nlYJ-O4*USo#_Kw0oa>Cacd5(ZJ*y7Rh{oIb-H34URmQ6+bJ z98mvXO_x9hr22Kf@l-K|5W#m0mr+RUsqS(_f$B&Uzw(~i*ng`5?=@(MzG3&Daea;l z2Cd_vc`{hpf8ECTFLiriI70gNZleA>j(5RJ-5&)1l5Z|S>?dR2UH%>$$k*S&KNXeE z`#eZtIGhWw@jI2(9NK(aMD}0$XI8xdv+9`#S-LDSb@-__ytoof&uZ9C@a|3CPjQU3NF|1(PT z{~0B;*Z!ZUgnYyQJSE2eJSDWO{BQaG-|`LF!G8x+|MQf;V#fb@%Kv%FU(xjckDhV@ zb^hEr6zYaqEXgT4?9%9^IH89s+-nyVM6U%u7-=HHxMtF%U4IT;QaO@7>(8A5U6YSSZVu zO>{+OeT$8gpkyII_Yz@QS(HFZ^G)T(-zKzESz1^ajbBF97vFq>Mdxx3( zLh*`9=&u%jHN@M4N$l;AmQ$va|MzW23U7u&M~aEo+P}ZUMgtG3EJLFGkr3N^nNc32 zf2Sx%0Oz;Nzy3oVdRf&!`!?+FgJKgH7{T@YXc7DhrwRo6s6QaQiEe%zry6>uH||hx z_Ie#^O#f9-)z#iDb+TaClt#+NvtesfH0m;ZlAj>iJ^yEi_| z_OF%J>x%*@8|9@{TF$)ve`#%p5b8JKk*odRrm1M5ehWKjTWS84d{GMOx1lEAU+s$0 zTcd-v6VE+NJYyJQdPd-NO^WQw!oTYMf3X)Z8-~tIEMjywFJRg=Jc)2n z)+ppYzUwh=?Z>L6|84)^ta*ykufc(pY{S)i75*xQ7~NcaLY(%X%eq^F>@sxEB|vUJ zN{%8Q+?dCQY99STSrpCNq;>7_ZSBS~&&4u4+Wjw5{!kTDBNM2h!dI&r>hp(CcHZXa z(EgQN9+S!h(s^fz?yAmTomctSLw9v2(eRx`8S?~ZIKo_HHa{M8t{!zhMjo{m9JNk@ zqRD}^cb9~oO&=M8YOl%R+T~$?h_UbL+`^g3f$cpV0yK-L*jXI?J7M+I2qql12SBMx zqnbybn#Bx!$TagZ(gF#`omYtxe*?ot<}zlZDpuBM8(`pyY2fw7G?dQFT{&57b{87>D z6b6)7EJ*u5Y0bTT`>Z<)qywIhFzbu!t21*;9kUwTmA|S38JO~OqmHXW!Vz#)idtn(5D`lYXPpv@i!5n>e*XXZ`@A+*YSR7 z?^+P3p0NhyzfK$R{Qq*5Hj-exrnL^YSN?v!a1pcJxcD;}78r84(_T>WUeMP$ticng ztZQ!sc|0-+$bL6wxEwy7?cM)r{78bmL{JY5&93^HO-ql(C-~sF2hO22O*s&E1W!Ec z%O}dA<<7?aD-ed0VVc@p3b#e=}UYqO5<;d_k%Qi1?G2$Xx>)ml)D9onk< zPcJ0`t<1O1#17crLShp~n@HLfK=h6Wm1+e*m86Z$WV(Wapelv7cJgYie%K@VFy^yH zZOs7sHAb0`*W}eQg}stbgah zzns^vR{pQ0>aSY=EF<-{(Bc!Sp#5)e{!i@o>xIHFptX{)iGlZLg*dy0L9NF?Lv9NS$wkoYPMHJb*jARq&MC~4)44yygv^c7&<1YB#dIIp+W+cZ}HU-Ln zGnAvWwYwdbxdXXC?h_HLM`vyE{7ay#OQmMndVef2fDrAztlZjG-R7XDO#jU434BZh z$&?jQw+TP?>@GqXZQ!O6+b&GgqMU~TZk!x$sbFj!0_7x({2msR-(JSl9D&h>nzeF! z@?&}rT0~lQQt&;V?R}9Fj%dz5rWZT++J2T**JZC)?zneSNqAQ2*8=lvK`Xl)G{gHf z?P&Ch?0%gf&&hJN;7sC^ALGfZ$l0RBP4(K=c#L84sgt8UUij|edeQi1se3+3ptGZj zZ)G=>^&eY7P(z;G4d=ry8)TIef%M@RNWq3ok=4GrV_EeL)V{#(F}SfmgUj$kn^d^o+K?kEUn)87+dqa%BecmBP)!{`kVin7qVMPNMYVS+Qg?nQM@uhSIGXvy8?lP#ck zOH*-``JQR=ZI8F0&ErJZg?i4jdz}T<=Dw6ZvdX`w@|Tjh&rM7|16G$$&f)O{~Rd( zWzf8mm~=`J@(V#}N!E{gZq`H)NIs;bhew=S2#UXEx^z5V^CkUS(w(rQ(YfXF@rWkH z8%D>VsAr_qQ2Ft8HZUt|3v&Zov$xB>GQz&*T&uF>Gg{enkI zcc-9T0IS3^?3(oIN7yi-SDnruH>4K}#kjFvdCz}K6N0(6F^`pWEzCAqM=Fp3kuRtC z)y5TzKC-2WNbyTg*@!K30=X3T+;X$fN5PfjRXA!56!4onL%7(VOsPisxomF=>GS`Yp!S6oTO6Bex!XN3v!x+#r z0an1;Jg8@G1gw+I{m5;HtBM`A(oHHfb5?U-(2qGIYp0x65XWk?vTwRq5i@VcqyP9V zJRT@iT{+0<0Rc)qhZG%CC|^Z;O^}i0jr>qMAAzwKU3`%p*CL&1W^E!Kb<)Q!5z2tl zUKBSJJeXhmb^%-vXg4>gDL`KNe81CE5R3EP+xWuSimV#DaS zPie&I!VYUT`SzE>4$U`EU?zZ`3nmS*3Cy96WA55H6=X-1arEZK!v^LsCCN1RpK!TEJ6-r{uMPb13EcfQJPb7s+5T?AfiAq$K{09b zc1^Afy5+*i9vn==6`K(&mr&e}4pig^@qT9U^ldt?CH~{BZvn`?5acr*W2nwDfK_V} z+d|q0_Vwe@G7Unoc574EsIFwSOhhfmJ^dG-E3ay8ujit7~qAb1J=r>dq zS2i2gSF_AFG?Q@W^L*tG08m6>(X5q@Okl=Q^7Ko#T4!faCY|t|s?9^>p-n<$q>I~? z7#GxVtIPg1v}nY>TdUnW5d`b!)h&6-D^HZ_?B-?1V(U1=juPDl_(}_uF{wx}!lfR< zzK7Mw9HJgm?+!3#Y*5ZcerdIacn*a!9nQT4 zzgIF0G@Zv3y3954u3?#4#si4p)`w9KrDSar+OG3a)G;IXN5_#k^*>P2viH5X#*GaO zd>XQmIIS4hV=MMLd{(t+OmJgQ05WSb!wb^K>KP|I!UMqVjJOs58vj_72w@7p5+;=n zMnqasveVM?0cvhF%L8r+R)dH8?Wlc+squBxSC^0c)9dLlrs6%pNTY-2xY7>QKz|Tp3xFCAJPB) zSGPicS^w%mK^f%4Z!y$YTKaU!(X=$ZHdp_?9QkbF*616l7WYRCImyr0;--YQcgpH& zgsfgR3lhabOt>_{o17wcjYLVUC%xAgnW*@>f9~G6_~1=a+}CN(*Df=(uskVMRslIKk7> zN57U>DhDW_UEtDMRReKcB&K`=^BlLYVuWm9CUZ}a*ea#;rmc$(&9G8Gp)BzZw7o=# znMzS=A1|m`VY@^~n_BGDjU+6RSRkU-TznN+Rsxw2I~6mxc8@CrWv`3uCMjh7SxOEu z$?;!#@Tq3l+(5LC%=LUZN<5@eigcG2>(IGVT6rU=s7OlVL>Bq+?h(0@OH|FA(CYV- zHtGlg3R_ve*(gc>*&a!0N;fNjWEw~w*>f(^Kh{&LgQI7R_v`z1)bHiinjG3Cv-c`S z7lBr95WYJX#?xxe^2KsvIOZ$LnLp@!^!mGS*@_wv##-vg{o2+-Jnt^UDI<)I>TZ#u zt=9>53@>oQm5evz#NIwyi`p1wd;b$Fd&~dh?vlGFjezc-{=6f2fI*;BUGAF7Qd7NT zC!NZ=`!Kgc@P&f6{EoMI(llaT)LIXIy9G#H=!N2&4?ilOu&7l$8g4pi zWs~Tk2-#Pxb4sWbU5O-@E+}dYRfw(wg5%*st~RMMI|52b1e{C1d#jM*Ljo6fm>pJ9 z&Em}zVm)j#&EU+PtaEcHMXZ&Fh56){Ho|>eKy^PJvC1Pe&F8N*Dz@9moaJU$1tq!D zV|!I~dWsRI?v_Vx?jFZ~wtT#vm?4Po>sEm>iBEiqwLcFgOK;f)8M@leTA`d^s_F~) zAzEA5();*tGh=N?E1LFv<62*NvxM2LCM+&dWc)mD@3*zq`fkTQ=!AD0=VI<}i&`Q9 z!Y`BbY$pH%j(aJ}5<>Beo8%Gpjpj(Ms1mNP*VBl6n4^0>Mg8!|@?|J?R;F&bzh3>l z!?NE%v?x3yHLeZgKu97|?TXReE3*z}(&nZ>LszeJhjA7R{j)U=AVEtbtDzSvrrLdS zQU|Gg_7rxNfgV$!`;=ZsW;gk*Z8D<`Ia zdCZoVxa0CXNcZ+!71+n(kqBqr&&B3ojj%T;N-KZG^N9L7+t;*f2}c>2Kl-0MWcDu& zdGfjq9iK#0W$>35fxpDD@AdaD1;_z?8#1uH73Ua?r1=*_(00=U<(Gh$5)3x3-q6K6HLNu zqc2j9xLAi-+OIw7POP`e_;$Wi39sVF>iAA++2y%Jr?Isbo?4sv&7Dnj->#B;hrH6I z(DZPh=orK&;U63P*kpbzSeZ#$k1oKX*<-#GarLbJ9@<|U3~NEccXF>Tdzv1-mrK9o zN@OI&5u@{Tt7z+il<#&M{WpOSoge2pFA0sVhdMo{C-|w4>n%pVryf$SZGXq~`O^Cr zuP%Go7eXJjz6<3BpY@y9sL+EfW0G~NArCJIotY(LrWD)>6suF0jwK(}fDJW;4gi1R zp2{ArM#6i1Vv`=4c1RbbpOo3T2&D%3A!FJ$)v=gmOu2|r^wSS3%KnKRp{>#3t8-H< z(_R~6Qx$zy9)TAzD8yi23-8KjkbU1PGmjLN7P6@u-vx9jC)N<)5ac_rcKgk=UjFX4 zXK?MlC9|EY)vMi;WZyy0Iwc`!da6Q;dY^T64#;x&oqGxBeKay+uQmttG-^t9r%cEj z+~g%D>ld0Ed5Zvdo-1g&9ps7p{0$l%5-+gTl{3bOd)hDsfzMI7nt^Isrc<2{-4aisFU4ooRsB5+&DLN~CxVJc=!h`Fi6=OXy zZ6^tK7Mk-nZ)Ft*7F&n+Tn7GdGA&Zp>?!f}^H|ZrQWwe66SkTnrbq7N`*^T$?g$oX zDgERApe^Neu-BmGyhHP|c^;2$n+Su7079bnZF`xS*t`KoLfutogaGCIZ2XG|^1B5d z%jQ-Ds`J{(V>|ZZu9JhDx_g)dGuwhkkn65Lp$8k%6MbE|+Zd{nbcR<4H~k+;yzse~ z^-e~f=;q~^Z3U{PM1IuKW&jxgxndPDDPt2{{`Ad%QEQIBqeLL+!-x{W&CB!4QNBtn zM8y7|vcDHs8WC)b-gcVRJ(HmLmHX_v$LUFS+{+#MJ^E>!`F!?HzP>2+6v`J=licYmZXf4l`_G0?2w;g9`4BteEUCe)p^Yn~!09ac(LYK`h6d zMvP=eRII1oxoI-deyu8E<`FWd=3>)3nEVa3rI9B?CT(mjypl5X`5B7MVew+=r(xdj z$$vy3x*>CY?zIXqs0{P@oXdpQe?*;|he*wK^Ecc^6e|z_}D)|4Q>${`rdb@R_6B0!9 zUZQS2(R+<91Y3w+qPGYeo9IOEL<=G!dhc8Bz4zWbTPL_{|L*zDJ?Eb9jA0Ca7{hqi zyVg7BGoSgq&otk0E%99B%c!o*Ltm>h-IyQy;LKohlD8@C;0EnADI**X zgUJ_nNE=0PW$;K!uye16+gWVNo&%Txm%1JqK${FO!Yjx!~Lv#O? zME{dO1LGoZVZLSSCnzc|?Q3&zLH|kQ9CkWyjgw)B$z?2kU`NW=@#w~z4)D1V3opPA z=i_P)1za6-QYPj&b7H7(i`!-fd9;=STZ9l4U-~C)gJ5(d6r8J-8rVeb z*g?c`?#A+c?XeBx_ik^6pb|QV>1Yq}7S3HGx)ZB_J)C`(+!t)#pra*{A?6H2bjuZs zCzjWIH4@h8UhrN!0vn)H;0XSw|1I#L3}xPu(xmaey-xQXXGf7yMHnnK+-@O&b}%r>p(sX!lq;RE z<11IBLMPptPvf6xo&Rx|rB>W&zV=9&q;!WgnIFYUZpWR#=U>RmBn>WX0d)@L(sdXA z5!qXjO5kDVs{|t^X?i$}m2Pivr20kBN6l4%JT;?)h|fxu`?~W|Pq-6&Cv-W+JX%F{ zZ2ur~{GiDxqN_6B=aZ0HyI`;_G;L*H6Hq%Sc>3G@E&dvaurVL=Vd2j3f9@7DYW26T zS`1qv+7Z_p+0S-gO<=W)!u#JmI827xKwJ*B>j5y!xA1UTLmeP`D#?OMEY+_VgF_Q_ zVDh76lD;qA=rrF>+ORcP<_5oP z3mq9W_gwfC^R9;tdG3SLREVaUXcgo`g3{sKpDZ5M|H5vfwr(vpzfp% zTQA(u$s0Hl6b9jp6SepS#tzJI19h{Op7`wLY^Sx!h7u;gNFy zn6N`^-y#HOu8m)??}moNQ69j7G$4A3Fjm#0Vzx*y_X}|GzVOFBtRL4Zc1~CY7_EJ| zC$5G^csbuJ6J%z7nbQ_wuf0#eGjLOR@BJTV;Q#$9eTv+}tn}Pvme!OU_MiFeKNcq{CvoCFgtqo+ z*7s>3-G}L~vy9P(qLAUzYZ!t%%kymUc`bE~ZEL+>)V&6N4V) z`0L+cADT?!;IgMC@V+fp?8Zfq;D+fX7ZL1eG8$N=k)j?pp+$Y#;lx*7UWGGuHOH4FeH&L$4Z~QHm2Fsov07aBCEdok{^HeqC!>*h3Ge zp=y-mI^4K(a-!ZUwN^`{Y$`@<+eIu7nyod!YFLhU5XIGAYdR0n4%IS0{QU1i!c*TS zX$f}{r%GFq5}s-z40PZ1m=D$BH*w;+sg{Y3+@ySRM($^l9f=A24{nNHZ(w7i2-+;? zWAnVX@P1}p2wjo8CcZN4p?iKW&1J|6>0!%!2!)mO7Bam7F??7j`t}pP`LrLYF5rcn zNC80gvsLB#^%VSnu<<|k1cMyOr3WAqYXcL@Ch%I#Rju+@Qr4zfcm zp}-_(tFykfEOW6G7lka#fRy}|lYVtEkw;GC{ga8w;)IuivZan_o06MXH`|%Z3lh-K zHH6pbwwPy%gwr|#1yxTRlTFvk-39z{*CLHpqhQ-s16<3hlm2bceCnGM_=E|IyQ-sq z4P^W#3i5mKX=V*z`FWUQ*fyS=YR{iy19%ML+Oc_bjPg8I`Z4*pZzt{4r1~t@AoL1V zJ~PMtad&S98h=J>0Y$8V-h<^LPx&{AS!*c~-JEcwZNW{FGw4NuVN;#U?H^RS{UH9v z)s_9Poqi64tM&g?^7F6GUQX~6-Zy6ueltzU^W02$$|8=LHufEKmPQRZ6HuxN+( zYfHPPUh58?#W|*T6W9Bz^>nx{ zi-Hyosu{v#lPV`A)qHNpw_&w{+u2-vCgxbWrV3fteOrGn^JEr4LqsqSW-paoe?+4j zqrv-&z7W!M5wD!Oxr8W33Sf}?IJ5A2Thu7!$%=|A(C~Dr9kjC_hdpVX|E9<2rqb4; zjdLmTdEe{bEGQ6gz9}SG3*E%NYAlKVuvHC!$o^o2C3ri+6qTdjWc3kZj)mhO{Rzv- zrO4)E0rYfap?a(pak~7qsD)XdbqS(uIki`ICD%+=01m&ahDqi^}qKIA94O0Tjm!vmc zx*}}p(~pI;ONwKH>{uIi=EwbCp+>^#bK3vQJ_x+UbUf)10Gn80-P+5>ZU%{gyGGXu z0L5OPEZ>zZ7%QrspdT)EnlTJ0_g25uG<yyJ|HEHU))z#&v)g?WFjCjaZqFTpW!D*-pu}H48u6i)(szyR z5RtfCl0dTV_YnWp2CE zcYX{vO88!C^GzzoDnwQM{=B-rSAT&uPf-}5MN2oWU8nd=aul&ZRwwYy>Y_4F)b1Yv zkMoj9el6!7gA0&0D4}=-pdDg4xH%tiV7n*!m^*(xrmSK9|BGOL@XwjsI;d;g@n?lRJZ*caD0B``a2Fti?DDj z=&TzLM#|ebZ?ieA8HnUx8hpsS&Cm2`YjQQ=ML^;%QyuS6yTGE!TBztrp7&l}&_7Pi z=Y=d|d+7YNt&RX8Q2EBf@cCZ^xT?m&(fMEE44^6?_Kkubm`1d~UrBc)5^DRkHLsv? z&b%*$uoeEJvRNrKsiT!opeyHVt=^T#2dA#gq*EtiiQV+7Q9Z12K4h~K>Tj|jmn}a& zTu}omG|y*BE#MtRj57=watT6M6tPcLB~KUpEV=$x zLPB-jg7G|&?AF?tLsoxtY28@`kQal;IU&7DA4+S0D$}ww6p~;?b?1S|-PhF)chy^v zgv@ymob_D+^xu7v0KwsiUy7`h{3FV1gWexcN^I&3Z!q6Dq$aqoh`hXpFx0wr&~4|5p5;@|}GNql{wl^ zv3*qm@KDF9iG(0WvLoboJPQXF{N$>_hn13VvlQEb!}|S$uY^6Zq1UO=;y%KN{#o;- zO>=cj#jvG`@!jWsy(Q;=4(au63rk)a&jQd?(7kc_{+7p3@_I*Je9o_l$;RdwI{+pF z=(`QdeDo=g@sX1+0VED>M_B!9^Y0;)XUfWG=!@>NuT=s*Jp>6-pI{@+9$o&P9Y z1(q-m5@4=?SyXp6+Kq-PiD0wZ9!|*Q#4>hlm>C0FkB-0E^MTAhKhS z&_0N;?szd(u&zW`$1k9>^omp?>h8NO%d9dw$ zw9U7-spGw=W0CduV5mLvQ}O9$xk78@#ilFieeN70Lh3LS$6oAmZ>)HF8m0@* z;M-flQgd%JX7x^}0t!eV?R;)argH4`(JchQKoBjGs5?UrO2cqi9 zVIq!+4R6`hOm_Q5{^~9OfE|+>7IMpK8-;6eVShI{9Snrr>c`WAXnuU3?(KTE=xf;o zNEU6DK4z@<2O$#peEr-fj6=oK3X?HW1RryJRM_4m^;Pfpn<4H|DulfF84&TC1p}h> zXB+HG`XJGnXJbjQ&)4ff%3RF@?mz@{0~`U!T>53{|70j}U%l+N*`g>EdUm*hwu`#>2ajBqU{M<|S| zN=^aQG<4E&E5)){<%=i+w0a|CsqR35wYaqaPUd$OhUku@T_l6tl#?<50Dnw%*7J;d zkA@jcFxI3HqGdA`i5;JbIh%(+1x{K&-$-Ov|PtnY} z-2_1DDlpT>u0XH#su(h8b0qKde3KWS=buR+bM>yvw`iN5%0}rJZ2IiN`e*7xVdvsR7tjPr%QWvtlO9P4L^W z5!egh!$tl!KVCfs{ljIf6KU8O(D{&3l=$55^DoK@V6}R(Cs8q(jzvA}L_G(Jy0f_r z>vyoDS7#1D^Quys;Q;W6PNToQY_ zk+8n&B-BLUaS?KUB44x<#kbRZ1l3)c`h6NU~5Tk;px6dNb{@k zz*`$9#^4|FV}XHat0$67)fBj)W?m7KDk{qx%y2*vd&6E_Wg-SZXA&4Nn4zBS^cFV; z+=X{`k2rBwaIs7ih1}aAnMkMTf27S8jhU99-u+!V8aS^+aD#Eh%&vO~ zd3uzr8a=h%)-Sj)>-J~&OmbO|w`Bn(Djrws_W(02xL5E39Pbto(W>XUt@|0!PVlLv zrze?wqEf+#Xrbc=|HiXk{jeB9dHk)EZ-gwqmz3Zo(JK<#+rp}C*~CyOpwfBRJiK-7 zBI@}&UB!uP^YN{=m^>Pi;=}R%(($u0ktTeGmyY+dW?mUbj_;T5xf+9eG1MTY7;POJ zJHPmwtj?CQa?I&6J3Y-XWj{9o16*t$>jBk^pjWq>mprd(q-n|+=NQ|}jOCh1`J7LB zA8&^a1bETo@Xg$`P-6TjbAn8N&{aia|K-U#yJXU@B;KvI)NS2NcUImn2>Z}M&u2*H z&N#tZj5o4I7oVn1C8fSb%2+-u0qeW0y2B8^ZXN7#+ZJAi zD20>)9p{IRcZrsnS#SM)|1I)0fqyq|+oac}zxcI(#is?$B&r4%6DcoXVC5O;<#kH3 z0=|ULG8ZQHE(o(%`?)LOk}JE|9E@-WPwW~P!_^BC!PnCxeq=5jE@|!;%NJ|9rJAJC=B;Uk&U6`hd~qFMdiXO^%2W53JB5Dd(g{UMneTInvgh;!r21`d;5u4|HfFRP;c^a0Y5T*7=6pbda^CEdMr1S0J$xMq&04PLYwc zwaS0|x6y0LNlffHf0&jiIUJj>noFt@`oMVH`$%&(eL91ZiBXcS$VMljj)zkwW2KqX z+hs~kg~u*HVQ&6TdYRRi@U(o6VS=~qJ3qG0E}!JbvK{%`a)1riz98~_^D(oBEl!Jp zofM(I7;fWPV3ZtpvlUYkl(0y7f<<(51&E=6(kAG)Hz6_x6+j|C|3Y|ys@=hA)$tN; z_D)KfyK$7dy&blj1Y9I(h9a{FGM0J+^v)Ne(^4Yy#Sl;cNxLO@&vLcuO;|1q_OYP# zBM5o__vEK0HK5MbrYmzlHYE~6?)YIu6%lfYRMj3EeurS`=ep-fvOJfv#2*3;uLDV8 zM);48;3bHzU|7qCppnN1!*%=%pTC>U8AzKdo7i84`gqVk7oZYJg%bw`^(kJFH=mNb z%&xDGINrGhR4G6}Io5HWhQH=ZKED(x^{~A{>82}Hl}R@jMOGU!8GPtMWzx^Ggmnl2 zN{|kHi-^*-tiTZ5nzKvSKl%I5`|8@ae5;nB+dn4s5>2V7&W5(3!H{P1W4<9?6R{x^ z7VtO*c%e5~*2?QK+h*TKSo#N3sn_Z^%Sj&OeuCwbApU_PkLiY?HEHr91NvXzLFMZ0r*4B>gokzDw42gZ|xmd70VJa3{Kd;MPA()1IcGsyuVApn4mfcb?caRRh`$!g2WpE}OXRS2q4o1t|y`{N=!h^cKdKttvIWhIK`RL`+r` zcp{uZRh)k@mn`id(J16dHx;(9C$!g@Mxl{+uOodo*#b;H)Gm1%lAMxp0HdNVm-^xl z-x~D2A*`($_DL``4|1qYe-6{%{kX~by!%htD7=7tDzJdwJ+j##?=K(_ztfw?3zGuz zOj}3A&b+uucy(m}umSUxLtaY_g=8rG4J?+-59}GiwT@GHoYuS<`36;#7qm3laMprq z`Y%DwteiM#L07W*_s`d;GBgjdf6c{lVtv>@YR8*9(~k50QQTN_b#bJra371sZBz|r z-tn8Q}g83@4dsT zkOmmzLlWPggC+SfjQEd*T!-ccnVNidF=>7Y-ZIs!^Y3HDZ7WMkZK6An3Gm|$sMF=f-_@HN8E#Y@!W@pQf%+z-Pl$_s>tj(W`8@c?&9wRxEtY!&;jcCDjWgr@*|JIp=`TgG$J)iGL!>BS4~#T- z(p2#1RYYysW*e{P#dZ~YwPE0{TJ;cXmmTG6uKX0YP_j_V&gEV9cu2S6krh7-{)#BE zenw#ox92U1jqs8dRP~kXiog6^>SYPlM03lI=EX4?d{7k5^cXRC->|oS`}-j`-70UR zkdMu3{5Jf_k(polH`?8Hp&Qwy4s9PNWf<0%<(jK--&YTyMt}j7uYvvY0$7EFcT`-1 zr>_#5N?ubZx7o}J!oS*%*=t1gZnL52{&*zE=&@dJdk?j-nM)9Ie5Ownfvd~8^>oS5 zP=m>g_@MTfJX{CPr5WVXF+e#vha zyQs#Iydkl>gP@dC34Nw}qlIeGW5k84U)7@ChOU1i2Lz4xGHi9aP!6#o=O(l@F0~G| zVX{4B@}bE8aJ>CB3);}#)3Yd%Ux>XXgz$guplKc zi1_*KkkCgv+9j6pXgT@>F+g7BSS7Z(0p}$dDh5-QLew6tR?EW&G~`d>C7D#>A4r;u6%l zzR~24cKQS_EWa5heknd>#w{v!!olAChYH*nG>C^{{V2)*p>NFO18YHnF7ut(o|&8cDpWJ4q>#Q94Cxedziy|7orc50*O1^NSilItsGb>;;!v!Gv1@io zkxJO@b+@=T5kXV0B{=dNMf?P%h)NXW=qi6}xVd_s-M<&k)cGRt`?zu0nf?Adh>8A}nIQ3_GTc?P zuB0fBqWAA@llimL(-;edZIw4~(A{e$q!(;hobX~Rk}DnJ%Z7QQ*vYAw+`AYG*~PbJ z?G#QY+zTns&cFRQtw_eK^z4FMG)t#AWbAc~Ajk}te!|Qr4p3g?VZIc>$;VqS9YO6! z!~{ulpUy#o5@v3aYz!l_eO?r~*b#huN{ckWWZWc^tbHpXuDzcOA`^$B|E(Q(lE zJi~MD(#l>fv#@GgE0eC)wN^Fl3&_=R>0X(b*?_Va#6zX(TJHgw``-bRE6-<} z0EU?WPUpE(iXe_Hx?)@zugM(Nv3Cwh&o}ek9EZGG-6fKO&f*LGZI&Dg$@VMP%g1fb zE8=~OqdzU#h23&}lm$l712=+y>i=jCQ?jrf@d4!x@}RX=;z_%Ypk)%5X-uj_TajN` zP<{SVw!HKC^Xx?4-^_)>`GITbJ>#@H`lyn-5Zb778?WEVr&EPpoVi6hiUi$;6WE-C z2bwwL4vI&9m||wV{Ln36mKbeOQAPzuSz23=iX3$_dbFMNBHJuTD$0eStlhWQGL!{Y@Km@rG4_-p|^2K zAoVeLP$0kCo9YF_`>-l=k?%}~{*Bg$Z_cWi`m5FbPKAjiYB8U=!1wiwyf4 z*E|0z?9_<-avY{v_es_l0O-*GHgkl zv`IJPN`P518eUT!!>r|yyzpx@xHycp)L+n z80VTqPmrgEdqK5@vCXS>W6C3S!4qf&Z<(dohe<}}lhU+LTi~)|*R@h?bRw0;kYk3x zw%Nj*KFXg~Hc5r`++_yy(?s2y{wm58-!Y!3p(l#ir8SN|CMby0WKIyyN}!itmv(Si z8=9dD;%z8ZO!UWfVyS{k=0d!5D<>qGzZ7K#wkcbu1Diab`jIhhNsD2|Hb16)^==C{He=E&9jSxS(g>@R-=Q5TThNNh283&p|| z?z+4OGHbJUk#~=od)(vP z56@q*JTdZSJYfuKR_kbjG=}EW)%vE&#)nfSy?i{8o!>Xz(i%-sqwOwwpRyS`voT5y zo=x*svV+@(-fR%zqMP$9MYl?xSKV5gYpa==eNe9AG=~p?{($DsBWUu~O4qs>3VIo# zar#xQVfq2ak(x{GX9)^7MgsQp8KiMpw2W6p&7a6`1scztHZ}Ry38K)6Y!9y`aXr+} zFu(t?oIb;2!e%%G-F?~0Z>r_k46a@+P@j_$hgITKUw7Y;v>O*M!hiI7A44a6swKg2 zgHz?J-xa6V<(=IQ`bUfIvcKi+{0e_i7p?iqW>eD=p;&>9gV+?5*zF{QZ z_0S#q$#dgk@ZfR?D>NSTWOs`mITfj)osNh!Rd$O(imPR`cjRZOCZp8G)pk+#x~KX4 zy{z23cm^(W7rm7!_w^(}{P$7~aetTHRdMN1#_s&RW$pLELW_8;r{{b*>n}V$dKjIS zZgloxvtj`{d4YjE6_2+}F?P%R9ELT^2 z9MO!3fNIEpuA?a?yL?bf0%FQf`z=b&>t)XM7#qt@9Z|^pyT`^36jm{zv2&f;R>5wm zBPfV+6r5f|!uLeXo)m-!JMl>%d;kr;r4Hbr||Cb!@Uh{S)>bQ3qP+JS7_>NQs5VQp$OZ>1o5F#~R zd3-^ejk1@lr-B~MQ2JT>X*w2}o06g>Yb(09)EGo2NiuCd=lQ@UB~|>1lQS+33tRD< z8K(+{_n7*WLoI6R1PAy^g;=Cgi!>`uo`F&i#Z2l6qToxutXSerI*g!$o?);A|u;gEZ3VKt%namJ!m+3}YoaaH-DZkahAe!;yo$i1ks^R^{l z@h|scUl_5Z@>xmAvvB{miQ5a#mL8S+aNE3mgk4goa&s}f{0v#i@Ogm?`N?F7@A$@$ z(e5J7=#Gj{13SzXAs76dvL_wCDoK9g6D^I?)W)G&-uuE;Lvuiy6z5gx zTcRcUAqp1`$dZuJ#Av>Jl2|3K*W5HHYFOqN8T#WfO1Jy(#~77ImQB6UB++389r4Bel+ww+)Xv_O8Fxa%1&1dRc`nr6}JD=KDTnY&H?q@mjP zizoHquKChW`F!bKCW9`b@%eBlpRcy}(`O4)G0~Xu_>frh5VXFRbc|S3U!Z;u3+!$4Nh6)lB)2%uG&5)?C>}7JHwe*p! zMV!Kr`C>BOzQYCvB;sezk9QtUm~{FCv5i5hU(x%gEa$o_H(zbu5DPsAB&{zFpNOE$Y#rNi9yLgwRp!x;DI@ zLp`ZZloz*&-me);dthmw`=y(wRoiVEd#M7jY%7KZrl{geGTj+{&C6986+WYF(o$cL z*CXfnUA+z>I~GZmya6OfM!a2>-mGVvduxS?){1-(Z_Tm%1YguxcxXpqFU&Zc80=*_SD7TuhdB7 zcWz5*Dq_@b8~;v5aKB4Od~kV}3cq4FW8pM{JWM&#m!yW=RR8#CR;sv}-~R=ih&!yeQU_dLz0byXgJFp)Wk)Huw0$Ava=aB|wA zhUPbuT!jPEDX4ffpd_B>uEXjUP@OyZsTNaKmWKTpNSB&jtmNMT5+>sf(Ot?p%ykS< z)hM$8qeio2W z?^2mvxGTI_oUPry!C}wn-^p*XXR_)ylXN$Eb2HV?d&6Bz8r;1(GgR?>K@`a#R}fQ% zDP438KWfIYHypib^$*jS&+8dMHx6;ibgSJLZFs;XkXRe(w}Vn4HQT{Y>|1xFKD|j= zQK$PZ?9N>4vRIRpz#O2d_kmEVfL-%x>x4c#dkv;+MO{Oe#K?k)G*U+A%2?t{$NmVp z#YEFi87GU-mG|)HSu$1Zj0WsGg_af>v^c54*38Lvv+6h`6?i8VBp?e~+f=q{o4c2k zHBBX+puA88^*bT*s};PCL37cv-X>3;Rvk*uU=rK8vCd?!%1yyUo1PlZ%yql~JFIwf zhLfx3&~HdJ>?)YF3$R-PDWLZp=36*fBeJ>;JtVj?TG7M1#oMb;|1}TK7h5i(+HT0b zD>TPS6zwu9dtGn>OYRJ)GBzRgccWsu=p7Ve?0*(T@>Vak#|=BMYBJB$xHs@)>X>%z zR5uOJkn{T0A0i6)dbS>_Z^{dRq>&#k+ zQ$b1>ICjqO6bq%xtoomx$D}iwZ;uV28IYQAIUI`|VGBbe?=E80;i*c_%oglMB5pV8 z19KSk<($9u4vD5hn=h&QT(Gr8E|!jH$xX5s#qJ9Xz4~jslx(D-nI|~S?1mEtO)ktR zMTT*nrO8Pk-{(Aqi_(#3SW{+|d?aywQ`;It0kKmXAMnoXdYz(`?PjGoso{dR8q8oj zL94`W4GySioDFCnTQ*BnoRP3XePOO)DBcx)m|uMEa^~oo0-#pD%qqWmN9&fK0Hu1wB&Gg z&eBt$?U40z*hFePF}2=DU(B_bLuxgHCCY;yy9&BV#UvE7DV$YX^D1q_CNmG-A`cr1 z9$TDqh{h{ASpQpnE6kYpOx_jc*r_=;<7fQ- zqn-sC&s2LZZ#DDrht(&z=2TZ0(?4N^aTi~TybqzgyS%ZNKZZA)n=VYKq_1te1h*sru0qxZj3MZ8dghqf+hCMQypolv|InZxw4_1C~hv8@+-X0G(WSPIj)SkCwZX!aM7mFHKN+CjshBN z8yh7uDY=$?JmMOG-!w$adpPgnFJM!jB@(e{d~Q2)X{VRr9W^h37ffESLel#J=1qpka;BOGEyR!JCi zu$k~STx_;2GofD!uIfljPH;DW8IDgU#2uMAREWI6@nqX@zUF9s(E9HG#P@py%~k78 z4Z873>i2_z-o1cvp?F_3Ct3%;@Rd~l9(=s~&jgLD_*^WX)!awEG2wh6s|z&*+k8kV zbe&?^N>G?W=b_{SVIF2UvSct^mY)@a(-=YRAF9iqlOxIeR(<6+i;BK$(!xtq6pKj_ z*H%hXS*}^!*5}kl!_CQgA>PGE_=ll%f*q-ZeNZVtuyyCdQ;@GIa(!fgK%ZCpo>QCA zoH+gbPst@wRME~PM+&Bk-M@dJEfPjQ_{wl?RBig#Mb573u|OyxzG3BSO_hZHo7raS zxcn!-zHUMLF+toE8xGPYyPCo(=n%t~+}zQe=I_K^y%!ZARa%PSmZ3^3y6b#P;?o;b z0Ys9kRHv__Mb~8{Pn+LAut>x=L0FhRTc$b5XKCy4jkQ7gyK2z21bA zCsut6*K>c1MBqwv>2%7W)J$CL& ziS37~XFE!BbmpXGueBJ5Czqg;&u{)H-xA&kWh5&njQ(U;Zzi$q@O3^e?^uZ@(B^*D z?zxIe`Z94_n~H#l;8JT@uTF>MuAHaMBc^cgSck7($(^=}2T&T|+Q&sbd_SZjAC7C` znMyGlPQBjAJHd_m$|z525-qehsdLt4rEJ9gC4#9666H*ThnI8vT0;W%IqmeP(DCfh zQb5{4ze2`?fa)#BjUqaUPp>tquyhdEOd`A}01ReK8y zyI!5lug%obs&SxkIzwa_iz$Z)e&&ZJlz!S)tMZGL7@1_NU{C43;?5MV<=sWoMds{0 z`TBUaLBS|cUd!fjm2m9*oA*E(D}PE@z?ebrXVV7}p3P0E+`-DcQ!O3WHXuI65_$-_ zfyC0fouY$5H}92j6kX}o#8;#q27Yp6q=V#WhMv+$n$ZMxLMY?zuGAxqee#`G6BxcHyCZW<5GisYQJZgeG zBkE?K*R*1{)j`U!cyXZA3_*t`;TNpf;@c)&K>n#pL|HFII8=`)UX(QQf|%ez4JmcV16+K2L0t$g^?>mENL^7B~)+TDO9k z_E)1zBBjDNeF`SC*lo?|UIC&rt=I>(GG&`W56Wg2H6tq@uGhvNL=M=B>OnTp?zE#Da zI1gJwQEKM08umKurN5_B;D-9^^EpXNk>NHy)@$^oOdv%-H>@9Vj>9^Bt$z|0QIAiz zOu*Vt4mqU~B&DwjCtkn&^|uL@t1?Vqx>M72#!yOe$ep%Rgng6tAnVl(UNZ&lj)b6!U2)W=pUqwnG3hzX}VMD`D# z(o*70Jd;jt5>ssk2R#Zj`#&^b77D)U7>iMELXS=zf<*>N{2=|Qa{=5>M()=TK|z*m z2JPYrL0HG*bInm>JsZW56tBr9z)y`;>&o99;apqww)C7mO;((zEsbwwU`c==kN>Dp z-cb{67N2H&kC_9$YmVhlaNAz8=Ng&9k=TFxvim*ivl;#EJWxu8EWY`mnuAaqRdpEG zsdzxbZ6)bfqaP2U2Y0eH%usVS;$<^rchp z4A(_ghUZfI@E~Q%5UTC6J_pyl;#yJq6!7ajkte_4QrD#|d6uzX#+mH=nDV{^HfB!@ z?DyY8&HSs4>Oyl}&M-j)a&fdgN}{PCMHbVZ6iDj6vlSmF|9!ier2*YOv(^%tvB_HK z&ZHb2qjo9{Ezfj%W1O4xi#_Sxo2mmrGXX&v?;-lqbFGfhzsA4qL3%Xw+pAlRGKoJe zbMNv*$m3f2yk&|xV_D!Awi^v1Dtok_zQNq)K^xV@cFqdt6vV2CR)SiHpz)$!ndwop83#lm)I-gn;&45_>@JW`3@K5Z`a~U0jVXOaDTk~i%A>t=w&n7fQ;yP< zQB=%CgO7BW+-6Z>{}dN)$D6Q2)EdIt`D6jsL5$Mz%LiMI#Lrz5 z??i>T@R}ZVa;yaJQn#}sA8jry)C60K#DC`lN<}Aw3w?+u<~32c;LF%cv@{rnW;d-w zusW)qQ9a%SJF`6)=@*A=a`o42?olfw1?iotA15c6tIH3^a;@mn$7xvh{~ z+l8P~Fz%JAuX(N4HAZE|O;&~<1y|dS0z9Qs>{^$l@oHKs{l=f#?#<4t(AJVY_e*%% z$V}sjSwhs}^)4eT83*?K1wo27F0y2hbInd-!Xt5?A@(`xQwQuWlDMA)euaI?i#JkX zJeX7zecf^`JrG>%%_=k_9C!|zQn+Lpn&0~*J_Yqwl9LZKoyp3T`xiMR=O1X4TVLG0 zL>-)iN&U`1yTK_?!lomQ3%oQa1t*#yPa%7C87=HFpa%s?E+lQ%q(8FxL}RlHRB@gK zmXE_63BC>gdNMrU+tA!C9E3arOa3(+m|Ksw6|wo^-aM{*?rRA?mXS4fU@cVZRnMQP zB){Ja*Dt~?f8VKp*M4eGrRa#Qisk7_Ohv%t6{wzyt#fgi%WB8bnDb#}!qz7)@L@Ub z)`R&MyQ4F+Nm1stb=pdaCQEz+5ea`|{0FW@n%fT}hCEz}GSy#{!m1zPO2B?w{)@9;h6XQx}7SJ zayo7fEPUIxDaduQ3SGMAm?KO4twZ@ZF=2i6dk=nu)dL=V;`qh}^kXuDr;pE;ou=Ap zw~HQ{0nFdU^pQP?R)vQmzxZxEoIIGmZs-k}NBDWNAf{G-!6Lg7IU3Q)MQ!DqG93Bd zPR9cSxkH~L#0FQ7YymberuaUce7bV$OSH4DOoEr#6SD;F zP&xU)0+^X$>&v>;=(5qs#KrZ4QY8$Ly;xsU6JJtEc`x@QjJFI z#JGxO`Y+lUBF$6wZmt~DhLjW8q+v~-`NjO`p1uBHJ4ll07~jTt)s}vXn<9!9Ob{kK z-{{x1{D-io&DWcZU|?PZex(s#!PwQ%P{ju3--|X+IbHLdlz4RvmeZYiW6HgbFcs{k z_#hkNKIE3+2Ftlx=SPbrDocID$W>w`Xf9NK_wc8-9AL`t*A#DO++?$Z?WOKfM9Y?f|*;PgRcUp~@4{k7O9v0uFU+CEtR zU&)-*zbOmL=weKFATwyOI|h=U$6{S_#(z|nE)l(?-iVJPKr$HRk8m{C7 z){97W`VHqaPy{8+iyYbPCR`p_-~Y~eNEm*Jd=ML`@1P2*yz!;Hl#Xk>L(B}BkU>Sy z9YCkD`cCZTbEsuZXGQg4I(e%z{{5N$e7QkS%Es~6Nq6p_>@7@SQFj@TJYRcY#&mEp zy?d)*>A{_jMa(ZUcbYpHlp~xS94)p{x5SqjidnC$G5TS1uP_5QvuSTG3)H7(HWACI z=#d{b->@)j!BS{d6vUzlVTdQbdtQw#QEB<1^?mKsBuZVV-vA3uGc}XuidlW|%lHLk zh1{1`H3420M=0N#K69j0%& zA$XX&SXJp|pTNR66%oF;BP(XOZbbB&^3a7OpmHI?M+?XkB)HqZ_f)vNkB!VDv(6TR zO^Ozu9O`=fC`3?7&8AH!0Y9lTrNk)BOL_J2cRiQD$woG=?d%6T0$7agX!XZt`Dx`3 ziAL>=h6VGNwKo?5+oxS?SyCTYRy1$hdh1f9m=of^jpKiU#);Ub75dZH+`kABG@~0T z0NcX#FAMCCt zTK%H6N3HM_kM#Tg;gf$^Urvnr7riN%901)t{8pnG?B5knQia%7cfP7f5A3 z=#rk0nccbqsv>NU@LqF_?rSnbXZZkO0|w&9{$=AA%U<_^B2tj6%UVSahS#>1LUsc# z4C@91Z@l`Oexwd)EIl}htmhaC6hh`>{eX)=@$W~J9LSX30G(@Nsj$3?)gTO=})A+gCyCu1G}}t${XA$kq`5; zt%AvyS}XhTXa;jt47Ez|_BrIndpXxU#&oHPDY_W0zx#No?DeTY*%}iN0)C9a*(D`y z`JLLj+c1~ucnO+@Y1w+Td4Nh{cmM&1g*$yt%W+|L$dSw zuI$;3)zZ4n{o$VLO?Y*?;*V3!R9DuG%dRjThkPmtZoKs8xGeCQD!#2@@Q;VZJR#(@ zi0+7ARBX@(ET9sd| zO9yg2m%NF7tCm^i!L5DAgMK4^eKk;oq3Kc)+hw$H^62J=_0uC@$*XB^T4a$+Gq2L(+XjP(Sacu zR}m@B9A1)4L*r!VBFm(9ijbVLm3s$V;P~aKPniT;O_?&=z$kt*p2DAZH_L>|tl-;F zPa}ObrP02F53I%xKe}wx~>bP2*!2sSM&>8PFB@h5J8bcgR!L-m# zdQGHT9db(vyl-fsQvk2aUe1dobT?HBU#!Ao2PT~+OASDr0o4l$i(p4oUVcpF!A z4$^giTlys%kDAOnxlyW@%sgT|$7_M`RE%L-$E;iTK3wY-geMO;?i;>!SB+kWOSkSg z9=&jd+Bv2lFu;_a0vCuCUE^lCU{UWIb*a<`H3=4Pk0N*`zMCT3MR3++Y!0U`oCTp0 zwM#uj{5maX2Tyl({jEZ4WFv_FWbor2J(b*S4!z|&uCi_aD;Uq|2<&TH+>y&}%N)va zyhV&Fk(sMV$Cli{<-q=ju`BkWItH29(~?lecXwYHj|Q<_io9a9k7Em|pQH>=<30R# zZFhn$Bs%%DjtK-CC~(-I{FEeF{s5}Of!+KXTl&<|o;86C+jL?sy~x+=qkuzd9PrSz z%`khZ;1J$bZ11seB8L^Fxq2rjH~TaN*erIiHFmb}qTqIH&UI7SZgm7yOSY$L4VeP6 zetF*|pYgg7@f`ajkI-X~x#f04qOz2X@FXR$45ci!ly07gXugg!8g5Eh!%-B>uncT% zo0_;%zT0pq+`i?B$k#+ACAKoKAtfp(M{%(Du&EkP2{ds9#*fLj$Jdp>Cp6aI-$|&N zOaveH0~T3CGAYAm(G8`xZF4hgKQ3Xif8ywt{WVYWn2s#hVS*H(Hd$6wMX@t9&8&nH z9`=Zw_IBpmZ}k1t<+T<-)xFUqfDGsg@-B}98HJwKWl;jpjfcm8;OtF)h1-((fv@W` zd~HiW;kn}bEh#CKiZ~FOCC+l9#~g)qi#Nwb7yc}_a23ep&-{%Pbknjg{<96m`~$NB zdreb9?q>^sczzaBEOA&Z ze7=8POrhJyjP80*yeu$qrDA03tGmT}^L*~~q`xm8TzO8i71^CGuDgj>MBcXV2hcgDkGoSk-I7Hy zA5$|;`8i+~ce6<`Hz*LRTsD2QCKR>C-{C#CabF6w9O!?)uam(C@jH=Srx&i;98{1r z%uN1+S6yI2)3#+fprAmykUHz@34X_t>;PF!Hc?xjxS*u(tf!Tu+C--vY8T*Y3!>xJ zSV@S6r0dWmOp<|$7kjv?Eux(`JN!HO~D_RPq zy2Yf}RQE%iIc(4C#rs*bLs}Q+;(fUhZ0o2bklz?Ny3^P|m}g);xNcrz`(>?hM9?-a zVqs!W7*eRTm;DUFz2bDj;pq7qcmHwX%HD}enV0NILlnCPoM_YF5_uPIaYMGIo1T?s zVRu4}y31f=osW*B=!lKtMQa~%$T_UdK+aIJ>MW1Q-*wQJFfEmu#&~$c9nyNM!A@}S z6~Fs)@wDLDiSWIoSi#W!a>Pq;vbMU8KH?x4*b_V*pEn|eUIx$blOCblgbYh?1U$F_ zuR&h}tui)#-~%D%xKaaoYBQ*1*P?!;Zvxn~{($MCTe!8+aN1qlUt<)XWbjt9Z$(z| zl7J=*@PG+srvp5RUxn2>H09n>D3HX1mZIqcm6+8F`XH?0thE3E=Oesm&hb2}QO<3W(Q4JYV(tTsyvPUItBsyE(VU2nDfFpPB zqOo>97fEs51d{2~s+t%u%|@4|7{{E7Po#*^R!n?9hVMSqqt5H@{X@je1lL>80oeVb za(m0-{si-)jnh`qbj+X(OTFa! z2r_nivaZx{n;RU*Od!$L#5qWB+M$r|kO2;tof#LtsTIP{T{rtxhM5@bf}{?(VQdQ} z4{Xux_cYNncU?8}U*lZ;t0wXbgA=dA?T-ypK5|#Kqb+QISo- z<|*bLGRN#G%Q*}3ypyo5Go*MSs+nlY>bluGjE*Of%}*xu>vt`* zgpcaSj5UU;&2j@rN3%`8%796^Q9yVyN6{FqEEC5WPt`FQL*#ug0GVfF=ITq&6~Tpv zlo43Oj@vvRNd*K1Y05F6wAsXEEaD)rj)sz4ykY2-)i=*hsAR1k3+c~!0a+l8x1~7k zz%PPSG}MQ=i{&pz>QS7$i0{Ou&(VZF$;O+@r24xdKTY&JF6=bnj)%oY*W^Z z=d3;y7>T3bo`b$+BO2S@sB)${7OBeo(d$NigmC;QS)L0uiv$Pxy&5xeKBaOK!?FF`R3HXJJjV5OSK~DpTpl3jVn&m)iXRZk2eU`FT48+EUxSmb1JF`0(0=Mis`U+TH6SNJ3>YO?BFBYJo~x{Nl)O zB*ek(jKy3+CHWL*HvVw`RB}JBN?+y9Mt-t0U@D39M33GNbf1^)W{wMH47(Sk?HCwV zcPI7Tdl?Yh^?N9g4VtYyO05tx81YDSSo*XJlA%hKF`}mP+Kb{Zya`&=)^sJI_8>X* zeCi`>h5Jk5%0U+Ny|F`yJLQYxyCE3Gl$6Qm(Jz~Qahs~6wCd4>kJwMX%coH(b z@`X*LWvPY0OR!yJi~TuUIG4NknEP=-AF?)4iB^V;mR?l)=xsXuP5-hplgS%3waX;m zG_nx%$IlUiMLbE@nw^TqXmDOS36oN%`K>i2-rRxRbg4g?nc5b_Evq|-l`YwnDE;e_ zH^TUIP)PY?!YAJo^rg(%lM;-tFBMU4>0LQUX&Tkvt)Oi>XxgFq34E)&i$?9TcZKg$ z=(N>! z$WV|jYF*Qv7?e`LAzVSjlA^yzH?gW@xuIOc@Q|AO&R;bQjDA099gwt6;}4oq1al}A zb81en7b2sXSI}7FNx|^jrrAV`f;bt$$jV0>_eO{HmAgnZsoBCp=DldY;*5l=4Cf;P z0J*zO<5k-;r2@-r>jf&h;*{Qe(6AK5?l_Lno6=V*_AAv=1E{L%RVC{wgRfLU>Mq^z zm3TGNrorHV zm8sRc_yJgFS=*<(PObq-n&fXJC&YM!;gG$$lX`%Uu^m_`oWj z=V?0<(Kc59-V(~2@G7bo4#uYeJRv`$OJ_DSQ3~5t{(9~BPv+azpOfF4Mv4vSUs|i` zeBGFquH&*@uVpLN%{z$5D*n=}ocd-2zsG^J(3@6t=qcpOtQn@hXIUv}Yg@X*68Eb4 zlgfrM9XFTT1|UvsZJ!$erX4%F4v!NVm0*#Oh2PG>BWMXM7ycxIgBG4sm&J}-;WtAI z2h&=nn3c|*gNxU9XYdlu{P4e64FrVz$lo8*w{-{x4_Jgko+weYx% z5EXZjRfKRCk#KMfpK-iToLwoF-=)Sx^=E^x+3%7q8`0VP@m~)j!1S>Ljpwe4({WwM{*G>IHohEl~3#F|T6o)3jwqgMyKz z!q6VM7TeOa8!`P(cIa34cO|u8q@?E|L~$}_iXNh^#oN7N3$T_gH#*`a33>snY*|T| z%(xT|3R?W6EL|Sf%7xUOJX+iSP;wi!5y_w904Tp1AR_Q*+xY9=Is6^M$UwlX9L?=( z9;Z+!ZorN4PG$n5^iiWVtH*lI$?S!6YZy`8gFx%rW)Fee`WWnNDXorzW(e%8!*3Ej?n#c z4fYyr4-rLCxo;rE50^y=B0hcP8&oBord03iNo5-1>Id9!&!HH<>f(M8*NgYvMFY1g zR_5lK>%Ze6!UIP9h`^DvCv%FtGR+hn6L*z2Ko~+pwJ~ZEb1yihUh+sZ;TLQ*`Ta`TihEisVEnWp)9=yx!fX`zUba7kG8@Zd1QCx)EZ>kM z0_{gEQgk3Op*r_aRl7u%0++@_1#nb21i~x6sFT2OtQJ+vuhuTW%T}L$@4jE-DOjE@vQmW z4rd)A=c_@VR8<*7Qv&??U@Q)3)(jK*m$IJPar=G5#=#?a7lBk1uXB;uSX4D4hOdf2 zQ=nfLy~6kgSyU?kEw(=2Fh=veUk{lDKl`tXnP%={xI-QztST{*kx-kSI8##h$p-fpo z7Vlp$4#OTO`ui{QY0!vg*$+7~NUPWa&BiY2KDkip<)gZY4X@UA)Aff61;JElkxb^@ zEY7W^Lyjjmh8$GriITPvo>pJr?t{%$1-82dGdycZ&2Upt1{dU2?1obkk87|_QC3nl zKZNVJPP}N3O}@u{q+r2#cPdST2qiv}B`;Fo+Q8rT?op)d{dh-A!H1TLmIgqDH~Dri z>id+-_nE8C$5Ts^piqwrG4Id%S?+en1bK`u!p8z$GM+^^s!5EV6NO6MSG*w}m-3g+ zBN`(%(`7C;rr8PrImYQ2`ilrCkLyf@hcbc}#b9ymA+BEd5}KH*R4xL! z=<1>)sd8>kfZ&HmwDE{Y&%O@?b=`X1<18Ci)H?-{B$w#Q%k`UhcNTd?z;K|7;^0%a zBbyuvDSxj2xXdtKKyTzA3NRdV8LpkU@*^bDliiu0O6~WM?vTy=P0=s!_-TNXbN4aA zZ^T)q&mTD#8B%ZVjI&cP^5r-P4tuG*7t|6tT5`)rSS0+~>|Ik6!FT#1?AY4YrVl1BcklZ2O-p*dv z-b$quUA&LV_ghHxEA&RmfT|{sm3rUT^XeQIIw}OsCWc9CvB+Fg%I8{k3M61vdaJeZ z2Aw$M=S(LeXmQ}M;xd#Y0u zLb|Oc8CSvk>bUjmyITsVDDN+tE|luCUX@ zN?834^|EKaI}lKW)_o2BfPIptDP>;<)$;hQIj>;)Yln2lYnxyUaF9=}3XU&~ivk>+ z?@;(tKMc3Z=Qpzo|Gsvb--X(yF|vD%7S=KQ2qd+|S%mR2M#}r=(6$RiKhR}v$`}mQ zB70mquS;G2oKHki*c+*p>QH)CO0(IO4{Yejg^Qr+2Nd70{(-X6=z39wEo$6d0lEv& z9IgQ*F^t{W8YZpNOhrcqW%eb?5S4dad>nSY1$wNJtrfr!jr}%)U5L3TrXaV?ZV$;ekdgvoaYwFHSIHhTVC6E*E!4DmgT`p#Hux%xNq;7y zN)Wt;j-a9od; zb&IZPKH{6!vC0U=bcvrv(>7HG0vP*sez79xc_|sIy!)bG&0YrHm)1jl$^s<~xNE&j z;UmBN(6g7x?&rGqnIiZx(;@~C(-fnu0%P*i@zM(K@BgJ>V?_m}y^}klD~4k4(PVKl z8ZK7QzU)&DV}ibgef_TWu8i-)M~rl57I6!$fcrF@gdopyf7KsxmC8M;nvRj&o>d(? zv*E_7LTxQ3q-rPf!@5rGG#?i{(;a7S2=G?nZEU(Jzmxc^q`j?rC95&i7@&^Wyx_Big!nioc zY;{z_{>7{b6W1-#2T5&_j#i7cbIO&fD{Z1u8Rbw+Tg&M{N1KG)h+j&Q6tTCV_w+EI zjo9ae!C3;i6>i609mId1+-Y1G*F&dMbx%UuYIg||w4U0keKN&X@+3&pfE!=db31mx zw7<=rF@UCdETs52B7r2SRH+Jjs%q9j;#<5-{^Jf_J`iL!q)hIyl5dA)$?GF&VzFFm z@6UTKa$*E8)OMJVihjiuHzNC-rOQ)r_*BOgkrEDO0$DEKqHIZ1r$$7KFp+wwmRcJt zcJnB9<PkWK~Ge1Vm(;Y5PV!&gVqHdYXCCNY`&D4w{p1C{UErrb;!i|@gd**9?CDyfSN1E} zZiQ@gj&;0u&1TQ%dAI=ti5~*s1e$(?()9eFi53C!T!ixEfxO!Y$P5la*o1IF_dY-X z8HJ)tDMH zTf4V)vSeP)`vQc4hbf_HZ0Z7_VZ|xK6D~T~4vqN2&n;1&umbB#@isne{Fl590u80c z_Ysv?9p>%({K4{pECB*TmN2}WB9cUhC%($;ODWUv1Lqip`~-_KVU%ukyq}&y7U?4> zIpQoP{f>DkT5R|whblp@=fjX-($!4GIe5@lge==%q_QkT5_>+fYlMM2LEhirs`>iv z(6Qk5q3-WSY9!dVIqh16@S%g;5edh!N6Xy0BOQw03hf>Y7mB zZ2}AdS{Z{5c?6F`n8ijrO$Xd&zdHC`249{Ob-QSUXI}dc$x7iTV}Ra~0z^6Us4)q) z`=7RpR4y~>etbe8f-&2S;3n%R#fWIE<7tH8K!w2Ud;Om_?j~5o2^C=XROXaNpA@^~ z$%lQwF#BEpUGos-HXpbMtZc2xo!$)K-~wU~&-m5Y#>2En^GV3;72Dfby5Yp>8JQ2c_*P5iSn7F+F4WZGln1chDW5H~>ZtgYOw7`R8^ z_loHd850366O9@z2~FM9-6V&#K(fHpA#3k~6{$2pkP*B4H4l=ybSxZ*=kezTs+h8}~H`72Hvrtln_J9%mQdc@cH7zFky0Z7kKy z$An!pc(tyA%0PaJFoGU6Nbj?!pPeCjp1qn!smj(M6NZKMHiQ6;MF3K3f^FDa=&j_g zp4C(~Tg+@XETL8r)6v%Jd|8;Of_T?Sb*YWyBUFWFhZ_U$IHx$&n55mi*%N`F8vN>v zwx4fYAgUv5m)RfNov?ctEAw&_JcV44m+@wqaW{uphpmU9&05>N7j2jV>!_Dc0>_Nb zlCgtM@`T)xZh?`(BW7g3OkID8;IU$!tNfX-XQwWCXHHzq`QoX zrizw?cv$y-#DBlvk7YPiox+d= z$vmMY<^`snqvbnvMsLeM3l_e2NL{9koYGrg2G5c4*sZcW1Er8BP@t zE!X%QKLlBK2Ra*HPOIS=xTWWrMh$L+7$dxY7L2YmNrFezy*o?}bHPr(3;};$uA$xP zJ9x_by?4i+Gl+AOl)dXU=+uWQ=J*CW^lRt-z&xFom2gGJ`^^sUor=r0SQG9Z>@wr(--i)g!efejv6_#4^bWR9#^!RCHgnu%yyDg({AG#^Y|o}cq+mTf zqT@-05odCdYt&23AiKChi}jR`l-e9LZkcJX8v4pZ%$oCU@1)R*CBlX}!R!h<-#?r> zp6!k~NRdHiDv=Zi^7r*w2k)ZqiPLuA%c>KFZIGqEufO z_S$IZ=CVs>T?F9(8muCVA9GV5e@*n;o9(yLVk?-sd@u+#^ehlDo<5vbSwAP;N;L

j>F4&zdOTVfUew!@Gu z((2Ao5}KB!u53$TjwX@9oIK1rOm%N;){3TjZHgxKSbVKIZ5cz0!w|LGt@yyrqIk^b zPrk=EBX9rVc%w=Va4!NDn4ynZh`ULU0$ll%ySOcD&T2C`g)_CV>jep`QL04ek+WqQ zU#B=g%EcDQ>nhQ@uJ71;zG1gIci;%wbb4`qm^ZY(*weS7fEpZWl(|%M=B(~Nk?a+* zs`E36-{~y&1>Y!)%&8t+iKqnaQH<4x{{;4$VxWO=JPf1*+ht!YeZQ)ClgmVVft#&x zJj>opBN}21yptS$w1y(J(~w;=Bn8*qe9KA8UCgzg+3^j%_(h4Yp%bK;I3j~hmQQ5P zCrr1q+FN8g;%`u1+=N=;oBvu~K$z%{&h%BZax~lRX#32GKRsbPaAyE93QOX%!$BH7 zV$oC%WAZvlnX$^DYQ+)ZFBxC!>Cyl~v=?lrz5Y@Oq{GmB#sErI_q@gE9i2Z!leshe zCyFV0;go3ZB26N~)Omy~LJpF{1u}lT1LB4CyEIttKMzwkpW)x}{gIv-TU7#dcRI*g9Ox8B9HiiCX13QjDh4RP%So3hMW#1iSAXG1EqFj$bJ#V+qSw6*LCk zzGi33e^vRwiOC)~+_FxmI0qg)2Uk=%2DrRr(s&)-{b~%VgYQtF6HGEV_`<|_T0_nCx1=-h*vMWBR_M^6-k#=~VnQ4pe^~^oT2f&gbO4gyDQ; zdMJBPL7Xg8W>(NZ6Z_y5f z>-AH7AP&$sP$P4pagfwkBaOk{g*-wpa@0_;<0+19_5IaKS= zcvUHNG|ht1d`hF-;%^h(u(I6VVB((nx)BY2f~!2Qi?>I?|M*;*9ror92TTOzg7ZY) zB7om&dIXGq@~6jh;bp4@zcHQSKNd;|&5&4yAKzZ&KYu2bGmzD*?xDf1;->w*@j-Kt zY)-dmG*%h_-$1V79La%Z;nz}fh}%(#TtvBjsa4Di%m@6A0K4HVIm5+5uh_OYds(~V z=~>!g8Nq__-eTR$$8NzrR=_12!@7O8XJD`Nt@0|EvZ`x9Cxbhk6HG)P1Hof!-sBpK zZB(ID`9T1^S=!c?1%Eb>7gQn@Dti1KtV|Jg>1+rufOQ**=h9rD-^#5Ns`=vnWDDf1 zd39Dd1FA6TUKRbgl85c`d0W9?B4&psXdl*+br6Y%*hZV+M$S``nHE3n@jV!i3Zk`) zH47#Ep`a}|6iWo~Zm;$x^>*Sp_6&=;BsTNz-6WE++Q0vXFeLzCtdzkj_$oLnC3GX#tAiBRZ1{k zxa_hjkVAPSje^yT0Z_OmZSiKFX9q|RC1BjC(Lt(8GdFt+Wv5k|ghO(VFD4{CK1Z1M zVL-3Abq(mR39H0S2t*2|Qo6ZGzdg8!k5SWp!y^|)OJSb2?}{M4+5x*@0I$-x2Nd89 z(1pWjl<%26Ew0ToVr(@bS%YQ0XPph4;r<=Cm7G#?@~_UapAlvlVvKEeoOgLVBdK*2 zMaIo`6??ma2}ue*+6*yBARm<(qb*2BZMi67b`P?0avObLw@Z$ejo|UL&#ihgM2KhT zvbQO#;7l`KWX;Xn22m9q^j};sGpt)+BubIqB{td3X>~2LG&@o{#Y}$g!Vl;AE1&^& z9pic!k({i*=+Yws=v^)TROzD0cmCI0`|+=?SUF!_w>ymSOGWcWBCv3W=p=k1?_DJx z;}<)?Ff2DYLlIK{EC1ff$%f8*{9uiY_1|OQ19=fb)Oh=t3%@^b9^my}GX;PDKEo{0 z-0vC4JA21X<^(_eRc^lihoGtVBy-)U(@#kbyZJgI*;Du-$Q%RV9Rl%{1a%bNM<=fK zoBGP~D=uzPGKXn7BTBS$5zbp87*>n;ol&n&T@mbptug8rv6OXxB2C6NJMtuNfv~DU zEo$M*dE?BF&G14|GLHqLGfM=hdf@4 zCmRsY&r|%7;15qI3nAGE70cBaby=wEJNOZkVKrT zrv}HDIg@PlUt6a$k046*`f{+I_qSaZpJr2Yqf@sdd0KAvb9g&E(X4;#{LagJZu^}5 zv;NCZ2s?aa8HFL?RvpHgs_8Oh=Hn~Tx7+zj?PWb?aD2U}&)#@>+NhL39lTSUe4A@G z`JZ`GWrBoTNb=m5JPs9od96hi+^_nH@}O4POfZNI)LFYDyxV>m1S_(xz8Zn0` z6{aL8f*ww|^pE!S4xYDnw|XalJfzgWbRz$6+z}#R;pH`^hG7bi03W#{+F?aqJO_m0 z#7>^)f_67Y|96_z(*2wxa6;j!B#d-^R1{wupe3RQs&ZLyYHI9^rAw{s!Ig=9GFcBt zSS!mUkXGz#Cmy4D|K8P5Z^4u7FbMPyp05wWUv1A~^RKFGVE3^E=ytq&>O#aRld2O9 z3m5&1i$iox(#dIN|FxeL$YdmD>=!kbR{4FB)&3|y*%bySN#0_%SLR$ED8sonS~Cm( zNMNSlw?Y{q0$1o8sy|ct&7lxsmob47_HP|mCe#g7lKlUwUiSX=rKRFn@vf0&BDy19 zqrJwrsv+?z8dw2d)d*n*p{Hb<`M=TWB=IxIecwDUKFe6m`LNLmd4|15oJ2KLl_f%x zbivrT|4~!%K>Y7mVPL+zE`y0=mrNa6)=gqB%3%J9Lp+21g1(!}?W&WqtsT!=@2B;P zsI#a~XNxdyugsq}0)`W20w16-IPa@wS(w2OF~B zZB97*e{egqI5)p{|A^;qx6Z&L0=@_$gK(<{OvY|A-PV_$MmrxBZgo4abszNka^4O$ zMBOR;9eI0Y#?=;c3Wq_qzQMDMcGCZk)BA7b!aYwWb(b?BqQm5jKSJ|I=hMv@Pu88{ z6V1Lfm;0Gh*2+Kpjr`NdCY}s=^#&^XMW3-* z%_v7_+e17YTkPwLV>K!Yhxr*ifXITF6v_I6XPkEkRJz_}&CNzV!f|lZUT|(=CC951 zG@S)Kr!P-xG9FXL1f5P;hTNWTjNDWv97(lC067m33QDR}t!DaO|5iWwpL73zDuBh< zrNwjAZ_$y{YypKV8@#N*F&;vSRX+Pw4fhv=R2%<5HZHrf;DgKs;fEz?g)~#DbP-k5 zN-jE+QRF^Pyie*w7`+4*NBf`YvwvjSR;c$)?SsD0qJpn%zlBA0m(XsE#4$1E-&F<< zR1Wlq|G3QmQ3)HxSGjS@-k0%fgj#x2UhsY{H6-Xm2oL1Ij5=9|<1?J?Ja@=i;ef%y zh9%2eYa?#<_Ul1rqjHOyTn`oZ)N={~DF!b%xnaZbV5H?{*?OVhI^|ctFPc^x9aD!Q zg3lP&;B4roG92@Hh8GR$YpGhL{(oBGKi*ghAtJy8;YRmIbAIc~^TSDoX5+fM_3c!& zx3jt*;qQOAG8Yawr9o)FTI9S7vDE$=4)1d(tX6u25LPinIXk{KiS1w1F?hc}b`PIm z8ecczAk_Bjz+*$pgkSwg=bmeut4`3(D>XG8nCGE<=XINZq4&*d@Z^2aZiYWnD6ey z@a4W>=U!d(TK(%}%cdXZvAu-Bz&{2D1r3P_kx2M|t(O1$aN$%&`rgVt_*Udi_i?!4 zqe-UjJ)FNQdAo(<(RoA+E2+U_!1YS6Q)>zi=98jvEd#Lm?E#!U z<`MzZr0D}_7ma$}$D?nhiiT4Mh3KXz(zpvBt41%&PeN$_*H~WUeJxEz4cy|n58*(h zQR7;ha9Pk_@Y?6eT5?Vi+?uQOcKNqr_-?C#fBLUi>(~XQeJs$Jj83s`Q}q%HTiU)MUm4mp4*EA`zFundMG{Zd&_!=Gx zTE>>5W%BKaWUGAo8t^}AF)H^flQQ^v%v6dj>4sIBB;>1>o-X|-d1K`gOxm!N zkHJSyB$m!_+^cJ=>*+D&^LenHpr>+%2)rUkJdkwOUj~g)5wEGkLw}Q44J}~3!108+ z!e+l+44fI7Mv1Ygu2Ke_|8aCr+O!L^yxsXa&3al1lmxn|gH6@1KmX4PDgT@j|BuT6 zx}20Z&S(D|GwnwmNufOWHQpUpY^nh2M;-(OXtrIFxXiutfSsx zz_a-^JvnRnKgvSN5VjtiD@!r2U9=GbRK$&@Cng;+rFvLoajlMa`Cn#mO!PGt?RU{> z2poCbslAe*__de#6<$$QnfR~!E>sbfNTvo4nSZ2#5|=qRK6{43X^d;_A5VF?z--$E ziz%g0yOA8}ZB-I2MWHuwh|XAiy&G&@h!jAIXC+9A>twN~o*y5AX2|c^jcK5AFfG1m zLKaBw5Z+wJtq~g$D%kM9ycDM{0#J!so+}5{H;mbeWIdRkZ~%vY)m|dgh$7oc?F^1P z{XGGPen;JvrS!7=ANJldF3N3f9F`Fj1f@%9q(@SbF6k7hAta>)1f--vDFF!yK?M|$ zPLYt7E-7gQq>=7s_^;c&2lqan=l$@0KA)8Nt$W4Q>#9ZXZoy+Vk(y`9w)iDgfmQ~!Xx>xuNqCapoc z=kM!`0x2~~b27*xqxfSj&qHqm=h;Z4W%T5d0MhQ4Uwb}&Bc7Diy_&$+&d<(Yj`^XW zrnZ)Wr_&zvq%++h1K}N0CPI@hMA4z4CmqF3%Z1GYpFhQ^gg9}F{DCv;#B+7TQQ6CX*!c{x)?)(@f`kdYY^ zuPB8Qk;T_fS-f3&sb9H8TtLF{bxx!SVHO#0bnmL#z2lA)e~F}gD&<8pPEymyA|fHd zaZWRlGMZ>1lK~x>{@pQL-7=I;>Iq_riYiS1LE0KvhCa0sq9YG6?^^S~BV2PV25A+! zh6R#|r}{Ot?%(24C%ay`{efeJ?0$V6r`Rs-J>kvB@Z~FfKYT{g>N$NgA2<@+t=%M# zBS%BO!7WWv%`F}v+thOR&RnnY+neps6TQ!ypHButJqJsz2J^~k9BD1yk3^Bbhd6;= zUT-jIeyf3VRR8kQ5^I|8^Y>^g>?5WK1lLDtCSO%R;L<%CHSu^fTavZLPoH>jt9INi zoT4lGAakytx~PBjQH2ewy~Kg&Qch8Ml(Imk&g00!myuI=AL$%voqMj;CfDAIN#(rz zF9Bt!2veM|vmgyvHFbN{q0v|kc)222twTFhYzMpnTdusvZMJSg?c9Jb(`TticG+RO zM@B1h@y9dVybCZ7ZL0II5rpF2(=R^CFkRYrKE=cj6L|<>+c<_<@@DNbJ{_4-X(8SI zwNl||@i&_%nNkwfXA0UR`i!5@#p>dyO8$^SqWj+#E2H?ki8&fY z95f4#9+Bh}YPPIV+nOpbi;xn+Q>ES=!QJ?m{*#`cUd8F1M5`%hiqxYGr{BFLcNlK9c$j`#H!e&^LWuM6LES#+R{Hg> zB=@SY`>x~d@h2vG;QeEdti8kSbkQmu^NpSQp!*05t{F}TYg*?*j_FGLrlItxe-V^s z$hj5rISlD{5`M{-pE2(MvXBBXKngj!(NF?=iudtbn~YYKK`(C3VD_S z9X3C6--*?}jpeYW0OGfp^PS7n@!)NQ__Nnu%%nGX%YHtMTU5hct3xU#zw{eb>&(wjau)90lbx7^(g`$r z8#Qzq3f(&i*8x>UZCCAv;nnuVN<8hVHO_2&%FkYPjp!zP+ozkJfQQEw9XYGxDKCYE`Y@&(Li{@gU=End*|nOcgS&r%$BZpW$B*r1pA37@SQIX8;2p zvp2j=*7E50T2l9(+30z|OPK8*RqYH2phtc#S-AXpF)KTbC?)R-&jXm`CDH)mlBMk@ z4eGBX0%43*v_vBC?NA!WVL|TV>)eLFH|R^uRJ(8`CykqlcuSS*Ht|zp8%%*O_yuGK zkqneG`#AUwtZXY`a??VaE%Zyr>X0CE0p@Ss3*@`pZsq$gJJ%h9MzKVGMsrqLLg*#Y zuD_3fVd@gWi3Etm;0JBAX*CElQ?DD<5l_aP2AzQMIoq-NN}tG0fjMZw$Oz`@hQKRf zj^jp;+=)Avevz@6A8v2fI_v)r%X}{GoRfD$dWTGmD%lBtrUfpiz_Up5y$Nsc9sEym zFa0=9zqYBKnm;6{K5TBeTz%})rz(@HB>2)&ZsDByd!--EAIxF|^vy?rR7+DDT7MNL zrrV+#^nnS(KE&U~%@{$JNM;4vS(vRrMe74tWQCn|581FKNCtF%I$1KkQ4tSpA$Bl9 z4*t`dVTh0UKpxLSAdLqU6+EA*S3!uYy2xZmL`a0Ty+V@h%E;{8WKy7Rh;-ZcC)Gy? zW@pwEagICq`y{)Tg+l7}iV?KkZVdis0TOnVwWL?0xf$h5IQf1}l3qsR#y3mC!yQ#?>E1 z|58!Da(H4E$yZ)bqp{LoAF&kA2&9)a&7j`;{_$>ellyhwfol_&lm8kz9oAZ6VLq0p0&huEtxUX3CkR{X(>V% z%dpm2mpc60ZlQ5^6Jf6s>DXz5V)3 z#2A`m8ZtC4ocCi9cr<#DPWGW&xktSD1|7)}g6}pqWp%uer}=8?5_|bV^6K_Cl{IBJ zJvrWo{a`*3C%wf6zQrzhpKH+r^z%m`22%S3P+`;bj4HtpJ}gdB&2XNLv9L1u;N~g4 zm+BKxk)qE$H#Sn?V!M~A;`vZ!YOIFY|2Cl#_w~p+iYN=k6#T+<{LjoCEVV2hMvxLMV#$}`=lBZQHBvn=2g@D+f!3-K-H z&L_C&Coofrvy!Mg)*5RQi?e>SNL_97?$aI~0JI(j5PnNLF@VvRlGZ08=HOEX9*xBZ z0^bFx556X3VY2y}bF*#7x^X5vatsEl@uV^zXp>3J2Jx%u+*Dm_r?9_Zzmi9nAUo>K z?$=~B9{>7J!>$%La8=5>t&bQ#L1*%L`Sto^#Jzk;zhn`)F>dT+BFgMgin5KSYx5Vl zoa&1@1+uTqx4)|7Vf9gkIcD6GNo6PJxq;$SW?!{`o zIR_&V)>G&1o%RnuCyjfxj+Dee-4G!iZ%udsHY~(Hw;+JrpUy5Ots`}Cp!F)gLsEG7 zP^tX6ch2W^`hdppOqZO7xu0GyLe5NMpL5Cgyr2K*JG^mY(XH=T4(KX)qQlct zYMDZ-lRWvuOfuUgV`{`Rqa0w$#qE4Jp2BfJ-1P8f%@eP5BX zCrN~^`Rmrn)!QJz;sEx`+uWYQK;wHE^RA`5X}0KrP_h=AkEuwW;t#nlntB5Asg?)Z zw&@7cJRtez+1HCVH${DPf)bDEdHgL%SfgjxJGId{I7Rs^Dqz zBftFyPGJ&B=0!Q}bdjiXJec~T@K2)D^PG2J2aur7;e{?s%HC`GFd+NobvuSL3qCYQ z>dVHeUogRl4K|pYmCvUWnZMWC{$Ll=Fk~^Vc`42$ER$G-)fG z9RQqfW)`kFfKEu!@k)(zr7*bY5Cfy*3Vt`>U+~%y8FsyFuc>;Gyhr$*L!!72Le?aF z#2vtK@m^Uk>2n2M4PNzrc$;yoI-?0?fC>hYs#SD|Y^=%1iulqnsp$OAzrE;(=yZ6O zxA;X)i|4P5ygEJLh*kc%4bq-oc5*I~;3s8hN&3~vALmu9W*PJ;JPDbRQB3yeB4A5L z@{;^G4qPW{Pw%6iN>_VPE+sHHRH`~=6d6aZ1&GB9Xxy*1BV;__|Cng3$vmey2$f)`)Fqv-bqAFR^;D=I&=^lq|hX zqEa8w=58jPXR7p5!MI(Oow+<^E^KzX@0aqo^&Cp)6LYKXUQX^bJ!cPXqccef%jk6a z`Cw&c^z4D;;GGg4wgerJROZjw8{a5^J!m$OEFe&#Ebue+;n<5?8r9X!TCcY-M0R^3 zV?GKbDCc&$kt|-yN$)JkunsLnzg~aY{~NNDtw6=)8Zi zqv&~_iaoQ<=t54_wnm8-l@G;F?~~j$arFYzo9K@dw7Iu*fGS?ZlRfC=uwpjZq+S~C6cCwJolqh74_`VOmf;_abXuZ%|U^WgTe3izdah~wI0ePOwCu)z~5N+}V2BbtH5ef`^)DCKKypW)H& z(Y|k!GS6v77VQB)Sk48A!DY?Y?!gG+2w|B+r4Duv6570DZ({alCp2ApfjfX-&&PN< zm?Ft@-%_^Q=f{&VH<$g>Vx=3GLnDQSEe9d|EcI)3%e7mECtc z$ZBic%~8)B+>u9nd((3z=$g-KY0%7vM9OAra>6$+o(!&<2(#Czmgpv(w(Pa8MGdCs z7#a+09Egraqpe6BM1>LL1VP@|tB>4o40_A@aKFEC@nww+{GsQ7$ydK2zjMUxjL7!F z=xZz{Q+xjKx}IP5Z>=5BV)euO;Xpb1W)hI~8l!={HE5!GNylkqJhv_i(dYh3LH06Z&0C z*~wCcurJh~W`-!5cQba5Ully|@IKA3nVLv^RZud`)bU;AUi3;C(#>12d2+sH|JiQw znu&K+%~(Lf>C_fL*Jnnr3R{F%&m5)pJff4bB;kbOn`rFm<44GLXLo&fy@AnbL{~3z zf+}iVM|6JsM1+2->8Y5X3fXXJWRkZN>7ceBIEkA?*6)SMs$7F8mjFT1}v9 z8Jyo2fz{oeEZ)ys)+=UWaP`Sa6W!=d68jB)T&t%zM&isNVzHj&y#$dsY~1h$5W)## zV5+Vx4Lm1vhbvjnT_646>AkheXe&GZ zQizH2{63uGFOP?BBR#lHJh3zjWq{5~geD1xU)HIiWUO~X|Ab)LzV^vdHl6m0&vHAz z+Pz)DR9|tf z2+L@628qpv_>FX_jrY*8@%@SrMUs*svn^#_?)rfG28|4^nYrWGcLP=xRCz~iV~o~N zJ;0jbt=N!AQGbf>F~WIx^w`5)Q8M-Tm4i+X^AdhIUhpC8UI~wpYY9K^5glLQS0_}e zmNu}cUo+U`GyL*;5;Eb@rtggI=-G9PAls2h`(dg%dPZ7nsp}!cT9TiDt|!0Lh{QC@ zeL})@W3ut#i)nJ>*6$|3(=$=9P-Xxl7&3w(u29-H;5%X39QF6A4k+XMt@Dcx$kcn( zwRvMI25C;0SO4gHgsan4E$+4Z!rdnPZzZbu6maWx=Boj}4DuW}8hQajzL@0R$qq%c z*8u%n)2f=w-OdITV+Pcp5#2-Brk+_C!WXKR}RnS>&J%4pS z$iBn0;UZelV3r`<^v9<~b_W{TLz=dR&2(G22Iq==VG^{U6(nBq@{%siY}s;n*T&~g zn#i=8{4uIeoz=_KpAtRy5k`%+H%dl!5ZZ;PD>vd6kc`=}Q{(x6Z_^iP*MO-i*auY?~0Ms_a-ibtn&9N3$Sn@i|qPiyqD zg+-1CG%Lm?UTsa4o|;e4)QmMB@{RJdd#z>@I}g3GInFWn-o3m@7>GP6zwx&}yS5X= z|3?mPa$Z2$=N&A*LGqFu&?cTE3r>f^W4eU*zJK#ve@4BGoU=uyRV;sOFE=xz8jG1kZW{$c{FJsWx@d^Aq6=OGFZ$jA%6N559fnH@r_dVz15Sc6Ze`u z_c7gx!-13C&;N*>7l;DYdE#337x6*cN|QK;oz0eC`O$r#D6!aJzh>iFO;6SA4cp8p z?@|)=Bg&b-=;PNC_5c@b$a7DQ&JGlCj&Rr;BsQ$ESMN`qum|n*=Y#wc=RitlL_&*u zjp!Iy0_Akd049kC9_p`%S{%$!f`V}nGI9x|a?5}X;Pie`fR=$iDTsqiFrKl{V$|o@ z(jYSPg`i^rN{NDoaB4v&4l93XC~8Q%JQ$*_HgNRv??nwszz{R>VXL!gH^C2@NPN7r zU5(NJI9`5|$a)rpXo7C&>?|lY0iP}b%Qn}1it5bA!U+L_Q0Acz3Jy7V0ItOH*0vgd zr^xswK=RIEx#}Zob0I2Va~t2f1yK+&L<7?Qh8P{W6l&hNSiq6r8DHR|oWzFl&=9qU zZXBqEI$nbzInUjNQQv631V6|#R@^?LIzRA@5@zZb)R}2ekq1%KH`h6#Su8}5Y-fjN zlmNSi-h*Ne{UAbj0MwrXi~A99jHVR{L;Q4qzaA6901K%&xxIrlN49~wG&WfDuRxh zU)JyfL?-8ke^cKl@Ix%GWfd0+WOL8K9)vyw*hiwa*#n_QCz#v-HG~AhgaqT8$g`6u zgCSq#B*-qKhQz)BLmsE+yP%>IwT@5*$gBMKC~(mD0vNVmgi<*FTF5#9d@?ZXnudZP z0U0#OP=?UmvoSdktzCI8h{8-n?F=++VkRw$R|*8cfi#(NtJF|iagG49T&dgZMJ)su z0wnCOB4LZlXY-YUSOYWj0R>9L#AyMHqmqB2?fHJ714GnXe||*~8>a~v!gFDD7PS|I z68IrIes%nek)wcL@fLYl0Noj`q=8Dj`n*{bwN6B>Ecl3iV?X`u&~Z3mcdd3eiBS93 zi~{?Am{#F(M)#r6{gbTh)fw|-LKxC`?8bvSVh0F|Ji%TJv<3^n9W zC{RN2i)ZUqpHO?rz5^)Bsdq#$21N>xP+{!q!~4B|0kHq2h_X9pC<;{qA6ctYF`pf} z1mc9HWg)e*G0cVsI4C&M_>zOIy+(V$fs)>}y^t$zBz30=1==_!u*z2S)h4OGr|^}6 z2*HBI%mcL(1O!Ozcx!1L)DV6M`b=0xH&L@79)Th43Dx6g5#n!<0K8uv0Oi>JMGUMU z97m0g`s;du6Zk~ed_EopHCaPz@QX>x=I`W*f-ry|f$OU%X$L%?Oki5qv;Y(cz)cYx zDB;ttGHSO&;$Rj*%n1b)Hx#S`$dZU)Rh{jG1d;`GsW0tOfmuU+2<=vFWE`mN#Rh{R z_ZZS|qw;KO|DcRp;G0~ta`!>h0TdzU%LBfv!6+mXkjVk`T&Frkts@4b2Oqs?m`*}{ zBWuVF;k~}+?incmpmiQbaAuPq`l!69|H3f&V3N68rlGG;ONWSomGSSj@|`j7A5i#% zd2_A+%I=r5T2n{ixfO`W*FtJ{e(!$?lB;*7uAty_ofC4D6{@iNjt5uEQ7aStr!_GEKCwwiFh@a+ z# zhSvdlBA*=p78#-7qgw>w-oFps3W*IInLB5C>JM7)hy!O1^>yb*f&Wj2sf6w*gV9|2 zsHNdBu(Fga!}J;R{sDzQnfDK!!w6~VO~uP7I00AAADU|jutSsdtS@S1S;HmpOG#%Z z&lwthV6k0)i(Vj};?Tg>$i;REP+!1rfFC|6>DVfwc)=BDsaqb!H{pWRtu<=W#0j~ZHl9) zSPZ5ML|(CzbK-BaqzMuTysWtYUu@|L;obaU5Ng^#nfni&gbhJ~o+du3Ya#+h;ar+)lT%J@^sJ^<4y1*E+Ht*0UZdZec|R8Z?&=Y;GaSs~=_ zLo-4elbho5nV$NS*8eaY{$Q9b$hg#3YkGnLe*q>~nS?ebGzG=He?s98=KVwG#6f0H zE6**-v;F_6xl91ZU%eJ{@i&*{f?q}i1pjj5{8CQ9pgfO4*#8Vlh|&HurT%A7LdxL} z=KWLW{Li5LpFs&p*gr2%f0zw_-oXB6P@@0OpagQ}PbK^R)1Z8^@IJ6S+;iDvtdZ29 z(!j^ob4j=2cFn=pvB^N{>Itu3ReXON1W3ric-mEW>r9<;a=|b)p}M<0C>>SshtK(P zhc4k#DC@Wao4w4i#_AY&KF#(NYPJ=|lx~mpjd|{kjoCL-4nldD?tLj#pbO%o-}Z(E zWQMh6u2r4c$+uufyDLk`gQU~X?p^NV-zYuSzl<$``kyPROTa5@zE{CG3QDv@hfj`b zPTlu>aXsEu%)C@6*Q(x(8QTKYo4<;Tyqe2VS8X43O~_cvN~_#Ev-AdlzXezR!xXpwo>2ofrp85^BI?h53r8{xVSjd*FH zjL(SLEa)cRAH7$7W^7*dbziG0y@c?z%C_g&CluWv92*_qEde#~%1c*-)@$umw)oB< zTJVQ=$~m|e*K7UR>NDzN!@ZZUDpRG=*jRUMX2`hrtJXa8-A|SbCQde}Pwia*DxGx< zZ6v>*qSi)W0CO`tqHX-ljN)Ok+v(SG;qaKmu7C<$-c;|dAy?O&ZC%@S-=4SnHY0>` zPx+puubmlXDd_vK$)oW-@bswH>ik0?-j8){r#1T_HA5#IUB`-hAUF2@QZ}WMI0wqT zP%8wbIHvb;vZL%LII*wbYFUbo(b(hm8zvKj6DMO6PS8%8MaKVen`4Fi2mb22RH&nX z#d2+HI}?&$r_cV>f3 z@L6g!G!&_FD#37tx9{#<^j^O>mVbJfk6YCNDrAO)cMHQ;kf2JuXtc_GuY39!b;AJH zA>e4q(NJ<-LwPoR4M6a6qe>%H4Xeq_is69jYIs zbm~*(33)||GVjG;A&^g3^Z>>)z-h+wXCmpwh~6uYTHT|6|t$x>!SA@5_Pmm73 z&msFG5CGPL;5W(RV0!Fi8?14MseFza9sJREu!uT?o-}C)Ni@%^}7zR01RE^1q|3XG>7M>whhS8%w@3%aQ-bnYM zdMe({I6W72K7KRp6VAGc7Nw~;Rin^D+my0{8Wc+Z)5Pq+@P%S7N65VzgICT8p4R3D z$KuU%e3BVR*X8z8{@Y%n9dF;9Zs;0W^$DEPR^tdT9!*{J@D@C}_?&OqYR}fJ${Rka zNRl)Q>iG6Y1dcA)I9xfpcye2rZ^iY*?(npD<1(S=8ToVrf%nj|>~D^bqlX2tOaK`% zK+-^3Lf`pmDP^~qPASX^;|)@P4*N(41oqGfN9|YKb|=k6kp)W+M4RzytV``J_KKHm z%HEDXnlD2l^-Q{V9xEY3tOgxaR;pE(o*}LF@b%fow?V|aKFl`!nxi6m`aY)Y+i#@# zLxaHz{MBov^JZ1Tj7K9&pH9t%n~O)6mnQCvZFq|wh&C4pTU*wDb(`oqnc5pH%3l+m z!Vk}AZkqQVC`#Dj9nV84n*w2gtkLQD;a|8fcz}*~`vTUT)&I$*LynskD zcb-AykN_bK3kN~9iux;pIqusj)c0SV#}z7oRAK)lwOuVJP@<dwDJ_U7+=_W?h?cS$*-fN}5IC*zvbWpER148jM zA>ULGeR)clK$hTk_nnGnk=?>E6RNGaRd7Om!U?>|D*_AeoUor#C?G?`nt7j6zy>)< zxn7(WiL{s-#8R~L8a3Mv2Hx*F`HP`^zXHKNf^#GMm&edIlTa*2f>Q!df~HP_8T;ts z4bMrIQT~!*A!7r0)D*ow2$hS8ieI7MSpAdc7#I zX^oGF-#KDwZNj4B@-q`x|&U01kP-12T)P($(OttkhJgFBmH zMoY&@$BauW{Kta<{kTkYyj~3}Gqx+vo5GvS15sL$OrEoB2CCnc>%RJWA8#}d9xp`P zxsrK@5kJ$o$6rIsFQ-~XL{cQug;3O0W)t^#Mq@wMF18XU$%BK$;_Y8Ke!d5g&(7ed zku@X*bMP&4UszO187|)59ZGvY$c|peYfM1sSoF!^CM~P!o6~CKPX6F*_0j6mql%Rh zR)JiphL}Q3npK($jodUZ-eHP^+<=?!rs!u1s6>y7^2fRaE`Pc?5TsX;urrWO+vL1d zG|uk##%7?X01D1ODGWCu@Twy-cOUxf_gUsz2)sKN7ah- z>&Y{x2B~mH)8>}Ccp$87Mk&5D^98&QEWde=nL{*Gy; z05$L;dDnhZ8oDp>HVXbm3+KHggLq(;lG;TSL}B(N$iN^}@9OR&@7f+tg$$`8=zOcK^m zbR8Ce4(dz$+A2bPH3nWcbY9yqRGTL_X}W-NGZ)2=D(M)pIofe{)Egl-13YM=uKcEQ z-{0*lE5!4(bR8T*hx(_}O5atK0j@|#H_!=0bp{QSa;9Q!@2(UAkK{gkK#J#J;E9ZGE43N~@ z1{JnL0`s^>97oghI$+t_V?2wk@vnIfr1hOYPX}wJcPn>A+nyopfJ}TG=+qV(Ek68e zrNkq+@sWqX-QCvJ{aCbsa*$NXH013l~X_ z|0P+irV z&WW)?PasqgD(u96c_5l0BqM8_;4lA}wSPEmQ_=q_DQzb4)591W*6-SUrT2S_5~!Aa zv6hv8*$;76u;|P}_DAhMN(h5C<5r@Zb5({F$yie8BkZqCrWFM}7DBw@2) zTrtdq z2uU~)`%@p{xngVEKv9gmDeqH4V(pTRqnHtO>y1oinJ30x%#4gOe#Vp=&kvHJ@~~Bq zAFYFkRTql)ex^UFsTlk&THbs~jbG_QF$Xg(&4zaaJ!{^I1mE@|VSy^8-6BcrSK zzYc_8hV`N0W#Q!`xVO1RG4*n$Wh6tw4W+A+b9Gad#VMZ&NY*`1TAy~XfwwHIOEBv` z*!z?&oC|J5)8+KxLLaX*r$74g^jxfe7<}=vy^Rd~A2gyFv)Av`+Dz)^jk`ZVr&+XjBJF*&hOuqS@n?&{d+1;N`(-*MSvD)-4tcw4$e zSiSLMM~u+O-AEN!N=C6vhUY|gXUWi6&SJyQuu@mWnGo>O5u-5j-JwqV)@nGW9)4Tz zByU@nH-2{pt|>W1eZp9XDXFG@hv0Cq9{DBmYz~iI)r;?=vJJYX+3($cxMtB^z3V*) zZT9b&HxI4YEY1eYz+Li08HUJUYGVyn&!5Cf92TR~@lI|++T2NfijRyXNQbN3T~3~= zvU+$Do_c)6oP5quzm;(=?eXAYj=A`{v9K4LpAz%og}?1^gbpma7(W(~j^vg#oT`EQ ze&G$9&P=)6{6(IPj%;b=<&`o}y4ja`iJ-Ic2w|e8)k<`J|L22{MqZLh$G44#Z_05% ziWj9VepLpxD#z5Xb}+t1fS&^*?SW}sK#SIJoz)aSYTe{x_&yPBa4^jgPYws2ChH?` z(VWk2MFx<*eRf_MYGWd&;@0bx)D>2&hR(#OVEX$s;A3`LxKN~n1ko@rN1 zr#^aK&Og}5Q!(f@R%g80B4~%SAOcm`QZ-%XD7$l!0^BH%JTK={y1kOe7xn>f?}$k~ z2+ojf4Az%qRofV)l5r#D*FvST_@I|4_y}mt9UBWfIDF_NsY#e%OD5alhmYac;#DI) zhV4rUKhjS(#4Jr0nq?WwyI=834LZpQvpe|SQ%R{FaK@89mLGd@con-D^tAGAgffuI zEX_L@0HEF0j4w1T}9 z38n(Y8-GtswVzdYa_Szk!ZbZPhOquo6-*L;&PhDrh8dqpz|9f~5|VBFCcOGySGh4x zV{Q2I20yO?$Z}S@@F7HY!rptpci&*aBg6Rlhf4i zhY9ZvJ(QAK)179zDi0ai2NY$K0{JubeH=r`ko6;gP2^-fY2#+$? ziq%B#x-=MvbSlHww7H2JY)Z#UrX8Np^6up38gqqD7%X(BPIEihPO7C#&O9UyWZq4> zQp!+J1or71Tch==HWss+M;ha4@cEeMaenj&snF!9NG!=rA4B%U@!Opy8Pt=@pHk{J zmd!;gF5OcnJA6t+d?_b4Bl-tk?8gRmp{fUd#8%n+f}Y;`zwg0tIyCrhCflQ%wptB^ z`+ZV{+sV~(o-ySQ8`$B; zuDv2-QnPsy;Dfl!;6^=cx62CBoz6)RN~m{yHo+ivc*IIb@l1pG&M|jKc-uJ#m9NhP zw>o$#ro#juebu|B8VsYtW(ioKWv7*pvXc+jc5CJ)iU@{BP8E ztmm-ls~4ULa@0ZHj&)c}>KiP=YTGwImC!#z$al~khQO)<*9PFh&Nt|o< zxF4h)WKei0FSGg1)xd;!AirK^+AHIZ%#Y+KvdAS!`i(4L2>6UX>5dN#S!?mLObs?- zzXWhBCC4(&KZwpugP)GzHgcOwtMhmbi+WnJ!Xlqf;LmZ|)kdpx-?2xkm?_BqRC`}Q zVJ@jInAQLK1~4FCJYF5CHzdtt9_#UOZ5DlX5j>{twDQ%98=D z+$LKrMD{^S_=zJ%-b61Im0O=;#G>T;uYhs8#5w%zgh24JPUuSuDbCebrv)>{iNlI3 zg~0d7`k2=FYcDM#0kj&`_$1MgTu+`2TNy*htw5d}BWfNFU>D#hn z-L*5qZ`4vl+{^eNRuAU~W#>%&w+8jTtqd2Lo9%6au05tUA6syyP3}3)y_SebCS^2` zd0)y{#kW1uU4*Ku`rCEcMDH^smF^^vM=m9P%@?1T{87+;<7vE|NbG$@{JQ2Cv)M#Y z{O@d@DuhWaKREm?nWHHZsOe8`N>=Xm}FK`8$;N|EL8oL=VNB_I8 zOYJgNH=44{?I~g}#`iKq#>g9foah__Ji{IyVWy7FH!14|puT4b_D$qtz@u2sTPT!F zlJMgsQ+$EQ<;m1dwT6OHC}F7rUD$gSi}On2Qy4ZYiK34-*_@0~{jo-TI%`@ZNJ6J| zvqz><&eeLn_mjJJz>1|%eeW?;@W8cB#QK@nUqNqR+Pntx5uBx^T?fU%nc2-W5Jn;=c=u z3ab%ewT*AfL0`nNzSY1IEIN`Dp*YbHe5#Ig;=+(f+EAF0XfwA^&ri9uSc(fmP}LO| zjcdy>p5ZY+9g(J5+%lcADp^6&bCT8ta)(az5S;UG)7+sBJs4RY$k0FuLJiw$i7Ln8q%c`Q;`&mcMI6ya;Cs$<6gY> zsO){k@n(DaZmaIxTjRHU2Ac}u=a2B-wdCNZEF{RPUAX19CH#9JI!!n@2!2yK~FwdglP-5>Nz84=#hkS->swW+En3qExr*zAk-bJ;ja|N0{cJ#}D zZ_^mEhj2E=*##W;e5-vt^_)qwNSjHC`ae6n)hZJj`dvyUPPs=G5L8TJOj1vHz(Ge! zqF=b%hLmfh%+A%SoFaDGe8Lzh3iM&N>N#VQl`6%Q-ovQ!l1VrelbcnU{#)bb<&^y>T7IE^$X7cZv;(W+QP<7}#0`TD^nvetXY^7vBSnMkOI<`=J+5fw)r*FlwHCy? zCKRTUp(^&wDNp^m*7YHxec!=7vXvI=)<;3gK9FUK_v9 zBLmN9fpZy21sT70Ip;7CS2A^0KK!>Q9Zrwtmy0~s!BLQ0s59R;LT3l9F+jJVehvoZ zU<|cFp%qJ&C5tz-#$1=N=}qto*W(pJ@@cqVV5gWkqSDFT+d*?=iGK6tTXUHY{FduO ziJg*WLA+-oww~H?is6;sN8U&Q5_KHxP)>h@lynPXIpg$Mon6wu;6vl9!BpE^h*xM) z?>T*KB4b5BEErnnM)!Zl820rF~mz%IbeHjB)d1 zk_B>n4LOihar(HQE_(8_{~@+28~sBIQsdfpx|acuWk+@rRw(3TwWRJ}`T>ntUChqU zi^}1VqanX~#n+19iufXE(2#%UTo>PTXjW5_^N6KW9;<(r93C&uSqR`SSvg8CCK&Pg zg>Ci0T=`7)Mrdt(w28j)NOCi!i(Uc^66$heAS&5sd_9bAAXT=@g%2b+Cx{4A3Bq$Z zFPq}rlow$95c)>izRO%Y-ovvZ>^X+P*6YE(=l$`e!L9xA`?xY@QUQLwVXYbQ-<&Po z|1G0wd|UCv5XNyv(hJ9;yQ}Bhsym&6Pv1 zVck+X3T83n(@MXx_;!Tpaa&@M2;)=DOffxES#$zlNgDir|ASCQFMCa^Ho`1HQ;w&n zc;&{FltPG8*RXk#@Xq=4);CbYITz5>C!JNQyNynFWWVz<)g=NogXr6EI84s)hV_88 zOF3aYy`{uMoU|xqk z7o+?4KLztNJ}t)mTcDv>$zksl5B9MC4Be}yhdVb*5oj;4zscshf1jorYh_B(qiiVJ z6u0R4kt`(6Tl}29{xJ;z%j21EY4_yV+OqK9w>{74rGfg?Y4~q}re#*8syt%oL_3Ps zyvA(@;Bu&&kVT(=g?}bu6>4LAxjs#Vj#~M@!Dgr*Gva=0md8WSMYW`?H(Pc@UG|_s z7jZFof5AadYMZWZ?s%ljV#jH*AT9bG_NIa&nuWG*9?6g7T-3Y%^kF zC3xutpZIr5^y|R+YI)~p<>|?q0o!S$4At-$=r5Z4T0?a*1;0Q=)BVo++?_w56M?y` zin*Rid`+T#Lkn~oE?*R8z{ws@0w&G1I=`GYbCOwXc`^@k+5!YFrRR2!M%pa1NWC!~ z0F3cmd)mDZF9WBY9>&M=m>N)AA26fy?u|-RXDimAR^PYFQO4K;{rOx45ED3M2iE_A zmtR<_rGa_e97PcMe|r-0!gJvc_2T$Wo{JOp@sEoalN{_{XVMPug2K}!DB`;ayIAPd zEq<}z!A~ooQH%yFr?DkaDgEhH23@1q>J(Fu)!vUk4n4c|r7T^;FwG}D(e0tRl$e?v4Fac3I3JvM{+2NBbXa^EVSA9-#AinnyNVdVY(* zmHdnUP&sJbmt{(DId2F3<0dZh2d*>ITRVxEaQ>&IvgE_@8utQL!~sZFzPNs~71zO{ zD5TFHLRQZUul?qpUF`c- zuh(H`e;`=NfjaQUd{4$BNPGTZMH|+;2(m@883Un9tH!Z~9Q@LC@{$X_*9%H641R8U zh#}rN2s+>44hBmr5x84xT!^t1MpNAx1IZQr;$!z$AHQn5%7gLAC_PWBUO`NCsqE(psf??hdC{)Kji@Q#ubB zje(SdnulL+l4`m09+Huxm1aHeQ(Ae#Ml@dCy=55f=3sir$PV=x@G^pDwusD{eg5 z9nH@RwgX0_rnjadMkcq}zvcNq{@S6for%!syC}NK&SQWoBsdaW8uYPV1^B^Cr{(L71d4l6yrx9R&Ny)eFfXSnnO6iI2^>_%le#2*(zR|IAT7mv5P*77Ph+sZ@J`|*JL{;ftn zwG6@@rx~kaR?zakwi2FYW0FS2=9jFN&FlYt3$p0Eq35{SG~vfIMDj zmxBn3uUkz~zWjNRoaXZwG;ID1YZFCr+VI76WdA?>h~`e4^o`Ys30HQy73((fU#vZraJ{}K$J=K z=JWBh_f^2RP|@Wv$CL1ZAGC%)byTXn)6h>(wGou1|0G>-EVBZLj1j2%yZ>PHb>No7 zf0OBu0$%2W>7n0$+{&QgIt!IMcY?CxaxvXEoSQW_6bYwm1}o=ojXk5VUiVzP~>msYLV2zIb()U6%ntu2P} z&&8F${L%Kz_21h5AAe!LI7e6yHJ8np^C#7_6L*5yn%)t~XTr!P1>sPY=Rde_? z`Ln-h%ep84LNNxzB-E00FU+MfMR2WhD@zF;C7w+zNk&(|XL$NE3Hfue`|IXw2)WrR zY#6+E{;%W1FT$dA;2mb5Ygu*ghWiPV_a@Vr0dm$L@6bo&w%xO9-`^EIoCN)Rze758 ztki3%)EVT*EM%GQ&O}`=n2S5=z=+g<`dlwH?cS0oNW1j-8a&+<+}8$(vGz+xTqZm# zuIK9bZStXLbEpV!zv*^$kLM(|!2-yJ+jBBFS=7p2Q+TFT^pV{80vekNzKEji;}Q38 z_wOkcKrv}BQMfQl9SoSt`~cYbpR6MyI6)B-VVt4(( zkglK>^B9tvdtIl;UB$SnrxV+0x1B&ayLSa^qo=609##Ud!~1h0Qe3ONzcMgqCLX96z7-qVoeHz`jszk4vx?eN`v0y-F1ht-Q%0@ zeQEGJs*w#U$CCGpEh^UA_Hv|aFZJq495>k>OP|zpZH0mE(pEb!Tw^gif@<0HH-cB8 zRJo9=pi&w4n1Fds;%~ynBrJ%E;-OX4apYN)y2Kl*TUR1YPjT(qC{j zDm8JlYg~N6mM&am;xijU=!=F9MKUyDy74v8l?fbkVhK;rW?>5xyvE?&H#RS_mx(Og zo2h2P$|;G?*@`oKWY*Vqp2KS+1SIT?P2_UL33KiDl?61#^rJf}gMhmC9^m7ZvGdgX zwTo|U67oAn34Lv(#Q42Hs;O3cDe*0;b)SpsYJ-puuwwY@BVIL~!>4h9$o=rsGWX4> zB%yCxu4)h~x?)<~w)iY5npHL2A&D!#H0ZT5SZ;9AZ;)#UZebkDP`^a>oRVj2r=a!t zu<$I4O6Juu(d|2;>a8y!M|URnGY`&3Q>2od`cvFS(OdVzK7{v z!o)!Wf2{5z!RMmqry(HuQYUO2WE?p|IlN28DnW@-xVBH~ua#3HC#mxYVc&X#5xtLk zDEL7e_^c#@m4h;Y)2F1B?~XE)4l^x#22P=*N0&X2Nt&(L`12OI70*e7BFEP^1it+A zK5Y99Od>r%7T7L9xv}M*kdSv(h$y~Ugn|5M@c(G~3csfRH`*^SaUdY08wN;ssvt4C zyK6`daHN!UmxM?o(w)Kpr9sNkEhD8Qq`U9<-FyE5Ui3s*u+`iIMY&R0-Z;y;h)cZYdw ztZlM7<6XYoHM+!T@

s7W4U!Wngt!>)fy$o-`C(JtFyVyN{#c4Y|OQJjBXw^vb|G z5nCS}y;~Ud@C*DJaJ~IMB9VmZ2O0sn|1uduJfLO9J};`Kq7@{s$5V!ku`zca9SyiQ5;l;Lf>F(OdD$J^k~tc>uz})&ItY0C;GB5@~*GTxgzGGu`2W2 z=2E0b_=>ZANlv|vC>Fk6`tx!Y`%Q#75`5vhJM?;Dje$dsdWhjnK1{0I>m9g zHhFEyhG=ddmAFKPgPQ*d#6@=!fID*v15Sdl2uizg)yB{=E(e)1{x0h?`pwrh0cRBz zjY5zuG;X-+KJUk_Moh{FqIL{=XaSRiO9W@;3TqQqUcamV2mX zt|faDnab^d#*D(!z3^|nnYx7l>3 z&#~WieoR+6HEjBZb_)C%{SZ#$F=(<>m&w^3j#s4Cfo}A1{1}DDzHBp~wVi6$oP_z4&-9M1ajTyyhP6Q z-*?g9N>8Vw6tzD|>EHQWv@DzbA~t@$eDY?w{!{63h8m+3mWsp}%VUAD!&QGoG?^p! z9KCx!5ZGKdJ_}~3g+V8VoJj~p;E@OSDyk< z>Dg*9)}EkKCO(S^KBd}x<0Jg3llfvWRtmjMvb-{mWuY0Mv03fdziT~EwH#k`{(`ei zj-I~D@l45_@Il!H?FXK65?W?Ndl;sG9T2=G0oe$_hB>iEMd0&(j zG0CG4ds*po_-i^e7&Wk5>L%Z=E0W_#NF&A2w=g9eXGUjiShOEgR>`O8#avyk+Yx+f zotGgp+#RqCU49I^n8rg|P5H^_rS6?!$h{%ktrhTA4+zGv7q;u^dJNm6)3QD*`FsN^^gb>WI+>>QpN-kkgU%=&}yi~HRz~@I2+qFuO zaHhNm|2?r)F-r<>b>GOp!$K-eV|{YH{=Nbk25|_oxT2;xANGh4yGfF?20}s!)ZhM>s6wq5p~N6aBmI zZ-|#+%YDzuv*xVt17M}g8$4dy0qBPl5uZUIak-M>9OtF7CeDgcM9w3U*W2rpg&ux5 zY0`l5E7?@{kO;5tSuu_b90d3+y&}oSYjAC_CoQS{>fKw$72SWLmHI^i)k}GR!?j9# zq<*3LY?Zo)l<-}lC@%XOF4X8tyu3Rm)Q#%MR5k!zg;Re5{G{o30HiO=ex?G zn7NOqRyaKfD{+p0FLxuzzBoZ`_>||)Tl51RVCE47mA@DyPY5r+{7Hp75e-}Q&de!y zsc=KnW}bik@_lv0;erDk)08lO9RIW<3>jj4_9&#Emc2p`NE`{T8@4iCM0eJKJBvEe z<1VHK-EuT0>n}n?*iqx;ZCi34Nls0?wuu1!_50+uE}?p?dn(2)8V?T4IL@7Kmiv5; zC6ulk5+Fla0D35ckuBXGTJ2b7TfVxJ*_9rgH5{}V1fj~xD=6zy^+GYIg+4q&6bJ3& zVZus~5#1>wA4YT^wKI$s3 zIBi+1w)iDR?)2Y9azHt$@`)K;Ov5n*vzkWN-A{_q&nMEfo_^6s9$pGGJ8Y3Sl3UY+)O^6yD@r5~cZ zXxgY`;}7EE`5o_bX!yX2C=-$-xbsF;s`H~HMgX9&|Dr~8jB1rw>z<-A()n{|l7na! zaj`4*=4AV{8*-JH*K|YU^p#7j04I(+40P$^`{(}-O51dbp~Y&V@)OlJBE?}7 zYk`~6`Y%F1=YHWxLtz07l7j<`XMB08kU=5?H~z7n12wJjEAq!Ra} zs8z-zbfSAloig_by+}1SR9aT%X&ljkhZD49;PEA-m3_*+PsACK;gORB@Tx=5Hgq>@ z>9iQd+UWmB9>{fOY|6qmC7jtf`eCNtw%Ywu`vNOwY)sI(g-LbXj8mU-_I3Z3SW$Lm zSK*;p`Hl}=bOD|ros2Y; zid2b|s>d08BzR%?AMmTWZT7tvIt^$g(7_7jno@kV;#I-U)QNaQF?fE!CtrdIa7vFp zb`kX#VZTN2&{6}46G6g{HqHyi;p$jARLq}s{~6FxxVzk)%^FMuz2;xVQG%|_JP)%D z)yIOj+g?Lz-goO!ft-Lu-Lslfrr@VN{q5$ad(ip^yg$!a!hX&#?tAJrEDNSSJty#m zs8J`%tM;xJjd-;|Jg5yGD3lbWl9mCGWn~UJ;VB= zI;T*}ULh@Ydf;B|RF3tS^R>I|C&V@{y+Txe?STgdt7AsA?$7dKG990Y2A+sIemQ2q zwp~ME^pr%u=)NUnkAN`|g6I7trN`RfnrV2zVN+ayvbD(2hBQK9Pq4jydNoFTGJqZH zkg_~$hY&A>BGF66x~$RQD`gZTCSe#Nb^YSIls7v3meU_KmgiPb{p&Box69g$86L1! zveFfU6TELOyp#sFpVR>NIMvmmc2>f z%a@$>hg#>!CPz&e1W`r5D|yQCwdks5)EtX@Tm5Ie*c1+DmQYiAd-BxDZ@zEUua|tk zv31zHo4E1{yzmlkL&fI3UJHPS9APA#^auRwFFz<5j=$P?dLruo?fPv#&DGTNiPJBC z#Y+Ne9#*7WrkmKAP$!KR=*U3HlsC+&=U-kewhpYs5gG1$JTBQwmaj0WV1i422VttT zM6Ostig>pA1NUYF;0_skYhV6GcM32_74vUihg)2RzaD=vyz}`NX0Xo1@@3qzAvk8Y z*X;Z=U?v%F8myxx><+65+LFZqVB=$_gc)E3<3)Em$w0+$xDN7hEDzs$YViTV3>17+ zew6F!{(L_j|5HfkBXpBYvR0fNsAT>#Jw)?Ij!%>2Wu~<7f5I*oPN51tu@_W|y8xh| zOc=h0M<)FMIUJ*|51P#&wHuZD*7hrwPFK)k?gyBQv2*&dj8#zX(PLyCXUI2a7E|vg zZ=A4NFbUNZ)ynf%l`K)?@AJ-iDKsZs_ZQ(?gQ)2nV0L-q?1nFgAUZPm2Jp=C-Hza$C6k`-S=LAECKWo>s zA1nit|4U~2Nq1grydQDA?BD)r6o1=*EVNeB=@e&4#&9-)!iC{Ikn6EcK^mkke$<01 ziQ){o0SZ2%<((LxYT}*{P5k^|+vj<@bi{p+2^jiyq1bxV`Q}$5R20YBb7{Ons<0?o zb3#lY2`1$H5t2LJos39e1KvSC3dok~SinGp@}FQsd-gy!Rzz7M_5iIR=hIdB=5acl zKXO+^Kl1R(mLKyVs&ejGRhg_)hJo0Q6E&^gvx$6V>Jv#?vwsGLayqB5d$1?TDA=E( zi0!qOrme1`7e1$vJ@7t^i!wNzj+bGZ$Gv|J7T2*>>5 zr8VuJPYD~x?(9A~9#VC@E)QVS_jy_3gjK0Ne6Yi`u1m0KS$*kA!FP;{y02u#H*wfJqPz1RTZkoe~bV0;bq)V*K*dgzhP0bOHnm@U31Q+Xe-f+*xQ@x^l|@N0mQsADs%+@$ybb5|vm~8K z9~l$k`)r%&0JY__@3b1*GY?7X%6P3*S|B$@fB7R}Og2au3b7{Kf^|{kGAaSJu)_eh zp)%{i(xlx%gwD0o2?oufoe+jAmQk<-ROTgvx3~HN_B^&W;8z?TL$9U2$p^oZ#;7r; zMERFN30~oR_|(fXh75ie4I;*tME2pWNg~~u-y8fVR10m<1_k|nEe&}MRX0(P0Pz7Cv3C?d`2aAt9|{qyA+HpTKBSLl^Cy5BdL6c`Rh1VwaKiMeR#Hi=-$2D;#W=BrUQC;7U zYZkh@%`>#0122()VB==^PV~)&6*XZ!z-yDFR`k1qKAWgZ;)6?~#Ig)qM3)7+1~E7e zn&~4vq5`CPPqW)7WPoa}O9?tr7nCq?G$BsaVTl^UKqE88o|N=!gfKGZwajHC?^^~; znx2HP+Jn#U?iKC_WHRZ2?8NF((;^7;UrAPz4RQukAT^d5XM`_i!} ziDh3IEG1)jC^v6(y#95nlrhUgv{W_l!g)TIYe*S-rJkuZ;9zXM(^B^$?kR~IW&|?s z9OLSb@5ZigQrxXTm;`1fz01h(l8^d*3c zK!Y`ojibnbtw@J^#S%}}kjs@eN=9m)QM18B27_*55~uWUgtHMmrntwetpqUMRXVq{uaDdLJRWQ{>66_?Duq3B?aLM_BC?zGq@e>6K7& z>;_ytf9`7EvR-h_W04~5!)gvFIp(}qKu0k)+&YM1kQJ9oaN#jY_k#D$uf6WYoO?%bUXG?{$)?hUkbCKk5<`@T54uU08;ARDOywQtNU8T zJkS9CDCJQo3Z*%*HdhCDKE;$fq)Wke(Z&AJa7!K>po<+=AaMBQ3aeKO+pK^Q?_-RM z*{%*KVn*B|Y(Z<>m?90%EPPx8d)!(VY=LhqIYup~G_r?n_MtD{?nyr+DTE3Uf0U@ed>r^tlhf#yJ%7XK9m zsdiz2MuuKly1K6hI7;4TrERm3(KHSTb<+jJXn{0mZOW2!spi;-$x`4;nWh@k(5B<^R$n`S zjj9nRWUCaYCX9~`aJ(*sl}`J`++6#*dh7cg*W)i0v@2)18DpOoFpeRbDqrz*AC(U6 z&)UU6#kBv%_Kj!W;DQ7>LM+rOzE96X2;yQAe{3#&U&+YN)>_k-pQ2tCCqYn#tCnA| zsmT4NYlg2jFrLQa{xDoNh)UA}6Jnp#Cn5?Mb*CGZGGm~^98?+wZ}o*5Xi}A`wI~_l zV-8IQeaCqyQ=~V_Tj9j)H8GyDZRAeJ@^?E=_Y}dm1}MgD%4UuG!M~8h>F>Cfd>mm@ zrHR~DleyU~=(H($InIK{xCu?bRREELzn*g}i%eV#p+P}DzX->sY5_#3`bS6Gidci> z#0>Uc{Q0ujvpE5dO&ul*CpotD!BUvev=_T`Bjj>Q!fN8C(8y%WdGJ{WmgB;y2qbFFIr@r4Alu2}9a$PPSs-8ALC zSM8@8fg7wdXbujNf1I}Ke`^snXRLmfbzDYBUjQM%>*HON6Li*%T1EAz{ff~7Yb&6V z^8V|uOPVPqR3=S>u*AoBj+Lr@thE0r&U+8pnswGqi0hb+YrK26ycTfx^%B=dj3^MS z?D)0ug1umz?y<>Q@U~g}c=2)l-8-e7Z4oa8)$@HbxD!Z-< ze_6f8?{5ZHnV;WrJ(b+~crGs{z%%gtAx&vq3#_aVzEH83+02_Dz>L{Ki{YVSV)rkf z9wurfv9`TPF|=+=5gH`jGLk=}IE|@g%dL+a5qv2Ax`w}IWH1{266RVi@L&!~6%YN6 zM-IIu#YI*=3{jjutsBArilac~j`csSCR5=lf`PVoC||;k!CZIg=~E+V=(Z5@-%%+{ z`(rf+Wfck`VZoYTjI%W-0&J zGTA)pu(kzMD7tf?!Lm{jZ1rJ-0tMpre7j=3^Vs(2wsX#pB>70s3k=htMzqxv8Upt7rv*7|EroeL zJ4LnF+bp^FzO&aAQr&e|V#CklN<9Q z<<8a`FQz2LPgZUnW$dg_kU1-RQA?OAcVlyk?q@;Y(#i!@XoNZ5$&QavXA4Ui_Feq! zut-?zxL;~YUx*SAt6n6rBQ%t=g6Sgi9NIZ8)_@-lj@wYq=tPAch$MU@5 z)0HTg!%0?B8>bA`4J8WJLpE0EsIeFS~_z#5;k3UpiQrr|HTMRpD~ znuyqvjENX1g>}!~rqb^OyGpMs+|Y6dZ)Du}s4i4|EK?cq-B6g*ANBkD-EMUb&7!p6 zV8C5mKWop|rB-OpezftlLaU)vtEt4P-VgYhkN~mtR7>W5Jfc1YXbVoI)gTv4i|bGNMxsGVxKwViv}5fhJJq~%6M(FWs&JNz$|B0zI8%r*(I41tg!QNH^1liSri2)u zjdBIM7q#mt&IRmD9qaRz(m!|uKJB;zb;9&38dd_0jkS0|v%B^RkrVu~&|wp-XWw#1 z$O41U<`G86?`eWxC6DAeatA+gq*4nzv=7!#X%CwrgR^Av?c$@Ldv)U9?3+HyOuv%N zJS|mWF}TTlQM(Wc81HNJSq$uZg=cZQN?rp4cIlWbp1ZGV?uP=4)WQ<8<@db_{;2%K z&J#gjEkuqMu-@xFBPEvQIEIU1APvBRuy>;5jgaeVAk~Oss#4ezp?p?bqd8_p`l9}@ zBy)nBF#BcSSf}27@@&PX&@VDT>N!zYnfUlvK}w}FqH76|d;aSN^!*b#3rT|++$hEw zUHf!s51Cc`x75eEzv@_}eo3c_xKA9s81-KUD@)x}{qK)<(+&3U z<2uVEz)U)gph?kMJSL%Uc)?{t`VTBxVt;f|lZqL<$6Q+)ob^MPsiK|ol;R7qHx2>z z9(G|p0L2(ZTwlE^;KEc$_8{YN;JqN9E>J7-lZRXPAQDiCh5!nxtU%in0_@3&5WbiL zs4`!TbR)qq7eczqGtT~*APIyrj*}%_o3wQ6QDx04R)DJN{hRucS@d3kK~CqL9CnDoxT)AJd$xgw}i^=EC_d6am9} zClbbDd9;wFh18>C9z_uG=FJ^0&kT;h4FYk1vk zgb@x`4{G?P=Do}FG|G_+eSz9&tjaI?D5Uw2 zehH?IM7W`=?cr4;MHV;<(f3P*ikA|*Mx*7+^A1)!X~Tz1Hz`d1--Qqtc>Wc~!}yUm zZ$zUzV{o4MS2Y^V=T>Ftw>KC8@s9)zrsU5W?jSNsfums>CWYMk>XAc=k?l5uvok8t zZF<#G&gYXr#qEjytANbOV_53(>%j{^a8V$IfZ6eg=VZzFOW{u;e%g)wXoUAs7}8Q3 zSMjE}sKT5w6wFqVahTw1XRZNArNv;@JJR`4zEs93WM*G5$|1@9j6ew*_lrHmy>z%K z?u7+*OI$AvmNa^YmCHwC%y?dQGMZA6P=*OE7L7}(gm$o6*L5{_rAZ4*hXuWMO8nO> z%}dm~vNYsnl2Af>BE4>enb-9R11<>&mOIIN8AVH~Po^t18TQ4A7=(yA^Dk(bNpogzn=cNE&WN)G^kqa#Ks zn*1AfP&s)tsJgB?R$D-)GT&J&GNlu0UV0T$lgbDzh^pEd?OKRHIldk?NI{5!sito)MH28aAoJWsGD=Vt*!xF<<`mZ+=AXVsme$Dj&4^w5%>Z`;c z^Uv?!QM=O6UfG=6NNPc=jltpth9@{=us-q@EQ~$&U>ciJv>LiVZ3c z4Hb45Toywdd4K?9etoYP2{2=G2#P%bVfNDPPlLl1*A~te+()y|f2%$aFQkheQerCD z1j|R?(q6020tFQ*At{vCGkyf;LdU01_C224ZF2t-*6Gg7g7J^J$z+bdw~QAoR2ml- zIMSql5!Y@)wUPBbBj|OGE?Ql6c()Z$zq#W=ZH%cRK^$fG;s|K58kcbcW-Z79*XGmW z41?8O%gNszbxvI&?Ni-=yC;OEEz1iSojPIi^_lXT%^8UOkQ@>>K2*T!e>-7D(fUY$N zy|3A*`1bk#yiKw(URbrmrBI)W1afUq`;f!}*_Kfo>F=T&9VR4D{IahY>zC7hV%zdB z+0FcWb1i2@Ks?vJ%pCq26}aT}j-?XK2~c!hmE%yU!aqRfQ_eG8r(1Ej{Mm`ug0 zy~bbM@HEe9?@mCH$XxM1Rq77(9j6BQGASa%95eep^c*2#Jg0SjCV75+VI6E$7X4%UR zS>hgTc(9iI@LCT1Uc5rid$FCQt#Iqzm>!9{Zpkw7n7@<_Pc#s&DnTR5%gYl>%E*y%^-+8udKLbh-=o4ON;0G(*U`{tr)3U z(!5;GZWA?M@PD&eJs$!=A-hTE*Iyu*9yXpoIA!5x=l)C}WC7#Kl+E6=?K#rPxaM^~ z5V9GqN37WVQ+af4zN~lKT4kKn3M4TbA1G7FC&$Q7je~ej`IhunIg$S<>km6Sv_r(OEjP~1MR8`G(3rZm6meu((PE@!V=w5823|O<~bby;TL;88F z`qF4SCQ)~u+D4X^f;@RHKhO2pR%JBy;rtROm`Uu9M=?jvoz8F_=)_|9!Nk2FN z76u4ehpe)%@?#@E6xy_$VnhrKSa6#m0hOEyMbcTAoNx#5jKdqY%&o=xuVM!a&o1on9gBX1SgRm20f|M6`aFA3pgk;Lo z=z1doY60x3hJAk@&b)Oh)4TnF8wGwFSb`Fkl#iUL#Ahd&}xAYdhUD+kb5<5Uf%RM#H}Q7GM-8Y=D49ZC5F~8raO3s zQ&j9VwlDj;96oSKQy0_Eo|ER+aLV47_xrK>RrBn904uy+dl1lS+M$Dg@jKi@Ruyf> z8ZDIjinwVjelsqPiGcQ2RZh)rf4gS40w)O>c(cv>;!)8+xq5^-C-J9-oP=GK4qklU2n;}Ju1TNk_>rqM7`7MsVXH`m68(-N~)cEZ9K<; z&!wJfFMVlVOwwL}w=z>`H}lK|zrPwGT)H`^Q(r{05Pw4a^SKIT>+GDpeTE!X?98aT z1N7UCC|c5K>(0FfPfg$aD!`XqQyqCwI))1qj*S`9RomzmPuRqOcj&7)&a_6|Bn#Tx zbVGE4w!;NF#&iABubdjFX>OcQUiGOD-xGc*bPP3686IaGJjrCK4;$-DVm-FqG7&8g zei4|wCJ1lgu7=9HFF`6>r4{-4Wg}U3T8<_yDLnYddVW>b~C~^}|Zr8nWO4mb+z=p@(n( zYx}tgF$DGgWU4d&PifUD9HG_l=I%WIp$(ZKP`r0)ifGvLYF3!T!{^$_N@b|x9jz|$ zt9UN6XPCuvWS`Y7$q09kUu*z>@|sYd|3>&yFe9Py=s7KK3pQC5yuAuW_|usVEL00~ z6{Hv1q$EceA%z?hU1*7E_ikucIGYuGiZYfB}+WH3tS4=`YlX05dn$sht8+6FdkH7i~ZO zfn%BnHB@?lrh3M2{$ZWXMD}`sEZ(GOcdPTPzUp8Wx*#cWXgPa zjCf6xhZM5~V~Q$6%diytONs5^5Xb6=JMPn=h(VrSM-)~=!`U{0B|4N>H zV)=n^Ncb0d8&}1ue!&vlK=v&Tt+qwTIw8ra%2@ak6CR<$li-RNi&jHDy*E*DPh!5s z1h|TZKlWgYrQsw&zfJUFUri$@gK@?;EuuTM!9lILPYMYLt6y2oYK~CFEGKZNM&P#3 zp?vVT2=T+F7~WR4g9Py=-boRw^tg{XylC(lCcHjJSHkaK0;ugHB{4sVEZ740H^+f( zhOTIP1tCj=zvSIe;O?Ll^hvNx0nX8Dl@_(!^Ef~^VR;0Q3zFq+zr^=Px>TKMfMuGp zhX8+>;RE-TI;W#xFy*65^_#khFV~SnbEgmqYcLs+PUzT)Dy`YL} z%It)MMH*F|+ZaRl`&5axdrbpag=z`)c4#I844IyDx$sxrg3HsXC#>JzQ^EHLgKg@h zvdJXj)_}`a;qns8y+EpHl-4pEcYS)lMyx1KWdji1!i%ARWekuLi->qcML^N2UJa>E z@8=461c3DIp#X zttwsd6XO^9CDBEd)7fC+r;ul*whxdb2nJPmuuGs46VYpk3s z7ps6BhUIQ9rn==6ef7!l+jo4!fZhF(5Szs!k@`-6pQmjiO18je{JAug)DFpsk`xj) zqbHwwQ^*M678)khk=d38usdok*-m#RnP-~#hS=*zY!F4}BmkTb0RYasUOJfM5e*wuGy}wo8OYuTPyk$|SAYQUGLe`~g!TG(K`h ze_TqJX*e6PK?=Czu$Jxjov7>TvsnW+6tt=63=0d zXh^q=d=VsIMHI)ws@=fY=7{AO} zLq8q^XLaf7W^z>aak{h)vor1dHE5lDTu`0l`nh73y_?7XQ&o%%4{6}B4GD4LiG*5V zo>q1v!)Djf$GV$I%J4ZiS9kO4PVv|kOBW?RVMRolD1Z?XFa2`W@gaXWq(@-wHS;aS zmho3AuoRT22II|f7}d%s2YBk6qF8M|O|VtL zn?aH!%1NYQqNgy3_{;mBJYwOVCC>ID>Ou{yC|HyvA zF_i~+C&T-eSMr+5cC|2L;NZ(^HuW>#Qi2C2U_*X z5VsMp1cB8bM%mwE$v1`6Au2c+WdZDVsPa;sJGWeOs`rdx$!!4)??xd|A9W)kMSKe% zMM-s&R|WcwDQ|qh_~#SMaFpqR)Tv{|4Mz2G!UKJ#U+t&a+YY50&H0+!C~`~aWW)uQ zfVSy1kvF^D4)8dCc-9r5DB14%cPH0PN58>j{AG&gUF1PRz%w3XG=j6l<#S+lQ`ZAx z@zj1oQiF9L+1O01zJSAothr}H5`J8W43jZGg6EI?`kc*N3!S>A(a}e8xHa%bQoge+ zt8%SwO(SjFwN4kvrH^U!jOrwk?2cvf`#%42XJ)2bK_bDiAEDz?D81151W9(BUV*Wp z;cN8;%i@Clq7g=%LYcn&&sMa;3dbtmr~y(GXo@DOxCPL+vay?Kwx)XGMiyLQrkD@C zlVgK}B#fyCB&+$G1?YYIAEmVp7L62T+0GpI30p`7TMDWFgo;jfkh%Gsp&K+_ClEsU4%ZLk}7=HO*kDmoSP_nZ(ZGmgci zsni-ChboWlZGo-ILB^Ju2!YP69t-RPGs;4*UJDH~+;*75`IrJoTd<8CxReX9%E{2jxD5?tPW1}<5G~5pqFE7zw@~e{`er}>bv@7m=Q*mbu3VEpDr=U3g+D+4(*3wHc^$SKx2&4=*#uwm=Kteg6qvp|EBM z%lq*4Ev{K}K;C(y2R%7P5$7}PsTaRD)W9E_1vuwP1>4X_a>`~aHpnckkvDX+nhSS~ zlsl>#cxMLgXokekyqLE#x(vWPecsFZws+2LDb<44TK{ed0`UlcS)tNRp4WOp=cCUe z=V0cIo*i5u4Jj|qizBa=TPb%Xv#dVB-;v=Dt+Hu@0kXWy&s2@rSaLmHxxXYA$en2r zk+A0~bTL4TJGAMt#*UfnR^3L(O!0bX+#_T2Mg@przxCr_&5%uF={IfyNDNESc2(pD@n){n_WDcKz<=$|;D>chh- zzatH7)UGbtjw@hX9&#^h_L4+u65+yTa)#AcEZHHq`unI{brQxlPL>#9-emGMd>dP~ z&I(xAdpWGiRD44c#)?#n4MQ;w-WXdgI8!q1vN5pZgSmdvE^f3aiN~tc-_!ntBd~Zk7RYS6)YZ$^oa~G&f)HOa`(B;fCbOt>-F-bj1hO{rf}OnG5Ju# z$@N73FXMR{e_in63?lZck;a$;rw-rvXlX}#0X?FkccFQ;v?AYJVo^l5#?N0Yb4oSudSkSpN9>f z3}fv);h3HMgDVLtY?ZaPvG*w3hN$Jds>`hBY_V1Y(@~j+2q+x#I(?Op2 z18pvdkL1f^16xkYjNuf^g&-KoF-tDvFR{BLofAh6F5?eipUuM~1$IrtDMOt2 z?@EsbsIL^glLG++A^b$reu{M6GUM%rxR`?q#EOIO0$)fczp#!wcwea#Np6YsqF!gh zOAJJkU~DAosF#I2!Ej^iEaF;QB&{E528pC_UlG}eBFXaDSLNji)02+^q*lRXz$ZZC zqq#z{VnKVabH2K;)Rof$p2b_22s@_r8Qv` z_9Qu;d9dsEas_!z&@PqR;BVjG9b@_ARWorXa>z)(Ml)z?M$(8>D*#@X`WHUJ8o$qW zQ5x3xT^%oXYam>IQu)k;MkDeqJm7+W)Tk-)JuUu*>g_Y9?@BcXv5n*YT9TG||4Ju; zIB+#;gv>6uR=mQ=VJ&8ICWg3j-F<48qDMKjjwz=|+MKaJ7TYo_Ot1fV>f%S_e_ZA^ zb$9Ksn}$(TRDPB2Jl$RZb>U<#^{=A9#HLl;V`cGcjIhcJxeVtu!%_ZYY<_?pR)F>S zcf=rhSj^GaBQwrK=NG{nzUz_SSOj$a!?VQkK4Mv~K);`WqMnHf^ny)(+AvrxMR`o|e?PB6Rk=?z~@I-rj+Q@uMGW2^UKuBQMGl|S@k6B+JSQK`(DKMd%#MGqS~f)>zCU^*C9 zY(xNU%Rs+fnvx>%+ea=xEa+S4QR;D(vuJx%+kw!`( zM#@<4Oj*Go4HjU-f5O&%ot%DrAPfBg(j-qO<|kjtDG`cqpi7p2H~e**j9%y#HznBO zOi~Lw)Cz;G8>Y9xD?7P$)cY5g9$V(9x$k{<-LIhnWROj77|e9trq)=EJbjA)O>(2F zt00&Y;$C$kS6flcZ6npn;!w;EdMe4Os@?edlHkB{wFkjDwa!HQ%?{G@AVW!GaJCxI z?Du>jU5ZbQD`W!MBuk^3pN`-ojOHV*lCc_pRZ_2ML!4xM z##1Bq=B)Y7LL_sblt9+A|%L**_|EQZQ-aas=O zu#&i%aG@niPwx=)NFlnt!{LWdC=d9p}$+niZ2f4%Qh7(-i#nA4VWG94Gu7wVY0*m8&xu-d|8;Leim_J5Bge17I}pVS02 zXA25%4vxGDUog~>%uwy?X4p6Av!%SSuaCtZK#Dz65*GP|#3U9wd!Bmtp3da9a@*ip zm)>rXQ16@Y9!k=;E@}(mm6f)U;|JTKms(+ig_^*E9r>RekxQ1nodWE2eY560pt~3W z__V^EW+C_aA^v9%c#`u`j^0|jSNX{rWImo0-tSOH%Ok;f@9Dgux7XxW>>nWmu%C8H z`}I9Dp&icF2Fm$Ry09xaonrsH*SC1@s_CN_Y_P9EY^U^?@`4QFV?1>hXcg|=XFe~M z(0N)HXYgaPWSD77lF=!^AVaE#bt^J|k>vSu!e`{(ht&~Z;$S%`DQZV9BLttpj`AVV zwN3yquGh<55x;Vxd6;OZH9cHJjHK(IHl_89jxDAw?r>=uW`n>Q)y1a7V{0vanhR4= z*j@7}va`;qeH`z%I?zXFU~mgZYYxs#xyV`@h9=f90io&8>8 zQtTl%ci5LWwHNpZ$)VPz~&#Ve}638=-o9{bvo%%XJ?{YTx%&+;tHH=b3(NK2oMlsw7g7Se_pMq5xZ z)vSO#$XSHAey$hOJ68l3{S$1kORcCTxBRwQ@|wmlmi#M%%&+6bDCDmt!vX44xgIf? zhNYadD_QKVDjUmZs-{q=viOAz@pMAwr!jp|`@@De?)PxkCr2jt5)R`<_*7&aUE;3r6d?6Cs^2YowQp95EgE_UF zio=BHhb4D}(vDz4QL3`u+d7apahhm6aJ~#IZ*GHyTxEYGrqUN69M7D5wONaNdXHni<_iKlKtxq>zMxUk|y{|#MZm;?* zH0Yy@w;g?_NN7Pf{zL_7KX%TCwvhIt3I1MZ^RexO=nM5J_K%g))kB`b8)s*(uI*Q4 zCmr)qluC#%R_pFZPKzFl#ZV|B4-rL9S6#9G;?0dOu%XP!yCjoWy%Dx0&?#6n9{|W9 zB;_B{*p8pS52GL@Bru08lAe4<+;DY!ehGU^8F*3)h#f?s{_D=!o@Y7hbNv$0f>E{C z&beBA1ie~rI{~-oPwx8^wBv;Dc8TPsfo3t#|Eswc`lbmvQiLuUIUp%_@ zn)S8)^Xk>D36+hGY`clw6lI$RQR0zX5M^?PnuPm1FluzgXxXSHF-($;OgrjVZtDPV z$kno7l&XOkVdfMybjkeXowz`yvTWPJkFypn9_Tdl9iSm&T{!bRab zZ#nrDYi-<_;*DN`?760MMxwE*lbWqL1j9>cvv8|$V{W@xxzPR-fSe=ZqKkd+V|<`N zZeMb{+7#ZP*rx{`3!Se5t#iHF-u5czHmoY&{hIolVYdcwF76nUNeB*Z@v9TVX;3OZ z+IpzsaUDf^yULr{L%t;gZJ*`}5OQMlMXx8hPI$SpiLqtf1hB`UMHKV7EyEfoo`8`Y71qg zx9N0khd$o=8v_|f=`Y7re1Ml}8C^ab5;65$mrsiP`el5r_e}H?J8g3Bs;E*<50}!6 zw3Ew@MtN*ra^empL={N+`L)$w+NV63D3P3gtC*rusmuLcqMdIKc>1K^Ntso6%l?XI zOLq{E-T)M=k>|wI$o_-wyX9W<^`(ec*iLdLZ+7=PC`_Fezls6kaqEOPe7?Dr_2u~_ zC}8A8cZ&29@MXp59>oj9rI%@;jT;M$JZp{{*6Mt;(Zg45-*_G9f4ZP+?b!kq!zTFN z#576hoQ;y13p~0y(>n;C`>?IGTZj=jM!bW*Altm4Z?i-TklCR{)yEc|l=rjMflx*~ z!l;>`Qfd9cI<@w}BACOls*ZT04bYCfP^S@aA}%&F&NYfX)RR}~;;)I5uLdGFDNqGbhi z!RaS~s)>=wuSsMsl6Pe-O|vf^(}}x5U1frTGEeuj<@E-ToaNeox6=yO10<0ZdbV5V z4PwWl1M?^+=l5nh4?snUTzX?2^kf~IUGO1S{^C-+F2{4DJd>XYIAc|5zfwHsw4q%x zZT$zw=e&KlU9bYCFMVjFZ}6*pIVelFU)?cy3J`Zj0OF1d%2BBio)IJRqnd~#cYzb6 zSZTk_-y8Lamf<1(UrE=c{F_e@6PnB4^~ z-}i*o_{9gwHC^}YOXUxZrLqCH%2T zBLAvp z|AOMrsrr9dR&HqjJ4-jDs_%AH#-pEO<804RSxbBXAe zV@&t~HcmmiSx>L{>9VRj$bnrgjql`GD&bgs{VLV*1J2|ez+-y2+1(c^9o1Qib;!Sb zc=A7}mI{!?Xf%?CNkQZB_jjz@7>+@Pi6$4d^UcCUXHE zn^tp)D=TgV#?#o-?;#Qsr!#bCCi_EG4X%TDAvz-{$dg0An?y3J!Jte}Edb0iNNE5XQmYEFPSXPk zbC;}5t8qWvR{RrJ6PSbG3$}aB7@Pc>G8PDM+<2yRYZ-u8oLPPp{a~8wz;>W6v8$f) z66>%k^M!O_)nldLWsns}o5fj=se}ZuAVW_Bt1dW}padWPCXK1>NlwI3h7t_q}9|XyNQVcDn(G)GPr0(9(!ec2@7x(yRW5~OA$S`^urs`$t zKJY+ZGcOl}Q(oYRw5hLAE1tnZ0Zve0vF*8B{u=0$UVW|uI+<3sKt@&#y!rmaV78#+ zYlofkZXm7TLKCd2ZhM0DEoJ3uqGxaN#GcV?Xs39vaoLcZ*M{Q<~});~0<-+JnhpBp+r z3Tv35w;vjy0r*R(I9-s;*e<9NgE@)!)_Wxiy z7LOAbJJZu?y)yK)9xP{cSHC7@YT!0%G3ek*7pPde|0Q*={4PAULRW^N)ZVQu>AJ@t zwiC$`0AIzZTlXXl+vPgn=3Nt!`)^>7H zplaFPy>{GHgBHL_{iosl`mGiV-JHxw3v7qa)xS5sGIf4BReQml0ZFx7D;W+c%zD61 zh%O{oJ<(;zkI_6{D^q=06f)awlLUfLKblpq!}?LC5k1=EJQ46YiIWzI0Wt!v#*=OS zc!uf7^ASZ6Y+(Ub(p9}fBsB`W+c>}9br)C)wgvqLVd^bF|6 zuazKh12%FW5< zQ4cI%;7Own;HSI$r|lh2g3)>N37PMwsIGkTnFy|i>Y5o3kY7{qYUsEIyfU7Y!loV2 zg%L@~m8}2}U8Fl`&LK3;ot{}e$4gHDvva|4X3d&;q!Lu2?Ew(0`?Nz`fL&b+A|+i{ zRjA6OqZaL5^Yk=#_x2o`JO1VYhO~lw!3KQJ`2`-dYa3{-aB^lk@*NErY4v6>4&8hhG$^}})2(s{A23Z=0&X;n0J3HPq5{z{J9|Tq7R49FB^{1J~ zv`%tgExaEA#^9#Eob3M9Bgu2}dD_YR#_Q)QE7e0815$VqDCC!{)dxl$NBdZ3v%uWFO!-^Wqe#O5JX)S?n{p05Wu=-tOzTT3o~ z+}lV(jQj}11BQHf8~%pVfsZ(VF9SWhcM=BVfh&MPE4pRhew#?Q4L7W78W3=;&#B&- zDF6pi(BfudW=WmD%~ZH1T%HBDjvfKPj^diM6Lvo^W8gF`y!?Q9%@Df3i<)i;ArHXI zvdaoc_#=MWxKDZ-^nzCYJX$jOp`AHVMU7`iE8~ZsmADFUDL@AzfR9hW#^QG(GPU7W zmDf9LU?wMv9_SjogWuy1GTdTS^-}M`;o20Ms0$@`K+Qlc!ER}Q=X|yCEUzpeu>O)o z)(oKOT=Z0aReqJcsv7u2WQ!qtPIaH$7ydjqqS0vQ>LrQ9J#9c=Iy$_hz36{VZvw zRSUhoBFkEulW-`LL!=XPDvblXB2;QJYQaNn#{DXqh`y-&*!$xV( zt-OX8Q|D*<+GR~1*?t_4t7<0;Kmp%8xyyol=8{@sVCtEd*;4oQoS=b6$C<$Ri`$@K1t>8#y|JK@nHv5MI4K#emb4Ng@C(z6F)b=B6^CKGm@h13(% zm%5tzFoCA1-+bJ^R?Y1?29%8*Em+2N{>4XEe+$BFth~_>x!w+mv#s~W%@`riU0;)~ zQZti9&>#ti4@;#@8S+0fTHT7_vS0>AWKqd!n92Weoh67$6b_Gn+>`QfLMUuW-KNZ+ zpzIRCj=iepKyI*x{Nb0+oz@cv1E$Zu>16#G;%&@{G12C=;gYs`gIOO4zOHkxx-0JA zPq!PiPi5G*V<~PspX-t=2-<_CDjbV@l@indaIMsemb_;yIx!%;4#p4YzaK&@P!X(wS?U+ zWw(T?rMS{0dWikMEw$+11TtgE)ViHki@Jp=Zk=cZRYbg;>$&@`@|EiZw<64ein z5K!d$p#wt|>ISzqUtW5~9osD|a`QNE-z0ZWo`!#>(HB{yjt``hxZul=?=Tvp;B??1 zdGvN8oW1Z}DN({VfHA%a@@F|2atP@rmMM4V*Mh zJ(yGUI}v~kyfQC#kDX+nypwSW%CcfM=uUE2`Om>QP-9TJP-*j(&?Kcyb6;D|j*TAV zL#dt2Nt~JO_qsQ@D1EuZ+k+381>#CibLrL{)1WlYGRjUa@yZy(0sAirCvled?)+6m z`<0v6ief0drOzT#71#!ZLMBYrQY~;khm*KM$SK<&ZkQ(*-~w#z`qggVc?xP5NeG!K z+X-gKnsgjPV|6O<6h01^li&>Oq>M-T_4<8z%u5;-3!g#V@5l&*HtK1=vM{Ta_I4L}LoSheXD)btuHk2pAJGSn%p6Rh z-#06?NUVaOr`(sQwrrQRYW2TOBCFK9U9}4w2{RL!_RaQ*+*@6cGymK zReZGr3+XFj^eZE7)x0YtEHy}LZD#x&UN5571CU0$+vmAaM56^su{a$&XYP%Z z)Jdy9%a?xzB;c+#*l)}0AbnP=wLKHirVPSK->OF%EO|0-lL(g)v-hDlle*TgZt@Z+Tdj;I$Q%}~gRx*;R&=U@omkRwd74PKL_{IRz|!zhz>LDz zNCLaUyofZ;-ud+^QTqHx0V3kP3uZFLhtayGHUZxz@CmosPaGbI-f_$JK5<(zZ1}Ib z2@&^M6zp+r`tuT@D{OojYm*^k9^o2j~h{xfa8_Us0?X&!2VT%u-^r-<}(vp-+A-ggM z>Dy$53G@P%0)}^F6K+&41+D-`-qBCdV&BpO*%-*!W0|sT^I0Nei8WY)Q_4b(#Y}ic zPxG&Xd!>mRSXzR)XW9T^U0(9-#HOF{3wgs)goo3OBK0bHpSC*-x~v=5QchKZfJ%$A z1$kI7#j~EptIOGe_7#;M2qgy6=4d_*Hdd-nz4$K~6CDYR8O&~3``WOZy-Zq5?G=1k zL=W0qVYe(xLtf1K##IieV2!ClS&_`@LU-k5OEkDmRui{wmEuy8ku;X)DMOGKDg(AO zK(G*qG7wLIqG8XYRk3^VVP1}-N#3QxQJNYPN&lJi8+#COTRzpSTi2y5PwE=|ExyoX!ekWJ% zkjUBn_X3jN^UrNzcthf8t;5phPArUnN7}fn=;p)(zDXq(wk#UN$}koJ7m0o9h?;9d z=zIEBj-mE4hMx*~&}*US=&mCsA)1VAeAzmeg#^S2eAPSOI>wrCJmeL&K`ncNA!)=D z^wOS@I|N}(Q4BKH@5y26^3T-xJcCoaN9V3cKs zC6tH}gfpc^`C{lL3fOn(+xU&J<`DF7DTUssr!r!{i`(Fi;$ieN#(l~%*ANdOn9ZT3 zXg5b3B;dM}XFO;_MJM6S$@cc0HX>$#v->+LwRNFb#WYxEiJ@|2<($;8nd!i5#MEwE zrnlJ}R^ky7@Z_S3+C5p6HAX{+P;7^LR{4k4j|#ME5gA_+y~KM-6qdK>n}jqo21Tyi z+SZ(=HAOv_4fkub!S=f%-JnnD!J~|fsPr~J7HK1_36FOHRmb$5&(QGZo&Vv%cO;t* zdS08Y(RKtoK8&vKs~BM_jQk{YTV9-U6AOo|w+}3k+*KG1Az$kliB!G${CP&U16|@i z3%#`rWo=M!!#VMNfxfn@ed6GC5(gLPFCD&Gbd&hg%Amzsd^cNsiu)Kp;KPh%?QAAG z3IF!hM{W2pT6q%(=8jO=#qDd2?m|7D5k@6aw?4|GEAaX zoo1XVD>9o%wcQX#?t;2mvoj)k$DGprq6jw`4S1+piRc|X5`RvqCVb=C3fULUTKEI{ zZztsjC&+od%gYazGCh5O*6?5<(^O?}LfOdH$$ol)GBD?A==(+S z{c^UIONd$9s(6DRw*Sb0bPOSg{BdcsNpY>%sCp;0gJ-3e%NnlgCH$~Czx;}*WZ$cf zDYQ`TPULDZs91a?wcwj;P?ETLlO(8J5|&{_M-sRPYjD22&Ay>G@6cNgB<@XKpv{R3 zS#kj3jvEUdv*>&Mw2&89d7_S#=X=Zh6mKH;GB&Q2g=_0O$r)IBblXsDIAHQ(@&}gD z=bA>hN+T*{b@4~=d+`TM`z@P-Nh1n2JJ>b85!(qD>s6b^;QT5g={Q0n)g$#{WMDo{ zAIX{3iSDgGF;t#qNSlbz;dx_?lTV+7P<0wLww2Mld%K4mL`b_|oaOFlw>TaKAt>EQtxWO?lBGrOZ*XjJ38fC2MZ$ zm~RPM^@jQ|h4L#|$8o}duHeCR;HuvzdSV80WS|^#d_4gy`1_RrRH;MNRh_Ypp(~bA zb{UIw#}@6}Uw&nG#S_qP$FbrtO7i3CjmE2Vt=F38g+kxN-;St1kAy`sNGFAL8P+Cf z;1z5G5>dAAr8silVp-@*jVF(OjVZd>swOH9-M$&hI<~+vGeH4gXlFe8S>11{0~f$h z!SXh+0L9$UtB>J+I%Je?gA^Xox=XN0}21a#)$Djrjph2Sg2_&y+9S(#@=f&~B?3 zslC=wLRYgWLCu3YhiT~uloNXbMc~OApz!a|7Rg1Pic^s5M%I|oph-hKJL2}xd3gN80DwLNGxlT|GEQt7*4Lb8cAd#9Yx$~fBy zn?0Cunjkr)(a--H=I2weJ4Jey{KF_}ecWtc)M1~Kcz)!KuWi*gCW9SYqWc*$6B3s* z+4xe3;YSaAGWK3L_94HP%`trm_dHQp7S!Xs(UC!$-8it{{Qi1ErbDE+GLPK(yFM4r zpLg*+kzg4u8JVyLL@b>f5d0y6CW2lW1q2oZ6KGTEcc2UOMxG~m6Fv{&_H~RR!XtrV z9)m@2L%kbBZZ}|1p5CpNC2fRrkZ_QZ7+H3y|jB8r0&g1n)01;tX+Fr{nKm;Kwe zV(4e#wPCDJ$hMd|>qJ}DphU{~O|7a5O6l*T6m0nW0`4<~mIvTwv|7HI`088YuH0$e zRh}`hd+yLT_c*Q|Id(kjHBoV3lz6oNqo~aTnPx+|rw<7ZeHHh_`aDDpd2Rpcc*p6M zFY;Q)^SyQ7C?;8|*}BP&+VUQSOs%Xgr{t=0tOU~alFDkkS9%y-=9ixDtrpf6dfQ!7 zm6A*fG?(U(OUkDeNHDQcWLz5CI8ddgA;I6vn515>v|{QVK^oE#{hjj*fT1R-gWK#P z{WoSHrvCvzn@n@Bx6+WrK<`HB`QL=FejsgTgA~wrq82aH)()Lf0OPJ@tyBz{mJXNZ?4EKvGlpS#OuFsA5I?)WT%S8A&3O-tcL;cpzABQdvb98Z@<`PQ zis1nIaPPFM5BstRY{tukD&<$I_Jtf+acHoci#R`WOF%LWLcPl`uZxd`Dh`h!;Q==l zZXlV}@=~ao)z?moPWgzN%`cByd};rE&!vM8o;@SIIyEj%XWZ@(XdwepdxUhbHHlljmW-e+{bh&ZQcRvTY_dIIu9PqWgXVE(u-pG$7M17;tN~%azfE7`v`mpV= zzoC9e!tEL)(Yu!6!%3RSD)ID_iQMWLOx##f=4amVvq8Fs$suth0b(m!YfOVq;^poI zjn>$v5kkMytKuC{Uu?%C=H}0ZUnW9e-R^+YosrKctQPc-QmA-)>1Q1h!A&wK7inON zx0P+BksKRxsX-2L12H-^v*q7$2%VLrW8DL6q~?WL(P~**#S2IA7??$)keO?>L(gg- z`06`s;_$RpymOhg+l)ru2J-xP3JkCmw!1_6`pRtR-j#KhW9~E=4&i9;iKX)f0ri2f zkvxi1&Xt6EPN>K6NT&G5RXyI+Usa_x@KD)?dZny+;qs6i7|;^#DIDH3?k$G+kUO$5 z2!_@Uamb+?9YnGgyVzw*_Qb@EbZPzAAY%11g*Cn!L6~KQqAdJp7+4HLo3+}k-!G(6 zlp<${`k+4|E|92D(B+7nNQ5#4KB14Ey-GPh#>|*=JvKlC*RF5RJC+6M^tTA1Kp<#3 zQe0{Ge%@@2kU`6R9Ah-3RCaOl_ijhU8cKPMVwWsl%~_;LSQJv=Psakj6?TY~L@&)w zh)0E-*%YYj5)N)z%nqTTWvBsAAl0=qIxMR(|Wwzi#+BsKH!UFUrF6*CRxg5cDT% z68HuYNS0HIeN-vMW#^mF*Kt>6SlZruPngjS&k3DH9M7H$?fW8P+W3#FE$Q?SFJa@a zHCeoY)W|~G?~qDe=%>Pb9(3L10FjHE%!v z(0?v7%+eNn{#P|OA^@1Jh-V}TQNLDG>>9dx+*1}!n!ZwOeWVdm{LxzCbV7!A{lU5K zZ0g=vf5|bNpct_8);15slZqv$j}4Q-8_OMJ`yKV`=cjYkRg<+Had(*E9P!aMrU%I(Do3kYhH_3olQO zHwrA*ojOdE>>n4MvP_BJY^~}LTlb(*9jiLujZW!cITNaM_%hmf&ARJnaKFfX^1%#c zC|XxM_U~d7FQeZ(9Ot)UZUw&L8*u&o?hGr&vy6JH;KJIW^CB_%+pS`|?fz-j=T*`9 zONN`N*HO+!$mltt=sCB|l=Z{gQOD*g=M$nw!(%(Tj-ALeBT={e5J06#ve?%~7-fe! zbQv830wtpFCvr6AcN=r3CH*92n*0Yxbd~=5^`-(p z%vCp$W*j2=f4m#`V+82fT4q+9|L^rG48r{JWeDqkz8eDNgV5+aO{rl1@7E`Ag^KOp zk|h84K>z&!zn0?P75KgL{@p_V8u9PF@XHxF*#(|_IQFL(KGAOoWR#isZ-6900S c|2vi77Obi2RkZ3rp#fitaw@Wg(#8S*2fEWz5C8xG diff --git a/examples/codex/images/codex-pr.png b/examples/codex/images/codex-pr.png deleted file mode 100644 index f9ba75ca129f37c41a949a93d3a361cedb05ad42..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 534175 zcmeFZby!@_mNp83AR$2m5!{0(K!Ur6H16&Wjk^X95Fl7^f;+*TPO!$^-Jx-Z?xx}E zoHKLI+&gn-p6C1BKkgs1pS^omSM93mTD$69YrU%=Qb|GP*^}2#P*6~w$;nEpprGJ% zqM%^tU|}G0O6diRQBa;LSW8GK$w^31DLDfztnJKEP-G*MG%>YQ`$@8NRjARiBt<{& z0r2AqML%M(U4ErvrpCtq7>zB~^Od!7Z9xis@Jpr2^Tj89CKCE?=E{eD)j7H4Q3ocP zD4?Ycp`Z{xb81my?+d<_K{?e3QcK1| zQGViep$%1`H^8j^`jj8bDTdIp)Uk}PsO+>yEd5rCy*n0#s_X(<^9h9;Mm5TeZzfL8 zA5Fq57h;0l?T-)N+8)LiZ^JSnwL;s?XSte(7(&eM0bG8P38SNAze}|+aQH-7Q+CdG z@3JActMnY2?*zl&m+x!W#A%riy#>ZPTZAxG^LGR3A@R!iwrGJHChIxvgj}`Hsj-=T zrlhHMJ2Oxjx)ZjVo!Y2{HTv+dg6Hrzp2|6x-Npfw!+2$_(JsO@A}+kn@DIfFXuo0h zKAR^?8IB;&y3q-X9(=WteQq8s{N_r8#hH+o+3%Bpd^hO?m<*?OBY~<=oJ@&IAv#10 zZ9~>eL2oV&U}H?e<2)x_7@^*4f?EBF4*OaYV!HjRCTETs-XaKfJe)gYeaGR9WjHYU z1y6bVi4WZx9t_X)H%yXHq4VcgSz$h3UXmYgsFa8W6iDO zVQhR|DKzjb+Q)w6ONEvyicS{xeVsrqScuRw1}${WBSKPCm~$lHrzZLnst4=mWn0~} z8To`#a=w{$2TPt}ac%^7c89B@nD&LL&p(_%jjwUKfuxzvKLcwCuUz(T?Ffj0W_B=Z^HJKURiEn{6h5zx^Q$f; zsrw^=@Ol$KU?_Q4TDALO1YvA>lT0v+gJ;8>Sb%fR$lm*l?S5u1CXZ5&eJ5D){@{z@ z5lsi_Vf05499>23_rB$q4Fkb8=xA{0zyfNQpZ*nj6(itWOd+c4sAT{|hp{##aFGst;OY5S=0z@KV z`&lDVBWuRFg1^E=N3mXRL1Q6kLHx+%=xBklo|5xz7+>%p{@diY?r#;iTHGS|XZQs( z?J_wth4{(&{aO}U*jpl7YTe3QWB1(Oy(=V601ug0no$@Tn08Q_a))tAb6d=R4||u= z+cr*Gwp+wF!7{yA=AnMlE-WviBq9?~@1Jn5kGAw{gsRsl`8P@MBu2ZbTd#KvaWosb zkI^BU#MIcx+Zo5ob@wj!E`M6nJR{nYap}AS>Gs}FrXrhmn{K@_F+QJOG6*x6B)Bnv zKV?&Q#x^6jfcEE z*4moex;1I*xlKGP182W_nD3uvnzas}Ji=Ze5@&o4uu45Gs?DtZ)jR+lkX-1_iNs<2 zpy&%RIiYc$-*E0qT7SLXGF3l(m5|<=K2-6$0=(QQH0+lEcWb~ENNdmVn+aZ9KdBze zo8HVUO1JFTMI9AsBWrVR+i%NKsL7&Bt23RD0CE94Za$o_Qzk2>jot|BcNBK`uG*|( z`tSIQgEIXUc^fk9MjS?P0vH2QAE+OA^c0VMkIzt8sd#g^b7++W-ctYMJT-{ZRx)73CNkvz+tkzS5z_?q@&u1)Tm|^IUdm^ioF!kQ+ZVQ?)%1BBVm5{C~%M{5Zpb9HOSI`zI5SV^nPj|JT zc2H7Vl$CDrmd#@!9&!n}gJ4c#S<6|wSyMul?`-@7_&NMky|V*Z=BPZ`FNrFNt|Qu{ zY$X*r+qh=T1cxt2^wQ@Ow$s(Krdf8ru=H@$O4Kx3fchf`=99SM(i2;GVlz^bJ+yM! z`r|Xwj0H`vqZLZp>m(Q6PrB^DQvGV)&MWZIWHSRtW`jgvM^8w5c)q|N3qMwW478*l z;pf%0Z7_3(leO{$x}{-5pwOK+%rt$g%$c?c4}mj{5v! znzfL9MElM*UUfVz7nDrLX3g}0*+A{?G<^-YFVFqajD9x7n{64!c3*M3_|=>!#a{2a zSxpNjpQ@_uH4onYybbc@S#SMvf}vqnzNZCga2{mdVD2xqE%VcyEg#oWZQ5C1h^UV_ z&^q`yi>UX9!Cbc4w{>}xc;0+)RS(hR2G-3r*=&Lb$_L8S@6~;HUODJAK!z>WcjR^N z8o2G=FX%K}A#OHKfOa@5Gg|iQ`W=}~?MCoT@<@S^yI=P|+>3sX)MRG6%elKvch6*$ zx>Z$FaFy$7$Ih_q`?wb!cQi2?DY^tpbTzaOwVT6#3HNkZ!Y8i{1@PT-j;P>9U9rg` z4*cD`Y6}tEBi4tu12+3M3J%ADWZq2`J!RWtieE*JY~I^f@aOuNZ7c7OZ{(qYpTu5K zrcj&+#d~cSFJ3k*GBRejIWzC5fgwL9Edy|2(5+V^siv0gqYYUXf*!Xot~mg^%OgAu ztqqj0>8n`@atHF57fT%6hL=rq6MQ8Xd-SXFjO>_t_s}ON1JJT~O@S(Uy;dioEAQE7 z>oG+xA6ELRxY1I~?SWG2906dSU2gv0$hxZ-%; zQf&|327?~GRHCrwM7ec&fl_CN^73>^HZzx>5|>2Mt&m)DO~iQmmGjRg)G?F?h^hDb z>aZKi^%2UiC;OACs1b#z_$R1ikwh~4r`isuOC_w5o*!=gVMUj1uBX5+`=4M*t9iOP zYHp}`ivg&OmBI90-`G`=4#pRAEjbGS0EHQu#zH|yeU0)MnLK|%Xl z{v#BWaBCFwe|ts&dH-|7A+JAm{(eXM7KVa}{6>hpJaZoX`_nj`IcWc$#?V2Qp@^wU z$jKq^s;187<_<1aKvx=k?LuS*wxg`J3knJe{htd}PKE9i+5VihnwF~;;1jp|{`QEgtq_eCK#58M z=xk2)f&DG}TN>dfR8&-g&Sn<;Dw5LwRvr0Gh{np*)sdfr!^6Xa-GhrA=xoXHj*pLz z<1Hr#Cnp>72{som2UlZHHU}5lzc=#l?MRxtm^xcKx>^GrsQ$ETYyxz16{4Z})6qYU zzyD5iPwRj7?{lBZKyO=vm0PT^Tx(fgEh5cLQ|5*5M6$Lr| zjQu~z;_rg~TP{-4!cPP_{_)m?p9HoC@gZL%nYE;%8uE^uW`7PeHsr^fzu%E*bn0IV zZ@TVKP()GWB*oM`Q4cdQ+otOl!Ib#gPxcKT-$-V>$*p;gDi!ug$_)Eat{r|{!Qr!- zn-x+uGpqrO)F(6!&%tF_%z{;&;;G z1)^a6#gmR0IkbfRU2&7YE+wiE8446s{J(h0GC#uaWSkxi?f(3iM<_WkCmust^R?3zzt_ z(~3lR^={&4)F|Y5mnj zg7OEDF#i{j{(A}hFChKT^Xz{C>A#;te=Gf8K>DBY`2X*KB+5_cxHI$$mj3X7z(ZV& zid=F$Xg+uj{(9iF$dBF+DEzel;I%vMwJ-AFe_l2J*~XD2`D3;Wa)i46(`*ra`H0n1 z#0v)G_uGGtfCEJ@DHtN$_$0!fy87=7C2rq+(xC$C|3dx0(NFne$o`6cQ$Ch^^pAHF zB#ZTP57FJP>iQ9TZL`{3E$=d=?ew8+Uelv7 zI&ZaJWq{2$BZllP>(Yq;ug`)}urkWfKean#Q6h-L(69fl%g^YHR(Q?@a!sI|R=Z## z&W(SH@*gt&_XO4-YXToPBQQ|p5s*C~JeggDCGdvm3HiPMo8O8$jK6}lvrm@8f0xzW zOHljmq6e8WDoj9Bi&VC>M@KVKYb=Mq(cFBD;`!|(!gkLrt4f0{nXtQ=l;vKLK2h-i zV2;tuh`{_Kz7h3D3MP__Buj?BNf~JQ#)C-zHa!9j9~rC?2%goq+Z>pp{rj)_&+`x^ z$dd?ZiLFQy7`8C1O{bpR{HBh$k$-Ye2nf8mc;N%0G5x`$w_--nQzil4GXuQl;vMgo zxI+qr0$qU_^$wdpM6v%!?|Ii|d-gZJr#Bzv(J@MI+N~+`U0bN)g0ZHk%OSU_sNe2Ni68)MXwH$1W$)Zbw`rLO^=H2Mb}Ja zjk0ynGtRk+df0#;2?XK6_*qOwXhBodj_3J=zq3DS6SR3vlxx>0C9|duTeqwTOec7r z6dh!f;W1)y+bzuSUwH4ruPmi@oCDp%a9;8<(xYm#(|DmU^*m{37BdigSCv53rPWd>kZ;-FF>jH`@oaZS)7ro;-f|@61Fh zY*8!@>$R8YWLUh<-nbMAMXko1E>mm85#gbqT2PvW5ob_+;+3O*Cr&&G@dtYVjaP}sn!dNQkVWw!;oIkt9eAtBD81IoX5wFV*B|mJ| zP&qtS<`+(lW!|S$&K}xQ?pK&AaHfY^GQnFSefRkS2)|*0;Pl+_o+jDOYPVuci^BxHR6R{%qvYLi| z^`2>Sc+F%^x{fypJH@n#Z)Fd(bm6A`P7poE}`rg)`k6&xo33*=?daP(yptZL;WxggG!PiFL zjR5lT_QtXboYP7fy`U6X}jQ*PP4mJv_#LqC^TD1gsR|bG#eW| zAOT1Mi9hZaggiG~mLtV?Lys;KTw#~iNkz37*o<6eLpkWX+MDhoRS`OkUXdCdemA%J7*dVMbh|e*%2oJ4|$;%mXqq5J(?(~Bw6oX zO;Qa8rSJ|_iHJPic$G4VXj*o>yTzXV%KL+%O8?~sJzIUk%`xJejgYELBW)Uw?Z@G? z54dt1WE1Q`vge$QzcV%PxfH=ib|OSSX`7281(vb~hai(da+&J`n2NXv_z4&PK)KDi zWpDKHYQ_Yx!A{%FX9M`&tIm5^=MqwsJ=qSum@H47Ea&xfwk`n=>^GIEl}H{fG+?K= z9>C7GS@sI9$uFVb$7zerA_-^eFP*7Cn?qcwzFoTx8y=gp)+fmv+a9^9O`fgfLf(nj z{!n_qGw2KceO+Lyx}V3&Yfe|<>DiPi3`5tj>2)9`RkV~a4F8-0mPz5Qg)Lk;j^(q>9UGDcR!EUbO_+m=saCMSp}?6wYS*L?}(!Q#&BfmHQXgJ zkj(cRnySL_HRoFBkfQz(@ycjDgv&zt0Tr#wyB)fdUhE4!S34?UB>Dg?-j1@jnlAli zMx!v+P;oLL#pj}9-HI4#tuVz}XM4^Y5rm>x!&QdgjStZKY%!9-r(fx8RCV#~to`H~ zE#RRGPpxHWb7uR-!%p6f`c;(XlO@fukE&qYBglNqsMap8tk5F``$t7nw^zH*XO2lm zn+}~Wpe*_=;|EosbyvC8vjO(UI3&5^k54pB<~Lk334i6!d~ls}Ond88lf=$mHTNxC zl-q0m{(ikECf}ZXSg4EOXf=Qekj*FWL(1p-CW-l~*ftLK?Da&$k-+8EdRLed1pKiX z=64?|eR;f70fSBOtmX)MITRiVxGS~8o>jcX*&WSmJ#LZ6u6+p$5|_Fs z?2nj!A5M^bP_R%{m;^E2c=4=I;pq30kkcw}FYFfg;rFv+L)Hn9FDzA0onyh+A*W|K z-`zL9^ICJpWWOE%I2LkdTJo#?MrXIg?-$9yS26?(d;i?SCMtgkdS9Eq$8x<`8MCBw zFxd-(xJrice(yNG@341-3=aNBZq|aE?yW`3euuY_+$P)PnENxfq9^S)9oMv>u^Qq|am?sjjY+c}-wgx&DQL2s_?8vFbY)b^^N_TE%H~fx zZo6=YWSD?;8yyNTo{>V}No=}dVYt_;*K)2jBG4@^?`nXqU;6I8i2QpSHN{J|1onHY zi<+lkyd(ANLnX60+0OaR_}#ME@&Xbz!zqQ)8Lu#YcRLOaU2xkY!m1Fn2Ak=u+^pr+ zw8!o&AecVZnP;1&#H{j(fE}aoV0kpoUM&jfmF&~Eb0*LwB`Ej{ifk|6b)$E`X=AjCgnQ}^b-QE_@)WEwR{bj zi~Id`^-UzJErH0(NB^bH{gWxmj4zuY?&(uhtch7_OagdG^nt>s0o}0rJmGRuQAB%u zx>7xOAel*`xvCFNC<2eR7|xOxc@vdkK`LZ}E~HWu!Uq@^3OQYA+Rpi&Ex1Rn zyQl7x1?Z-Fxk@a4t8~5F5+!sp?=^SPs{gD19^k)i>kKtz%54Ob#$L#Vp8Ae1HM>k6 zrcUnJ)aN@1@?foR%5hz7z7{rl9z!bf`(xh}6H$rd5Ahi<8%i<}uP+Ljg0ECL6-rO< zqFr`JYBy53d`OmECu3&3=rB`j&3@QvbIK6X0pI|R6F+tW+vhWh;&f4PMcohVbFE&ob@T|m@k~B#Wx_=euJ2zT^D#+f!T%Oo^ zL|N)xH$nBjbnVZS0A>vXW;Wo;@5lGjfgJ_vWtzG(RNDChZUb}!9<%xi;R~ob}`5EtB?Xox^u*hhf1d9!9lTZgPj=QwNQ4%BAr{t_>MWB_WOwsZgn@8j}TzW z0Ht_L${7QP;q(*|P<7upywHPZ5+^22Yu0n(p_O!2Iqc~xRvwqr^>CkJ>59OY?J7Oz z+&KXvE|FSs4L&#Ft`B~X_co$98kS56_|M_MqO`cAccx+7?|+YbH}R55UM#v8Y*sjj za=IR{@q`1UxNopXOM=5arfhU8rb=}!SneI)m6mFtHpx|7U!90NOr*fC$IG={K1)XJ z?$hx%o_8K9DW;QcwJVnR4wwAA`8@~MYg65SD78DtC#LqLIat?P!3*0@<+9wXY(*nl z!+F76yy(C4`uaDzS?ap`K!w-E*u-{FlEpk@E7YsX(Q9p(Vur$y=9&E|dc=UDpd9?# zuVS+E=$>RaeLT}+>4%0&)934xhj`(+_Vlo4A_DP(1T2HM6;<2$*Qt|56Nnp32IBqh zhZ{zuUSGik|7)1lz|QBR_GJx5s}6~|!u&QE=CTVp)aFtPcU+0HdJy#e-r{j0$Fhp+ z3aeRtU$*)q%6aGMS-KS_*7`E`U|V7_DsxBca{;=2Vf@b4hmqZL;?x^D%tl|U7hUqo z8ME(Sc?fABl z?>b`P3Y>L3O1_VRKVKbPClTVhK_=93gv`mRvDUTMlDWX4oGtl63W_IwMU=*~n3PIK zmRZl5>l}s!TziN$Lz=*Yzhm)*4Q|8*&*-T_+0Qe9peVzDYe$9Mv*M zmyIdVp{l;O6F!nr4L0>f5F~cY{?-M}eaiyg>h*;@A>qJ%Ef`V=!1G-^>n>Cv7m)NB zbbW<;RR_v`UvOT}>XO0PDEuOzSq*F*j zbr2#Wq(FH5};_+g(4 zD?tWq@3Pca2dnV2LOn2X?_@>pJ(bvq;f)Q^otfPV8baihG}z?Zfm+D%KTy64+Y~!61>a9NS&aGz4M@Bq-xP7%f9|w0{QXRDqo2ls zbXR6%6%&tws^OFGRg&fe@jhsE7}$d7O|iSDso*g8uaDX=ORYAHVtmnx-=R5re5p3g zDB)1;cA#N1_ER|yt~;1)71edvq!{@*0Dk1*m)X;H!!(%SUY69GHUTOf*@;OJ9ZA?N z&tX&+zde~QrEdEYF@M}4KxwWFh361LfvaZ}2sW7FnhfS@8oCh^GpT%M zZ+qKziu9EO@;E@SZym6G78v421X4%}%oiNse+5EIEm#u1zg6Y;zk4~H4hme+sh()L zR9T#vX}#+u1?CL0w+^-RV(LBkB4tqLagtJZa!{py-xT4{0kKf`SjpF*j#&NmW;H4x zJuiTKvnztf!KLjUMym66>p3q8wIJBzM`_-qgpcTq@%jo}?**(nYrk7}CeK$W{;HrY zAl9JWdZq30g7w%h)Jqk3pi%jVJssCo0=!a-$y$IT(c zk$?BPCw5;qVQ_=Z%i)JS2O>LxXT#$H`--eJ(e0=gkZvo&A^1S{-nTMHc;^HYE-)k6 zeRI3B2vlF%_qhx^xM~b*PNul2xs&{sV!kN2)Y><~w&1g{S)CnA!Sz)-h9WN$9P>)5 zCkm6bTRvcWKy%7mw59pViCfvr5)l@&Q0B);l9B*Y7anxIhl+X4#g4@<3IrgCZ$}bO zh>AE_CQ~WC(jt!X`Yq#EXkS+@HH>&DqO$B&W(s^h4{{SDcg%WoCdxQ)6W@mZ?p?k219t0uT(7K+Xi5Q6mRD} z6Wks+KzHIkpe1=e=dHR9#f*WdRqr&oPSeQlaNFQzWEbVm&8Uc__&`DMY`)7E216I! z`-^sXu`_g{BTF-x2N&OZ`S^gLGeJ|Av3{Y)|G;=Yz^}h2Wg>=W*=QP-VRQkGwY-9- zMxgu1 zqYSD2C~sLK-=uM#RB~^P|^b&!l?Kg z({%Q!M0%6!(6FhB@S0B`&YvL=M#>}i=NEQL!BZ09F+U~Y3|UC)RCV*kYn7PS(=n+L$Y~*!hkIPWx~7fo?e|2DI*+{T|E?>3Sw~Cu8P&4 zL22CfKQ>=-n&Bt0YzlhLl}xhKdG@Ouopj-?;1LWuLcjlD^;gW`rqB1?YCXWUnJj!y z%&zw4O~pCggzMl>$Pw?vZ$qx{=9JN2xU;PI24KmXjzpao^++diJS-;tRtZQO_Us3k zf~6p~04tn4BZPz%w`q3;ULSCOF%~RkwGnbQKsbYF~7k>GOqM~k0HC@OHJ0i zqV_oVqNd?dUdntO|rP zR@Y&=R7gsw)uVm{2)b}MxTT<9%I5KCaC5+IuwPZUBF!Mzvnw00?%Lf?ADp?*R4*%$ z8C{_^U-VqNoT_?ZGoMy<0k8DP?x?#RrtkF`%L?TTHY9&Zx8eOsD1|@(!;6GbXsTAY zB5U5RZC^A>gCnfXaQF&5m{j1P-}I?~Q9VV|2CRG7- zOePthuQ8KEn|xvo#N2+K{c>{xe?3w)H)l);lRN?lE2`bCZ^TFnlCiV9EX$Kw+68-((|_cwdG1W#3294zs$MuPvNBUZ%wGt2+Dmp2K55zv?1esKrN`;AYv%IAW)jhD0rAx+-EXRkOzR7Ov2Vb^Ef_6!UM072dLcTJP-(~I1w5MFLuQ$A@ z8Uy>$%O|tv;gTE97%&3JsFmjq+<{clb-Ux|d)pZdul1A2&yR7G zcT*O_KUsaL2L)%Bs3iTOJ857>a+OX3R}daiz{b5!kytWm(fLMW*3Jq2g@JSWS|`s| zera~7_aS^jMD{J?+8f!v{h0h6oF@)U!9M=e)VGu?L zIyJ7%mWfENtuUEwi_|H?Mj8KU=2l z4y)6&wvK*rnRls!T(>EIGmLgUojr2eB#za+>lGrTZX=u9IIZdU$UrR^b)(7T)U=e3 zpCy&1UN((IQ2g=1SFgeDm5Kb=b&z(0mb$WbM?D_@;OSk*I}>bo#YFs-{%@d9;e5xX z-Al-s^|r`+^z#Y-=nP)9?CUKEh=|0g`g>F(lJsE3I-U(!3wYOjqV=m*v;vA9=^> zbv5Z*h6(A?YP@O{c1N=0knxqg(E_90_9I>K@=ovs!w0ncrQ_i&Hu;D*y&vQF9SJ|5 z(fyE)N%x+K)tj;Wlr28X#Bv6)uHC&z^ly;u?(&~xxL8P2XAwt_=FHNrwN%aSK&wm0 zc>3PsSa5&7waUE_L^`khWL+C;+A){YoS?4>nLG~On!ENy*^^mPv3W928#R)u6{l*$9btJO#!Lk${P%FmzGH-y2x#iPSZ z-G(#-HBo&U?pRwP{r)7iPCH)i$v9hz^@0k-{BbNpMJ)iwNb08WKAJ1eEN!WV#a=K_GP1Cx8)< z{jTeEoILOm#{3yYWT<=UDosDw`R#A}w}An{Ry^|r!Z`tkrE2e}O4o1m_Yy<(gnV38 z!t@xY>$?tXf&sJy`l&@^s+N5u>i@(qw5f zo3CjgY+I5nH}Y9_)|_;8hUoCj6Z={D%p1sJEH2qyB+^Mv$nl`n zr5#R!KRIe|sPP=0$({dZ{euQC?Co%XqCf1q^(67s)+-Jl1US&R^z91l>t5nXL&WES zr_fut#?cFCWpF2#ym0E=T9AP`Lok6NwIGLL_GYJOM>KPgHI-8SiJ$gr>Q+&~O@~~< z{NVJv28R{d7lyvo4a{H@u9c>YpXx3}03FZ{tVn0N^zyU|ZyW*|9qWxDz1qi-&Y2e1 zHnLl2iqATP_7|bzOzO9|HnPrD-y+&R?Lro}`V+u4YHW`M+^K!b?M=AWfLSq%UBoPE zIc;9RAHg<{qsIqS4#RzfCsLY^^{Rd4;vA(E+|3o2R5Cc%p>1K;3_s&ovQLaeZgiVTEqZ4FrQ=w_z~N&jZBw_b9Z zRG+4DJrpPDEa~>1YhkJLO z`x#vgPZH0O?;oV#Ld7ddATn5}@co|Y0clWnyY!U}3?@}-r{}Tk2J`Ihd&ljrCpx8} z<_B)jC@a?US$+@m8xj9tYM;G24P?{J?Qr4Jg2>@vj_+(+ff!S@k<@fotJmNQHq?#l ziu7QOHtXevO!ESd(c{d^Yk{&7eK;=vLP-Ln%6eyLx$aOFmrTXdq+zx|)el-EA7l2FuxPXC5@M={7)Y1(hB4x3I;^ED&d4O0`<|%pC7H zO`jIT_2Se`C%Df0YJn;U@zp+^N#^S>W~J(Dk2&}{Z+y>`4Jn?`ZH|uEru6&j$a3{( zfA_ctziN|5EPAB#*gHNC>=XsN%_AIB>E3|0yN={UW*4Uh_rUlvcJ&y z_6{;gRH{Dt=*Z_j-g6qoAgILXH>RNGYW+jbwfEx5ZR9E}ZB-Y!A7OI!SrzC$tV4)2 ztV#ZA`NDlWNwf4MR{pA*Rq^_RV)?dV?=VkPU3ux8X?LFgn7iGFR55XiO8FPpt01nFli{l4bS@wi9OvBB>k`FobYn{zec zNHu(LfBi0(`^BPlcAEGtMZ5Q;Abs4H{rG37>uzS_f$7CA;Qii|JX`v(SAlMdX0dck z;A4>0AKYe@63giA0kTe@sWY^O1BR3VSRFQgnq0&Ayu4{STbKtwHs zp1Q|JUgzHrA^BYI%fkvquNuff6iKnKb7XV*%|xtz#GA!7_o&mFp`T41RcU*SpSCUo*!p-Wo>MIr3_tgx9uUUHFRPxLF5t?OfD{^+zJbrrJJOyH zv-~`rsWHc~nLuv)NaK53R?#%g;aroe>a)_FQ0@cT2H*Rhr%#N+CBtH`vp8`ZO;a@s zk{tNN;xs8<^WUm}>#W%&{gCsaVpO3!?lTOz6N7ey(%FUZ`rM7lb*uk{9YI#lh+yh? zxt3b)ZCCe(MC*G^rX^>VD$Zy9o@5XpB|tdpl49K1i~!~`3c}}1Wj0^*YXP*syQ)CG+~`P}kRhI8R;HwB<)2q`k3Sl{9Av=E^DJ>R;9Rm!2p)g-dMeM|e$fhOcU1Uk z{v7rlv#d9kd~?pY)xQhFW}+^XxSmWl&ZLVuiCG2Y@5dT&ZZx*`gWGFB?TGF%&4eN+ z#o&m?{Ew3>R?g_l7XVL?@s?lPS96)BWg6efoeHbv7B~JT{jCjR%er+F@|RiMUQeD; z5aKroJp1O9ru|>^b<&y}@hJ`#?Dt&`2!CFBs>=lic>A%CnRDnl@zDyID8ZAo+^_ zYXW9%z7>Be66Jn-UKSHjKYGd}ykTiZu)nX)5?pRn(^@_W)zb<0a~m@nWM(n^L#5h# z#m5f!5*e4ymJ^$i3$KXZw)=pli0X!fjuNz@;oWix)_r5O;wxvHy<imQWByi(8J zZC4Qvr|)M0r@v75Cn^NmTfNHMO%CE*EJoj^Tzy0l^0}GbWo_6HJgJ+mRE>2%S}5S? zmm72kO{6tEr)fdr8-2>78=;)Mb@it!I@cCsTSP=6Ro|wHomBl6$}By+OZO9Cp75{a zHjiW^{DvNlMs9*1x8ndjufgVVjYVbZlez3JY#B!3j25e@9H8qFdm?{IOO3MH40+u; zFYRkx$psO@BO$~c`>O9nMH-okoKF%Fm+(?;xXskl2vqpZ5g=*0`{@cYC|se*Xf)U& zI3C195=G2R`2FVz1Gg*D^~WEyIO{%Tg3AHFMAP26G<;1z*b72~@g+Zovavq?&TqsI zr44{P9^v_$@llT|3Mbt2`O+-1k%>gDYg>&FDJ_3EE-KBxzxaaY9<*`6!CkX|+NjA| zt5F)~RQdrg4q$DogH8l_ZjO5*EsRV-VXC$%VLQQONWB|5UV>pev%%rd5LzUp+d;g81 zjvekERV+$~htuECN62!W>|boZry@XC0bl_sKP8|&IrCR4LjB^Yy=7c2*{^~cC3=S6 zDgS#68NGiqNQ+vQ=+0m_-L3}tC7RfFVpBimi&O8)2V8D(*N~qUvBxVNJ(>~R68-%A zbxULnAE6gVYtz74q1?txz;5ePTFOc5w*UhkFlXE9TL!zPb~4u*oOMP(R%yt#gxm9_ z=G97jTV|-N9KFpD+ri@f8 zYspy+Q((o1Ll%oeQwWmJx_B`Gu`s1j6jn~9@GgG2>;)!wg%mt+rI6hqA;-BUH8O-v zO~bQ|b(_PHV|zG(k-gJ{sBvBb3n)uxL1~SkM+g)|R4l4jDup?y`!oW22~KW0+ANb8 zvugIckq&gx;e4%jin;&JG@bNQQWO16G;NVKXs>M|&k38@EFz9zzNel}a?4?l_lh z@}WPtLhhwIY>cQ2pj+^PncGP=Iy?8>ViQ@@rcJOzy)s}^n12oEI#ymrt&~WDZFo@kBgXU)3Y| zX|delX4DxS2053xUY=*WZ9$KWy6C6^6&4I8Pj-e@X_slu(N-#NTH7P$KRF~_Rn4|$ zZ@y#Ml>OaYzM7kg{sL*guriKLRhDSBC0!2n~#Tj&UvE??YxvT9|&n(^DqwD*14|S&Uo#2Upa0hbCbo|ud1hkfms|d`|67fG=$4Tt`5fJkFrU=`%A5_n01n5U%OOg^bm0FB#^u)*E`~c zW+uflYZNp-34fNd-!cz^RDV%dS+%eXF9k9k_nc}M%%!y-$+i9|jTPP2 z_hzYG#f8PYLp@J_UMOlf#2CPVD*oZsa7DnGh%70T&RkyzQe~r-b`3oXbAxY)k02SP!L+%sL@~~OtBY@m@3zH{Y0Y^_tWYkL zY&28}!WN{deU72E#(KEeHiOK~Q^{2uv_5~;h_ub;Hcp0`!t+9RJloC`INnT8*y6Q< zD&=s8S)B6yLGaylx+epZlC5%eKqO>zUFj~wepZaV*0fDL2`23@u2qOt*hZ>Wfc@p?;{mB_%`Ff-Pa!@vo&Sn!fm5fC&@3=YB zj|;C8TQ)+SEXKLAsXP=WuXLz-frxQ&bZ%cleTcNE@+40w6b1VT})#%U#qLS(y+W@0i*=| zNV%jnN7ihvmlxc&kC`>6Y&#ynS7=X{V6$||yjJF6=%45MRjcK#kVQOu4M#!BbyI*< zj|=Vt$w1U^$p8bEecga7@*GyAY>>5lEE?VG03hD{T#l%Hug?=~a@_rTd6~3|cSjj1 z*d7GP@{sU4eEo*+C8}reX#Bfk+ev8xP-rGdTDG@afrph!vo;^iyIvIHq&6$N)FK-9s+NUeP*a z6(k3Uv(FpE0!_v$Y^go^yj(^Ep%%%B+G2&~V+w`VWAIG)@x-@ix8*ee4j+Yz2#WEt zU;NF+S0=YLhIq`DPwi)@8v3}M^i!7}gLON}rGzAy5XbjHtKE+(Uyr?r=k`in3x>Ta zf*Bz#bdGIHEH@46&ax6dEK$)u*dxd(|CMQ=WJ*@4U+^huqF)0bKg~6D`A*G-Q@O2U z>GyU&gUpmSywP?;QQnN4I_7g{)~E2N#Pogzj%bV}!go>ou?$Hwcc0>e&YfdPZNWBprL3?b=;gu0)iYP*8 z7xpCc9AVEnI(vFxGT;!Ux8DQkMkBkUS@HnSX7C%y&gM%Qk@9+=B0wMq0NCZ7=Un{- zavVLk#u%(9IRLC=2tgoosFVW*s%^j%$*S})*KxtG?-9iPwIuttxUC7VLP`1L0WjY1 zL>rgo%wsAVoflpqWskjL53F~nF`&-x)v@Th?I^%#T1?=um+ypsVi^8>rp2%hbG51U zE2^QYj@Gruk*bsXeq0ml)1dph`r{zkjWu_%_?rG~!yM$U_-x!A=FQTyVZe#8@w^Uj z?%K>xAb$IMRtv8QbzM^NvB^tlpAjg=2Qz!fCBB$X{Gxu3I|e1#AF|f~A&Hn%%5wa< z^uz1go5N$B3HB`kXWsgDl7_?1$n<@B><|(jSpXY1CU4nTBYLoWK7}3OA*3Vn7d?ua&orHpd27pZ(^9gMJt0c@?o~`3cw$3Ro_>C zZ*p)2FlpjsCXzdx^VNzolcb+q|J2{Xax`F(S?R>F5=2as0mb}e%d^zXg3 zx&iyi0N~>?Mbg&{oi@C{Nvcn8mb0%iM+w6 z&7Vo}F0)fSv--D)eHMCP<1~UAtOp?X`9uweVKKs|dz%L2R=Oc;NF1O_aOWC{SjgRY z<6Wh3WFof-6yDNJ`n9$ekJbm$K)ZsFUE?km9S}f%(!EePZHl{(@A|g7aOchoSFcf) z%I46Fp>SmRd*^)IEF{1u;jPi8l(8=>yH9T*>>-SBa zQKb9I;>Cch7TTXxkC2QP6}w{0Z786 zreuYH(>|rPL4oVfxYRbIySg%M(zfja&d_vM`@5g-@$9b~9Zb+{0PWJK=9atb9Fpil zo~@!pT<#f76U7Rwd1*{k$ubEfl1{LxPr-)$-T#?oAY#GqbDSRG=+Sd7tc zzbx4Ro3+wPZjclG>uTXGMY58$3&7-tYYtY-7Mh9+)wdlrzv6h7;f_6KF*?I`ZaMQ~ zKIxTuKw`3T&iPWyPbilzDe09hn>MbyUT8Lspv44J6s^6<0zy8J3pgSGSXg$*A+ zNp5WhC(LJx2j86to}OwEK+MXCk-4icRBMdVjWtY9!!s-lB9326k0me!cjgyHYVH_R zyPXC$f^vKD6wE}8!BJVMZ|?CdN1Vk8WK`kICretF%PHkU;#9aARcFfm<+6?>=g~%r zkx$ziWvE;wrdQyg0jrazy{9M&d3A^17&i~Pz;5@Vtw3ki0f064EijKe3*(&#ELQY5 z&-ZOAZ(d#OVIHijPD~jx-NX)@G^MRe+6-@&!)?+3Abbd;xML=cJ~(K)_bk|9pIg^f zwDgAZb7-e|y13P0g0A3X%Q#;4AU*+DwXxDHSB)Rk0 z&U(;mc&xUMO84M&iHH>-vU>Y`m^pqr!JH%t3{$~M4~QJlw3V{}a*#{IJol>~y}8q> zO_fna%~x*rc?kXwT_-JZ&4U|9Kq=)X5eKn=2}v-K$Appr0DLkVX+KbnTe`L2fgp|c z;Gg16016j(r?i{;;qY(94s(u^amoga)|?gsZvfs?w_W7xQy(;Jr5*_xt;wl~^MW|` z3S)7wB56r<*9nZ*vJ z9m_YwUU^s5P3{T7#%1bmCU@oeEv-WClfZ+c*DG`tW}_zd(&eUd*>@207D*EJEzL`c zm9Iq_s(@gE^&R7*si*CXNeoXFX!bmcj?yYmKNPb99wcPO3d)bzv6_Do_*l|#n8w#%EdZp<7rL%cE zDu2jxol9%WnF7D6hyi`NU&z}q>>fGv!Lza!57bKXs~<4swPKRhrOc>sF=68r&iv zz7KSHvOk9VI~+3+u`}K-P@c#vt~r$Mg4AJv=9sRETrNnBwHIj_i^MS3N0XfAt&@(t zFA3yrPAc5r^5lljHE!)lihg0!6N{%eKEEVk)oc2!YrY@?DnwZ7Yz{PL$jE zVNSz8YH0+ntc>%?C-LSMB$<|}hOGz|rC|#c9C1Nyx#Vq6#?}01lbfYX$?&6o?gdeB z3{#BA$xFn`!`?_>X3u>MsiF3R4MXEXDbK?MI+`*X$FMytA>`_8)|tVcT7_ES>sDgS zMP$KiTp6uyR!wW4SQf4MEg4MHhWCHN8jKIi7>bt3k?foX1hwTi#UPi`#j&y_$Qhm< z*65z`#~hM}mVIM;d}#mk_8)a8ZljDu;UuRhIJq3U()d`F`mTXG${v!j1<}c2q0_mx zSdRD}L6RI5DEv*_{cL|VqI88^})LL%$Wi+%CwI|p2lw<=%FzS=%z+ET21ts zxh+oHy+uoG(i|KsWfCgZtJh@SOk&J(0H2P6OG(-z1|OSw0+156hq_%r^aQs1iCAHK zVYDyI8L%lBt%ik%S#j?nmbxtGrc@vh=I!lYbXiVn+dm8@=KFTjWZG$04V*S8&CQ=K zwMrlalAVXgC%^I^pTbLEq(n%bz)2Lr*wK&KoR@|h00wFFs-L_bM;EK6JOy+FrLXGi zHn{O~8{@Pf%soK8h}tG>9k{Sd>~0L%iVb{SM2jsc9V?c z#*3lZ+M~i3{BNg4G{%fmnQQ=l0{gh?o+%HH_rk#vlcwe)Hxd#-yLXsG0u854k69@j z2+*N5LkN%2GKn+6kP5P()!J1L+>`n0Ky2w5Ab zy=k+Y`KtEP)=!!6`QJ%MtV9>JP&~@m$uA zBjA5y&s5Z#+4#ChE1@+nzi7U?eE;=t=)U3CryYR{8S=#_^Dk#R7oBmdCD!N)%ga+v zEYvj89Jnd@{rAQ;@ihzRY7SJ(jHHNo+$G=E8X_ijy?l6`H#_}}X7=6ttgNhM5)kL? zl@=q6tZ2oj$H1;Acl0W#FGM;~uF|Ktiz^7|B&Ac+|NL&L^$}t=JN1%sGjebB7*5)w zoz@R7ZJ`{mXmC5Le@2lAm~7pe1x%I(al(Osd&#B9;VxRHWic@;d*U#(D6 z&d%|9)dRf>$wzfBOtEGP+q=MHC{NkLdrP8_lR(eX9CuC&umU5s8aLK+t<%hpADur6Te?tNEJQJ$w)TmUDETl0MvY%VhfzeN-!ZKegX z+?NtW98_3ClAUr(pUkhK47dcP0L<~}&mPTJ-dA}y;Na{K?*K!#pSs4(F^7YQua!W< zrAP06*B@121E|8*friVEveC#vq=UOS-j|h-$w5XVLgsOgF{ zBhT}zK9>O)bD1At&$`@@38L|o^*0ugK5k3|ENWFG_zc=#*`7Tb4cUR1ZI-P~EWWR1 zDSjJFber;XfUxFseQpB0!?FRwd@LOTP8OZ37^v_a|Fd5`>HLG#`5NV#4@<3X`|*B? z_Ui+c?O*w8Y7EYBt)6zBDqw?no~_eki=X$w`4e)yji(>6`0XdoW0CTOKgvF&u<0k_ z+?WR)!6c?DkM0ufZ{@4%CI~vR;WM3nY9g?Dg~i0ztieAjTj>5V6PNii;~t?d2E%x*ElV>X?cro_c!Q}I zUGtfM?!naFo+B9$-!*w$Wb3w(C zqmy2f!%H9*?TA4X&4Dz*E?Ob_Slh2pQ85V>x$zZffKbj$vXS0mA`g#-3VP~75ywqE zGfKY|ibQ-UWcobAQ}StL#-f-q`4$|QWR&p3O~__|sScW4&s_a0QAp5)9{w;5wKlgC zR_PyQTCgr+S1uwxtDY(|4pcEcJTd4X%fMLaBr1Nn6Pm&=mMxzg#DInOO850?pY-D> zW*HMatC77?Fi5Y7-gN|1!4spp=a9OCbnlt1G%%S%{bzoT^KP3tt*@-P0lHSux)hh? zqwm!N#SZU2Lo{Rb1YH2YOgx)oozqI05^^(pGX!9@8fX0|D@%S;FY_}5kV}U;q#bZr z5c&my$sbK+6IL^=XvZ0iK(e}Jx~THqL^dHUbQDakLDn#n5nwi0A5Mn z)@&tTV-d<`L#c4}>$9$0VQ)}?+Dli&T))U(0qPMnCqT_vqX&-LdbAonGdNb+#Nc|U z0TBD1$xJy9k$YWA?KPBtyCFq?1G5-GrRY@hRjPYgn)AsNKc8C^H)?t?dj1$WsX7^0 z+v`@O9+0nDk#vpFSxW0DIMoQ!TQ8M3LE%kCwIOO!Kwwhjj zMzR7^s_1|MY_(xL^s_0mKp;0dE3}Qwx|tfpD_s19JBaVYxyBAS)Rc@UCWM3!pZ3mg#TQBSq+V$e5nQXP;_ae#D+g7ASUV5r{;eDd- zcZ=N`w4YW|O(2g$ds+EY+n$szzB<1_#4x5LSq^%DPGCZ&0lKbsmOS`ZyY9}pHr}je z07Xml*?@knkilr@Ly>$~$H>AXpio@n7C*|u4Z0@RgAF$_3h3-GY%?9q6X|HaT0JJ;KlI{HXc8S-TM0Fpg0aZ_4sl$tJV1eSrUXScc zk2Vh(rkwfeJ&T=5t2>2F-`l+s=Uu2Gn`-`}>-a08B(P|LaF}5c28f4LaBj`Q_nQ zh8bq0w$)VXkN7ZK=uWl9J&J~2O{TSlS&kSW^fP*9fc;0=tK=OsCG;VI76FS}aGev3GU;=O##qaQRsXFt63G{2)z)*;4~f2LHgin64w80Q>IW8HWmWel zqam4)Te02(yQ1qF(~qnHj7c}luT$=gZ#Epd*f5c%!46KdtXO)bTYuW=$x4i_C}s`5 zz|b1AQPZJYm3O=$6E(w_9M+d8ureK&{ur<W3sl}9W~Z`|mi*F}=4*aZLg^O;uE z3C`lX^|y?sBeuqq?bR1xmeJzvMyFd5qWdQY(eu?zmX;bItkAa5-V&sR9j;&8{UaOWn`$;z>VX?9&ejuC8aqW_$q} z&s@*>e6dPBi$@8lT72*XYf@;bYCM1PZJ=slW%YCE2NMbbt$ikhY(&nr#Q=cAO+zD- z4@d+v%U4z(!CyGTWi8fd7^X@wm!@r)td`Z$E+s;U-WYa3mY%~1+DwKd#WGQdXF0eo9|7ZBz? ze+ddM*}LH+YErSd`uqZDz$CN73<;*NiFuzHbv%Pl0d1}rihdNM)cCl|${Y6JWK%jV z37h9OC_9>N8+f$#`r4Revszp*r+{5%=pEGiLzb1VVvm2~+-k}Yn zFY)jzu(+mejY?O*BSYjhrK1;GH-WMPwv7!~baW6W^wlx;qYiTGP;64c*IM)lyas?Z zm z_WSM}NL`%Y=-QE=sr9IFomE|O4s^WS_G1U&mTuT5Q;cqPOi+H|`^=w@$Mw3+ zObC~bA8f4{F1c0~%NURb*XT-haC4LQIi8h<*i$x?QnWv!`qOxG0F^_3uB3Eb(wuwp zG@f$bMYn4W_bMlKl?k%4{2b&Dku}uUw~Yh7#r|kskIxr~Y%zc{C}(RBoCTQ@+~~(No-vWTXm8IB z*FHDg;MB!~o=7o9pGbmq^=;oAWd-Ma)r_BN5+e>-S<2M&e@ITi_5$frTQ(-|xD1T@ znJHVWGDQsF=#y%>u6REu5Q7wo+Mj(me*K2ot`hU`biOnrp)K3L->PM@e{Lsk?y|nN zz5_4dqF}l_S7G6{YM7^on_~3y=)xmOZ1pjIub*jQ-o*e@Kv*STHS-PvzaotZ7~(|4 z?Ns47UuVUqSIeLRCeft2BX{dYH@C{R2IcmGpj^^mY$pzs|>6Ah~z+=keHD9~I zx8?Fxi?O8d#`YAMpWr4(DM6=OZjoN>4PDXZ0aX;i2Q%vYY}T_Vnhi!~VSFhiVsj>P zM5A~L@2Z$Qq!j>ApHd_tU2e2;jNfHeU+Vjxug6L~v?hDBnvw8&B7W`HmP@)5%nfBI zT7K-rllngQuvOE}LVc2>i`mQRqx3un4}HsBth*EvX*@eCN{oS-hu!DKnv%-=BVpTn z9=dgnECC^Lo!Lg$uvPoJse7Gt=PJjuuPBr72^y9hgoDnN* z+Yi#h|7fF4fE*Qdt*#KD>Kz;8tJZmtJG>-KJoSjZIDw)`{pPl`ZyV3WJ4bPLG2WEj zrBb8%x~#ox=6WJ|Oa|id51ZbD$AfiUkOvL}%w#8J-b&7j$)$KtIcL{H1(50o~GKq9`3;qWf3y^)Z}0T z+we*8C?Z@<$Uh(yR#Lh`H%v9&OwUcwLddN2b*6D_!Zr=(BDkvgO_3&DNATg*`cz@V zcbMmv*-9+JF2ZG+Ez@iR2Z6D|rp-o9UpE&{WcfE2sD*7Ww2=D7ejeaQOzj;3k832+WcMpKk`?B$GCrQ`TZH=4K}kvb+lBtX?WXWO?R=~ z5MtN&d}{M{8s9d?_T^a~>MTVMt|(shzmZT@rH2V+Qu)6g@_DBaV@cxji7 zF+b_$t|@w59!MBiD8*{q1rT)Uc$}ytFFS%}Y9jY17d(}1FY{P+@_65~HXbW2*p{Gb z5QqUpIPxfVQ>?}Ys4G;ArEf)kI%T}<$Mxj}wk;7${Wcp&X>`YxU(8lM>B_9 zZM6Kn8^x@}RVU7T*YaLcE3#(z%ZlBh*EK?*bgkN=sFAMQ4O9a}CAy+r6~>m1}NkSq?U7oXq8;3~$(^*Xv|Vu4sXU^-A#PH~<4_hAN-4n)F67Fw-zD zb9~g`{ZY_<<+h>L8}{<9j>ovlqwz_m3PG2}TckpsubELjGmU#Judk5$KyS5Fn*|uA z1v;5OODgJ43bbQ5s9UIOZ~qA*8xuSo6**rC6?$etphBP9V|Yn+R9{*W>otmZu0gfZ zf>=D?j4GpZZR*9eKJ!kAj_or6HxO!n^e!v#^yqFUZQeXt3Egx4eOfQ^WHm6Xcd?#1 zBHk=`eZI_kWj`x9h>is`*K2#4tCwD@me%(1YQ@fDWAY7eO8KYpLd#c~erhSNQhRA9 z(d;teJ0sJ3(w317E3R8}-x)vPKLu$vP1VOv@v_;%Oppj((?3bd(M5Mib;It(;9z^;6-vf? zOswO1mbrDkb~wW0)wjGdh}ccf1P{s=MZG;Gbt7g1)^v%ES! z91LN#eDMi3HJi8FbUbp+b8c@JRN36dFd#X1IWfwA@V>RZhu4igoMr?>n{a(RyDNEY z^C;A*G{o>4n>ROZJWok22VOba=~@G@z#Msk+FL#;YkSU4b>E7e>(eN!lfPVdTk~>mK_+m@mD2{wM+XHzMQ_bhvrU+&j;5^etEGO{ONJt zlo->Y2J$>oe>P<0X)Ht_qC=i^lVzt`593plvqhXrS!&nF>P6fYg3w=p6u*d{tF3 zXXCCj0t`vjdX}OczdQi);I#GrcqO2XsC*7>MN&~v&_{!vg52q8-HgZ3BQT2v(M$Hp zjK|TQx%_Qt|G&g>l8!eQ7@RA7LZQL*zBb)sSqJHg@vR0-sdf|zIvvi@xZ?Aiu02}?!gDxttpz9SAT-wLc2l0?l* z`fm5WIT-^7eJ@VgA-c{US~NRhk{TF5Tgv5Qk|v5_XfQe4q>w(zJ&#EDcrvB`0AmWx zUrd6c!5*h1-=h;ps$v`7;_Xugkh3&O?V@LEY)8um0e+q2Qfsj|_-b0J@4mWEgbHgx zc}AyY=oF|1Q+<#Dx<@@`n>)xcu=zaLo^@7YuKX&$Mf`+&ez40k9O%IHI4^kFL>&kF3$otavS@^j|*Sy=$RzJ26`!BSVg17c0o7`E=oO5~#P`mZQfaI!s z-s8gR3>?D}&hKhBU8R8`aQmV;bsY5Vq~7L2*o6fJqtVuWk~%qm*SA~Z)@>9V;eY&r zj(Ypf(i{E98*ir2r@UI@E0+^|X3&xCr}j zL;5rE1SN z{yxvbgqr&j_!y~7eALi(dyZOfKl?Q~nKPO@e>fwXqmgk4iOw*K?2Mr5!26k&rGtX=^~LSGwP?-ReIE3sXni@P{;KLctsmd@>pNv;ewWL95J-A4+{40{sK za-YD+`QY*`C?6L!IoLzW%P*D0+*)=Gf^Wy{=PpZk=6^UC(%#2c%z( zex6fl0bQQXTpuh;@$YyM{5Mb0MnU(*jtx&s`Bw}2{ki{QT7Ulwf*UJ)^dHvopO)Gx z{Ym(=gouSV@b5qJ&wmqP3V!Hcd*>%%b>NsFS8WqB1@46Whn@YCFVTIWB!BWx|5-Zd zjDXPE%xgB7{?#7-*^T|%<$3QX-u?4Z`iGCyrxm7fSAR&DB=(n9jYENxZ2#BZDNIpt zqigv~Y2|4sc@N=!O%aaZf{|)^q;dZjIzn)xBONb>rGUfU0q{)J$Ont3kG^IC z|J^m>y;kz4_aD6pFtVTACfGM$Z{ZHMCl8IYQ^Bx_pVy+Z7RQYW`}+Rf;w2sY<^Ne> z|5(q3T`<~t&cO%Qt@+^Nzi7o25jSV*>M(TbFFAeh7mvRlrBwOqH*d22HP0>VdmlJ; zQtuAxQ~&Ww|E{L{Pof6X0T?dIHa zT;1UQORL98y$SU;TI{F(ymJ1Xl!R+;WLHEj(frpo4;>$H!>`ej+y0vDT2%nqeI(^M z5B?AP^T!Jl%>iuV5$`?;*I$!e(T$P9x`%E4%Ek#(yuMNVA#Z5OUz6R(8`)_~gn0fn zwS9D>w!T>W{(nt&zWIRcmU!)s|C-u*fAfw%|1aJUI5qZ&WNjoOYqr5v#(JTV5kbx$gx|H+Bhns9sD?`qw2&g{!lM*uqD3+A zT$K7@Y~hyMoCNp{xL=Fxr}E{+B=vy^-RK98?7xXV{!^-o&U_m`(PMY{dUfW6ICZ@A z$+q0L)gGERGxT=}a6)Y#V43*t@(w{i`m(t>Orf-b+C?e<%O?m^e7qA|q#Mf)Wtkz8 zO=b$9k%tNwF~jB5U6ubRV+n6Q=e-^q`l~7$T$H^JD-Fivl8Mw1 z<%nnccc4^XAWw+P!k*l{)0Euya%j9{jut2#^Eo(>{L;h)lzR_I3@}cWDw{Ry2Sy|_ zzpZ7u*H&kkM^}4TL}@-^{CIukab|E9TjqA6Rt6drRQ68^@uZiEpyBFM6DgGlxuQVq z&P5XHb~EtU^V_JiRMFmRWucREe+mr!-em;(qs4Ek@o1%)QXY_R!wI~awg>ekEJf>E z0K7f}CIL%~&w?kT*QM)g>)D#6D(p;Yf>>#M#%^L`8H?N9W^*ylYKx22{zmXb*E$2& z!_^okE6vRjLzz2+gh$~G_Rx?5ooW%)LTy?vG9gNOg(NQ4%WIyk*TK$tD$)1X79#GP z>BYDlEY}v|ZK8&dUWn1j#-|xRQ%w6lmaC#1EC2c!-M#5UZQ?2HfJZNFDqpG2NHNr@ z>HVhk+0kILrEy?0{9BZ3V~rIb4kHMgBqBJ$nU@1!NW^_}B-M{b_WKZJ(!)P6L5f6F z`73*oBAsT+o=DnoU>uD9QL3oS!k}&g)1=L8#N(dG9`?9V*9wIo0fV%+92lo|SQ5&U z1|8NwZ3DlpTjW<3z8KeH(|k?2_WT_e6VU7Bny%1iE_Wi{3j09w@o}vl} zx}`ryx$tRi%3!P~5{)maP_MY^!S;gMa%V6sj4Xlcjy_=|SyPA>9BNPO1lhhUA%Hu2 zO`P+Kr+kXn0YTuTY^J-=GLJ5nW+&}57sbs>5qbvoMdK8RMukzX`&fQ)k>^$HQBN$G z05)Ry29W-WBELPnfJviFHqE5(a3k%(ifX$Z;wdpb=2R?oE8dAL% z=ODd{hllN(#VA^q?)xEwexxtbK6K~R_D8NL>BS7qzjvJV!Bsub(Xh=JP{N%Z&IV4sLj25 z-&aSr^_1PZ38nR-y!pGsw|0+Wt?S?iPdG#ek_8Bd+KRQZV}Q{xQX?5~Pp%+IOke0( zn$Ss+ZX+EM4J!tpQ4Pk9lSX z?WORO`-#mCDu5$ywcOVYNpck50bH)sIR}HzMCTMOcHB#v#eN==H#!d>bAZ#WdLz;G z%LVPr3+w2dYL+V5ZH^d^ehVg=h`fzF|GpmS{YmA=!Y2_n=lMIZEQmz9@aX{_UC&18 z%0pNTBLW!pZIo!LP8o!I7f?ih*kH{WF8bc!mB9!@1RS~Gb=~7DrEGNk-ByHe+o&g^ zCnvluvjIM{*pF-Gz0a>lG8lqKe-&ZE4Cy{tp$GP>=474N*!VcRZ7qQQd*MuSw6F^( zcu4(vcSC{AeEYHgzysnT<>Y~gY0CZ&D;uF*tC7QzBxPTYVU!vvJ1J;PXjqt38T+|Z8?9A8>G%TTyaVUd#lLb9~*d6V)v%cRrw>!>Pue?dd+L;+p; zdM3hm#1Y0(*{!ERDK?8Fcy@tpmo=y#^tp3^aAZ#J5AN) zu0Z-N>$@0ubPhS}0MaIo2UPSX-Z6bBzqotEFGY*nu;1{Tmk!aYoS}{Rso-(Efvchs#F3ZOn z{Y*uxZo5xoqhC1nyfJxz;bM=~^_H;BsTwF7?+gvw{_?pEcpI~s^%CCSBI(@efU${3 zMF%gQc>q%fT?iJeuCm(_hq0y0XOb{>rKp+teGen1JgQG|rYI*&x~GQH4nlRYDP_|3&kl}@31 z9%`YN*D0D-I7hB>k0~|URF!~^qBG8!UB8=-a8vy`N^3nH75TppLtCX5{q+>o>n_|U zvRiW@l*5w?*XNLtP~g_M+HLGk7T4||k=qJ`OHZ~9)OX_numcGf@2?{GC+9onT*tAP z1P3`v(`?2@;$t+SyiRn>JcX*ihzEuapEz=v7++XB{Ja=@>e{IYI zf+rRg2E&>opvfTBSh8IV@3KHCw7-*jsmH?e@1;`ZF*oN^wSOLqrP**nfYIokU>)Y` zL>_9`q)nKjVP}0(2S|L(h(Rl>B7dF9HJku_@JP()+S^B|-ug{7^h@1~bUBEsQl^ra zR2YS~*%=72+f(h|6q&PGZhREOoVE5&yeU>U%2wqYPwR{THJh@;Q0dgkY zTNI^eQ28qH??Z&I5AUFuT?l?lzW2y*yZ$Ao598t{56P@?mibmZNehudJW%& z*kFkE#c<}kkCoRG$bO$xsVwQo%+QggI_14NrSeyf9lc(fJ%%6vq5i#Bi&?jm1u8VL)Kal8+=cQLO^Jju`4v?MQ>Hc3GAYj!)&aDLr94Y(B8uE zCIHXKVKdpEs(b=_7yHd7+E$W%rUry4;;?|JSIhUCE7i6Vi+4O6<+7S%u!nk4Vv`AS z^r^A;Iqyy>l?gEM&Kb;=ajRd`8~CkQTtaljbA+q z@zBXHI-{sx|JHRBrr)@AIeo6)bA0-@qCW{|8&F0NE3sUvJT@OTWYC z#ATIk9GEqdZ;CzY9{ni5d!I$bz;%6SiMk>VVj4$gJWps3?%N!t(W$G*GM+Wo9yb}Y z9xX~=yTejmR_|;bF8h={-dI8uILuQy0lK5D_PqzBD?_=00(?D~JjwQe%ubPmcq3lj zAN&h+=lGnJO!eD$KBG1ua}_CUhHGC7I{2K=XgAT^oa zq(<@oCbbrk#rA;ajJ^bXl9915@p8_+83#(`P98eCMcO=TFiu9t&EK`uvyJX1Z_x3{ zS9PakUGq?km2WllmaL}6xjLxD8L%)RbWjhKYOIc8CHx*}P-Q(w2gtjWs}OYQ)_=bz z8iuM<$LTHXk}taL#eY1f?^vx3E_}BW{UzpC=t0CnQl0Kx^OSN{71rb0g4^>d(y`2@uNW`mRlsW< z{I!lJt-vx*TY?icUR2kA8f!VBaM$|gzz6iG5wn%>OPC{ZhK|4dgxuUpJBDGVlHXyS zgBR|X(9?HW@(KD=qN8*VGi~;CzqmdR7j{vOSC2jb!YZk=Zz`+R3)e6BSvc`731tYR zsjxqLJ|uo2c?l)o89<*Aw3)5Z4`VE|K{!dJ2qx+4($E0$o@P3~ai{il?ZuBm<(_0o zI^`~`a;thx+oA*-GlKvkp;i@qt$9agoTgwAic2kH8ENFq_zkA9s%!%P&eHFA&3_3! zg)Q+ZBq50K6OOM>3>SiPpK9M6CW~+?c%?jr`!=z12>LHEC$zNrV`sYLmf69#d9>mk zJBT){37#!9Xy9$m!QNOLlz7qL2!4F?Wb^0jm85tbsrUX>qCl!*wJX@E%oR+~HW03h=G!zrvP}`kz5^nSIVyPDz*VelvcjLl0WDcm&#Fap_mD| z>5ha6vDaXpbBe$Mwz^+rq|cHS{fJ@wu%eUHO~^6Nn9Ljl}8 zt0Rni!q{{$S&qx={~_!>qnggzeqkF{6f1~S6;OHyDFPOhj`SLkE+rtn28JSxf`asp z(nARdBtQa0L8NyGC4?d+KnT5;@NULAXT8rk_r1=C`Q$9c<^Jz|UB9}HWCw-el|fwOieyVIc%(*D69v9zD}1iM ztJ=}+L&EVIja0AGWxEXAK|%L#xogOs4K^_9%T~Pa{QKuO{)L{sVOQex&^ObHzp~2L zD#a`o$Me+NI`Eb3TA|cy+VOeQ@b)COHZAp*aQ@lPVE6X!w^q5OS-xn3c55-)R!yml zX7N852H!z@(k*sx7h-P?sKV6;A2LE&@4_)&09uQi?(!q7qa}5?!bFf?gUm`Gm|XCS zaIHMC%ORpguM@lg5VNa}dGpg-F`lQIO)+{%Vt9)D8c*;3{PA+#V=skIAjQJC#!GSk zD04iS{JjvwXee4o=7vm1z@)GY3+lwTbwF3S7k-hxu&scqFGVm`bYGXwiElCo)ZrWyR5 zh%NF8qekQo4gu^z8N(|_62V9nnLz_ z+7NW!1v`&@EM-qw?dUKv6=%ZG`KIKOU3-QPSGi?ps=+?vXM-}EHjXZt6ha?IEni+| zEla?jfm*sWd~+t2ad6LdX6;tXUVC5k{ib#B*PE7r9&v&u`mlIsQ*b~hhHFk9)84Jz~l{J?-KND=2X|;_`4JlkN)#{JTbT?Hg3kJd)%3&DV-1%VWRIk7_KHO;$iP2( zrX}m7g^!?IWsXv_)wk!6EizZ@RN$GtrAwGhnM#b|`4G>XXUzY3@%VW*Q0|;<;rLa@ zF+WCz$=12~yz;8UbX=W(R&K{k=@vtQ;gtZ-?s(Cy@>L=GEylJj8;h5U^Mj_9GgPjP zeXPIb7%-B#En0akv7CL`swuj?D0gl>`NotoJ_eT3Cbt>+QH2;%vC=Lf2)};m zU)sfMsQw;n9^JscF`luA6d0!NZ2f*2_Sra7yr(_x+jy;2t#-@r5qXBlQ`S0iGXY@}*^~k&`p|R#rSUro>=-DhA)|KH_s--UPmT7DgTWwB%+* zoDLoQCYNSC3JTFE&t@5|Y(vwRHqCJBDGnl7?x2VVb>{lbk@i> zhP4}2eWe4zM9gudh*Xb{<7y6Zws4ytV`Cf$@-M7gPdAH$NQo)Wsw6S4`FYnp5_w&D zjZnhtf!XrXsXe0QHu$x&B)%YlvER3Dkb_2gPUKGfY|)zO)3CU#%(Ux>)uYSFztn(3 z;91L&{GTEj4|_^giO)9;$10qkac<)nOkmFB(NBjcAE70%S|I_Q&QHQeU4=1T1g~N;403o{H1&Q?j>45 zKN_8kK`~pps5_A7R=9K}J)(8+3stM_bXaJk$MWdCBq3{&_H~X`Trs}4>=}(y!5O4O zql;CxKY1Y@Hj3+<4{T^s2p(BJo+_nB_ybp%1vUPCqW{m8re2@5`^gBeQRU~ueU@`@ zOlH#2u5nAD>JV|h_lud9#9o7jvh%*&4oHM~zgG<5C6A!uam9v)xB*!`%C@XT0Hwyt z=HxL+H-G01)9$71x(j&0CA3cKXk$Pd=>-h?7MvNLy)JEzHG(C3;3^PqO)iMV<0Ntgf*vCP*?M;L+l6 z0)&~oadb%(+G7Ytc1Qi02?`|o1xEOpKtfWvbvXL_4`2*?K@ZpH@bWM(-_AS+l=u^P zJsew!SfsJa*;q88Z-SLVLt;Zq_ET=mO@XfHiL#<`ukMU&jcPw zE{vAJdM7>Kew6je;L6ws4d(y6Gt{dde<_03&X@RyGfBH8O2Y~B zhcRD+*|c)oQD*uTZ@y9QTV`AN?yTI^{O0Fvl#b;F2*J6Lq5SdACoAb^Q2UxqZ^mEP zg6M#&?twvjmxl?xNM#$_{?a;(`N_5iS1E!ydBl~eJ4N32Ha%UG$Y+O3yF zFH-hjVt6%IqpdDSuHjE(V9CX!JUap?s`?ZO26MC`k48;?1cTHI%^aoqT+0osb5eVhKx1EB2w@q*me8_|cds~}!PTLCwV&j1C-8p;hz|~;5q1!hr=|$Dw*&Os z7acow=EK(kLi418s@}BLM|BNr?iUs@_~B;!yLI9tMw@(x#9BJ~TDrYH;yRIE;j?)T zbxPtRvm)(+2|S%2!dmL6)R)e-Z*O(+4lZqf@1Oz~*)m9L&r;)k@Xxfg*#lutCBfG< z(=xu++UQPcA+n17Jt;d+e$U8OP*{k$c+FH4RSUD4Z(qh)TZr-p0XQ!18M0ugY} z0^NAQEEd>qahsb`6_~R&763-{5F>J{%i zHEG??d?|(XTFs30&k8bI@!wh;wj?j`pLVgiHjt!a3XNiK8_XYw(gzN{`}Dne>(%^# zMNoMu|9{@Gf9THhH;$(l{JDA>=6=(ZS0qKP_d!%^%4^dts_SXO?bacld^T973t%TY zQ?ai#I5zALIn328RC^34t*Q8F0Iw?*CyUN(YO>D^+myFujIc$wA*GfPhP5NQs%hLh zTzV+zIoEdp+CZPFmqneQz5}db{n@;KewwEWU%)~i8Yq2wfBSRj<%M}4J6->k@BW%t z6DEka=LPE@Vc*!nr}W?o%PL$YJbC_8eW9Od-$oIAA5TDpmyV#qyI`hvvhw%bWcOj7 z*Kd16y0XpRxj2gYACf)W99CXXwZ-{r<*MCrik;Ch6d-GNL(}#?9$d^7lf3PIHxT{h zJK=X*6j7XK7wKOreG8}w`{=I7g-5?48SYzE$SD)q zZwuEu4Fnmcpa4C}4GQb(Sd?R4T&7+(m+V=reb^nivhXOS(%}L4uZO8j)>F?}6uX-I#>#xc~}_nWV+)+h@NR#PP@R43Xc{cL?9=6VIGoAZm< zaJfxP^$0;aB3&H!sUSSggx}&o;+`HNwalkc=sk;U9kOcP0vI=~8o8z>HsVbA;{5n$ zM2kt#UqLzCNctUdZyjF6-k;2ISod|*%bB8g*V!WY%Q+%Zj(ZyltTpNm+JC^V|5Ill zC~*C}HS0mVDgzsm#?#sVi|VK=d#}BTN$YUuGso)^LoODnC-9|7?`GS?NEAFu z4o~1~IYJz$3t6|bs!Oki>Vm_yNy0)B`U6JndyCyywx}blDj{pfWSTO_RTME^$ zYEmVP(%ZP6)6Gpqj&UGeqT~H+QoBD4fZP1P20)Wzt1Z+RuyVwLe0F}{t?!*T7d*oF zE;1f$MfDevhj}KgJEpxTATzb|(e3txzdHWq|5_9~`n)xSt(-ukHr`s`&1k3~H_d8oJ6gUAFIr&?RQ zn30C*9i4pb#+4d(L{{e%kRG%Wq$)3;_A8#zVWliR-Y-f)HN8Gl&u_B!PE6R91uwIx zhjK7{FkC?i{cXNwax+41r(_UXdS>a4#KGI0FM3Z}aAN&bdRloMrH&)a5J<=*XzrPuNT!{Qg3GF%<=C%Av%`1Z0b3B7Nz$s)N1oCAP+#Jxg)8+ z$lw|1jQfpB`5C8LW$a?kTsjsx>WtZ1?g^rs`x5HAzs6@5A^3<5J zU5QIB8(~@T8x7-36*FC_qDC@(OWzUk!@yJ@c)$<9c+=*IuMJSER6Q|Rxk`&HkQKVgUP~MBx&j=gm!U8rP0AiAn|I%eTr)RZi?I^= zmOmfqziuoLkY1Dcpt3l^P~*m}kMvS%Vov_j^ZKFG^Euir*ERDzpx%_DfCsMmH**=% z(;LiuyaaIo=(dPRXZmA^xGEi}sFmB&tll)|@e1w6nIsA;;DfLm|MMDa{uEeF*|eYD zB{EaJrl%E95V<51fspee@})_6nwW4?(5_`xzRL(`rp$9+^nv?Yb}UF}wpz8*8Hy3)P}G}|aat+V9{9ta{Hf?N1RAC5Pr#dmq& zZPf`mh=B9nGS+IgNaC2w4x6eRfAwR1;r9i19dw~QBv97dF4p$iK%_2jvSKY7=|;y5 zXB}u6A?@ke4<_L%yUxF0J@|L$BYRpH=snp7TyJkqm{QQNuZF;>NvH8T-gh5j1*kRC zANfaU7wG@srrk{CH7O{Xq|ydK)d`?ij2=ZKtqb2YF`}kvmj4YSdnZp(O(h{A^wn>K zM#^vqBYv`!--Js5)TlBx^fF1JMht$@rZYXROV;$E>uiU(^mx%nwG1(Xv4=(@Xxp89 z6Ovtx9$w%L!SL>JM*fd|CLT}YV-if%V9B;}Sex#zh=IB}8UKS2Uc$hK2Z1XLb;A#f z@g5@37gWQQZ^l$w6MVIiSbB`j??oo8EL^u2h zpyG2BNqkpk3`G$4bChqnDqPw)`#wqQ|zv9%m=rqTia3;(4hvIzoe0Nqc1Dh?#7K=w`MIUuGq={+^S|tLIo90>Ri2PKZj|-;FquYiOgB- z?hD#w6YCPuoLbgdi>1oz8;n;lr^6?&05+=O^!KGd`ik=vf1w9nfmmJY;G(9FOzdD9 z6C&gcFQ$M~9lb=5Ou-4a3S5$*-j?&aC7_%4#6rSr!YHc&)xLvKxOBd2-<-L@&pvZ6 z*m-5bES^=&K6!oW_q+A%mBM$aF&2KHr_BWe_dIu^|hd>&Bb*Kcd%Pb5NuB(s{ScEk~_euY`t)*Bt z#dfejT&}o`1V4_7gm+DOvdx zA^5In#c7}LYq|o3bbdj|^nritAeB^+MmJd;wP)Z8P;8VB8Kd2X4I3vlI`R>FHy)@l zhPe?h-=da%9kt4IvfqCC*JV@1U)1!eg~KTI42qFbyjUH+$zJ|q_+F}+NMX$|gi*V} zj;Ba?AMbH~1z7x#2-{&fihZ+U-&{t{h*-77f5wlbAqH)4>;av}czmAjDA=v4 zyfoA@Bzce7^LOD1`U{|G%lHEwx1B*(!ZLfp$&vV{j+8$iN8ovVaVe1i{20bFL+r;Z zU*&axT(XAKc8cTU+BOF4aSF>ahrCwN?8;oKDI)QD7inX$N4rMxeEXVL?1|xq{qi(g z57aAEb=`K?tl(NDfxJAPkTPt3@8FHqtH9w}smrHpyJ2#EbRlm(LbYgO;Z< zOTA)E`7wV?5SvQWD*(!|EqL-EEj#uiw(#wYm7s{~V%Ux>c*5@b-@ZHe81OTW3UmL8 zO(+-gHVKncnBC2fH20!JwGhk!3)qH>>H6sYYRtn;w`l@nV@6Io6=+V%V1GYpnsja)fS? z_P|PJq)mF~JrWLqqjF7wEPm}=RUNrC$ElTD>cRUkk}^W5_0=2ZT} z)=zM5TK8FQv-py9J90p#^@{-q+e>u(aV9L=NGQJK_NF|cHp@k4Dm>F?d+$}UxV`!K z^NON?YM)vjqt!U5sbVxoeE#Ze2@5J{q*dGq)t1`DBpcS5B=&2Wg>!M+tU`^o=JQ3{ z|H5;poH@mm4P`mM+zNj?`E+F2C-Q;%d=^0-Hn1=d&$V2mrFwRQWKoFVIPU;rRi6pe zVC1~)XXL|a+e+167>b42-}s=zV>*^Rvz>K3-id69zaqp0Z`rUccrHfz%HEN>r6BX> zwXC`aoLlRVkc=8WYWhUaZl#2EERcoQpzkE39EHQ7#}Tdix5fxTc_@%DeM@xy^&D^bS=90;{*x(No7?sQ+5HH7w@eFaiSb~&%33rH;a+%K8F4R!2RKB%SRQPl_3TECm_Mf|<(=DoA>RlAUq?vYZJ;jYsf&v{=slbNvBC7}i5T+yWHrN!{# z7fwY$Q^c!k+NIVLAQVc5n`~| z(Z8!(^+vG)bG6F1c5P;to~wj+>uh>ap#j?+QNb(yV@Qi~Jm#@iS)^QoMtL5QU;pO{ zQtz>+rvrfvt|>95)(c@d)^Ro}$B8;<)pjn9mp8oKT7zJ=6@p)MP^WBf-b>h)ox4{! z=m#CjRdU<(?jP_rg`0aPJ@PNI#aPo@)2-%*(|Ilr{a&#uaAjE79nCgS9`$haBA8ssgmnmRg38mA~#V&ZvS=8rE<=U1IA|0IylAB|s@-`N%4-A^U z>9lM*W0){khNAo4R5BkrOrOAelVr~l191!wJ}^1iRR_KetPoo!mG+3-c`SA_|=0E2|R~SK?0lOuySEM zZw?HDV_sTy{s!I^rPb&ho(r9HD*DQB`VkC3#LM6u3|eAR$9439{Rzrp;=!e#N4^qvpd{?Im64-672@=yLntxV;p%CLLM&aX*D^IF*z zgj*@rFZ-sl#sIx01VLfw-=uLTQn7@(vep#-1qu(wKWp3%lMwc*I0Vg6ttm6(Vm$aP zBRX4rZjPCW3i}>IJ9OO1^pqHxRJi%$S!i2?fom!_emMF-&^c(zxo%C>@I^DAQ?Vfp zt*|k;G)E(=6!wW_ZtE4&?$-RsE)0UkoGA|%80|QGg){NZ5jt-jR#yKP-q+r%h>$2f;~&^mE$ zWV46Pv2*6r;ZALnr{P-!_}l6IAl|Pq(6Fnsk=kyI?I+=xXJAcSLnZ4fr>cCnIN)hayX)y#x%tGX* za8Th3P@TlWxw^*?D!&-KZqu2}8yxCquJ1^i-HhH$vY}oNN*nsB?6jNV@ioL9ZoH}r zUpLVq<+e2~)t+lh5+*{eOn75WC6S_aZ|ij*o&fTSb0`a_ zZ=r@M^#qwTP->sGT(dk{egB;)wrB`7jD`M%)QG4HJvIjU!* zQsf!pgoFAvax&#(Z|zHU<(ONoSoRF+Z5hWIKuKoXuD^8VynLrWW(k6sLwg!(zkxJ(R;i`@b~`x{=q2fEBK#S_HMW>+Wp#_ z!jf!kQOk4HuEZJS??@+tWd_mq?ZASl{mdZ^Ga%Q;rOGT@58sef2?4+yR%(|k|(OLc$G4w6Q9i%X{mp&(06W%o91=QZUsx1@sioI-+P8~ViH@}C-jZkG8S<_u}dozx+dR<#GMavbq;%Y6g2#;p3 z;W*eOT4=&srwnKo8Fps!8lnZvp6;`?p@$)C{neE^DVJr`QXiou)7cm$_JYNp!BiVE zf1D5D*Uqk*e8&ccz1vAxSq5>9U!PJYN@Yo*g>qi*DgCw@!o~8? z$T5=$+rphAZfm+LwMYKGK(Dz!+(XB|`_yO4$4^Rse=ARo8R!9KqBX6?%G6QCK%x9< z0Vn2PjJkEPx0QhL%rU?_UMa?oBdTxu&sc?lYS1=^onlF&y4R@L!V^BX6j&AkWdHh7 zpUi)TwSHE}37np+bQoa;c&dl*-w%-EW}E;aq^B3Q%Qj0(MlLj$U=JNz!heDv zeWLSZUm*1dROWs;&Hn4dB7=<2I^6L_DVsaa@5FYsv6R!iGh`dzN~ek04kOJ9FGXcd zM{*}AoL1%o_Tl<3BbVK^TgvD|33F4=jl^%q_L6uukn*^X-kSM{8wgPZhe)(_O`>C% zkpERKQao_wMJpu_2=TFZdH*G6g-)raAY*|WZDDUpb!8Xm#HW#f?K#a{n*{IY+38NH zd&Huka<1N!)T12Fd(>NPbp7ww?&aaP`#U|zLp({sR=wn4-dkq>D+^#d!89NSbQCZ2 z!c-M9se<+VF`4*HT>zx04Wdc-%XLrqLVsMUBW@hO~Il{3(||0=owB(`mn?pTE|tx^xh6SpH<%d zToecrBqak(sAlipVi?ON9iymizuexY=#}REVj3M;rORkh#6VAGecfmCP6$zKztHHd%(4NDN`B)Wnl)XVh<&fZ-Ug+)Nv^>dFvBN{KSi=$fyvc{0gm zafn~@+nQA90EDU@i=c{Ofb{1Eb+pya_?|$anQSurBkO9A%i8aM%Ug{x!X|2A^hkt> zQ;MjQ)egN8A!iauy{4OcSkej&WCGeFl7FvLsiV)3zn5|8jLZoOycp>R2EBWO#xf1jveQzaeQ{(q$}i zBQjRrcq{=_C;Rn}Q({M8Di2RSc){C%#7ye)I7tV_76;M&OPWf&mW=w*LqWk-f3tf} z6*v6?qT0bF*7Sbu&5e2NOi9cGo?VIbuDkPnd6n&R%r}IO1t0)dl6^DZ_VYr9ZH)1J zKm!&=M6s#G>R)G9IHPXeE$XUY^BQrJ?_dY_CVk_6)X5ZUVzGhj z2+EQ3F`DEwZ>1}9bw04-((?dd%8LHl@M13cLkEcPe6u=H2SHnGiTqehCDhF@cpIk7 z5d59pS3B!>U7r)`Ix-r0=0tA?u5=ioVmgeN35d?RE+_lS+7toD2$-fqaY`*6mDJTS zhuk&s@rT@CWJ8OL1WbGZ!~QG17lcXkO{Hz&M~P`!J@n_7!SP)BLSU;x3)f4M9a;FN z+b471&2{L~G0b}1EtypNS0^FdzAoZS!a;j0S^(2E{m2=!-jKg)7znsi#GFm0{z^(|2A874-yl>U& z*3@;N)kKqEsn~&iM^lOTVLzzp>&k@~iJM zVPRxi`PYpk`1o6(x#E#M+=$6pG=|oHNm!qW`Cp^7?Ym2Kz5ZRP=3tJnnaWk|MdHK2 zrD`zs(qB@kKl`~IVCN0!gfo{kjlXJINq8>R{Cg7@YquSJT*km9)YrqQ&e z+snuR^8xoyLQ8gQI} zIMkO)HI&dvg#V+p;_;L33I~hg!0J_CFAL6gt}Ed)N8fJ)34zmlq_&0>*ea zbvHzbT!QH0x(VN7w-M~Er^#(y7o{3N^Lym|{jxIKtu=ZXOL)MkB_p%kZXzpPJ5gT^ zBa%S~a@zq_cv?&&&($%VanfL6v3*lY2#Dov{-v7Yu2Tam7QuX%A1sX*pp^HVJTxDF z!?M(y+nM-4%7dl!R$IDgIQ0U)0XQbDI)if7I&zs%_x?a3dzXj~RHv$*vihA6wEc4n z{a*@{r<%v&6QH=;o|lr-EjED)`ViCXgua-vq$DJN&xP!ND)SuoH%bP3V0Z%00Efl? zPtB5VTX@}sPL`pO70*BFN27f*qX1dG)QuHyi1iQNm>X^zQKc7p^Fyqi5>D?xB;e4rh(H6NIBR&myFfmKgLBEEKIHYDz|hGd%RaBD$8E(UpEJwC3t5` za6B0-7Ot#*-!kPZWb1799j1T}qcHr#rz1*v7pzowwA#LM(jHm)BMrA@{McX2F*9NS z(d-Gtm)1YVb2~y=3c>+{RC;*bp$-p=tgc| zj?{xQTvuf{OpH<^8wn^h)CNJv0&AHy7Q1=sbK3sKdRJ4nW?YkS-Vjl$YsfK!IFBx* zCyFDEl+t~d42-k@%M)`lICKP6%niDxKhtYOI$GJOuFp7+?kEb)man1-9inql^aa^`LX5%*) zgp?>d>-MGX%e~Ap3Y;SY1>m% zJ&pUVYXT}Mjs+H`Q$*XV&R;%~rJyVV#l5f#$yc5;Y{4{1?r$p9b{))+5JpG75 z47xJ^DJK4vpj?#hnKM7+{GKR1e=3K57Io!`Bo!O=ndjGDeZEO0Pc4VWV3sDlGHgCU z#H_bHmOa34G45wFmgH@hzT3IDc<+dq)Tlh2fjtsjQ;PG^1pU}=(%e>l=O}9)6`#e? zk1K;NQqYzZTVWTJtD54U;8bT|_c(W~^m54H5%X}qqA)4tf}4Qf$7E49n@@~ozRM=` zQd>9dG523m$Jp`Ly&jA&@!W9lQ6|MN@1kw(ebdJpa2FlC*np7irOM4eA`;e8s!=MV zJ^o|4+(y)v99C=7B;7J>0IwIHjlX>`FQ{BTCH1p!8AaRohfuZ z^a=}SUv7Y1I$RS2tl-5)$hS$$UDT#$>ocU?l|Ua6^v)XwyesR~G2|Dp5***(v%H2< zqPZDapz00BNx+9dFdoVu4c*fUcJE&`!FD2?jsn%1dM_CHk3XarwiM{_9eR(b3;%uZ zstJvXZEyRo!4w}b2A^v@5_S0fa~AQ*Vshwgy==Qp_*Au5+46#)Q4L3syfaC3YI1I+on^UUh*frMu6Mmf$Rf>HDZhwA&iNN?P8q(;e@y^#6bYkc-? zXqV~AyA2KDmPdRDpy0T?=v=f+I{e6>tNdwk__5>G&z2a?Vjt*cZX7AseFJ&mJR{T+ zB~xo#$A4qfL0a=g~WS{AiDe;NN`qHv+Rhbifryzdu=l^|PW zbIz^A-9-u?_wK_6y}%VyUSl?b@7AhDNooq%Ifj2m|5jx^Xl3z_#@S8RPF<&sSJyqN zs23b2!46r{3Utc@a?%`svIMTkQYw@TQDPaFy*B11*9_D2YL)_wwcl<+Zqfy=ELnRV z?RRxVt91lTRBf=VeD)Lb474ayNfprDpgy0@gE;k79BTZ)pxoLS|Fgj#RYgt>ZwGFq z7BD2Fxxhpxe!?g9*?3HMgMrsHSnCqirD^e?yfh*6s8`?C!d<541oM)Gb5-LDbhr%q zAO!o4WU*-f1zEOV9tLF_QuF=8-mF0YhR_rHMtC%n5e8b49u6g^3Nc1Drz$Up9J%2f zcBd)$Y@4oRPIZ~pD#G1Ohjs9pmtJZ$+3ZsyIc#{ms~rOLX1ir zvMOw64i<7iU`&hICfzAhUQW&Im4|YpdOt?dn%0@%J~%(Acj=lk&d>%b~t2j1G)G*akFmEgU0&9ncfk^#FxndO++WZ8l0E&H!W zP#F=Cek*!)VQ}Sr;8VP1SVT{hqe!USViqah77B)Vd4fC=*c^L7%sUTTWdm&`7 zT$?mp&=FsYsciBQ_|;|i`~b6zbhaA94Oo2{0HDQYCDebUAc38^(Jf!M{`RZau*I+ zK5kCj*Xw%H_I2}nXvP2u(lNI8n)3#JGXS&I8ufX~0w2+}PzfE-Zk5(*eLx0)=Z*hg zr^o*Q!U9jw{$0pQKEd&~^WgD`RJKqSsCf`b9*g}bc>UIC!|#2*kG`niJ0bNFY#an%e7Wxy9s_A>{rsIlSBIGvi@e7x&6Pb_waii;skc_mydD^=Hgd~u5sLT|eT z@_Z;4Lo!z99q6SktNK>mp1O@+$c?K}=s%giM~|?%`quc4sGsuLD|5?<4vQFYi-9MP z@cs@!vk?&A)V=NPj1;m=;o%-lRP{RcEf?`oOr_Szu#iM)z0)(pzu$5(lhMAyYWn&z zyUsIZ`ZSWxgP*&wJ4!yAg4iQ}e3fn)+(3(`ZTAx0^t?@*fd?4}X3aP45oQTygV}`cLq*{KiYEJ?7VAKss|im=0vE5w)Op`|9^tIw2ERdf3XJg1JqaOh)TbtI z^bSlL|8vH!r+ptTCr%F&TT{|K6FQlui^azUI@UQfx-=bW4aTgRco;_i!(-k4 zl-UDAG1QMyR9d*=_gp9pPu|A6N6m2im#Vs8AO-y~YXd}AcKwd?40_5Ai&q#QJ=7^o zqLJw-@AAh?+n1a!d>Kjq;0A}9ha~LOWSTj(Jx^m%E6T0lxTvrcY5~xdOudZ<17Nac7qv#LGnt8V#~LGIvF;1y_cfkIguu- zOJwPcj&#nM41Q8VI@oz2TZ^)DW#q^v)$+EsRL(3ByA?Dfd05SI`Z2xK56Z_Awtcyu zJyR%MMLUNTY?=$s>-!_B>763D*$U(yS%{|(X$1hG-a0VIsB7Iu3eBExqGTdUv& zA~G^o5S%vuTq#k1pRo=QQwKS54nDt(j)vF(RxS+AiZh`1d#mkf`nZ;?MJRTY_vXjS zBeiOeClk$Er3^7vlpy|Wb?nJp9j4PYjF)!JU+mP`p~k#9$~PC}8pr(&NL~shTXg=l z`smF$@GwRK7lKxsx1c=))^1MZ028U$W|6#_uXlr zz_JyyLOo^eS%oddyY%p0-wwKpUD#dX+dbuG~UU5 zQYLC&!*-)75{V*_e;s3utC~5gg}dBWO<0OL3!R#=c#lNp${F8|oR^h+x!&It`=207 z2KO@#4!f|q9g9OK9nD+?vt+vSfh&l*04|*Z)^ckV(Y@Nq1k)VnzF9S5H~g8D?I7x- z7GU_DG71AX9GpE(mez0x$b*hC=uFw)Q|*grP7uDFo&hXT{Y8#4=bJWIWz=e5`>1}zI(q(lcjW*B=O=jO zZsO(Gek_gN;Y#iJ+4I$s9c_Au)_o>&fELQTnWoYL9sxS_LE za|Ug6Fywc^-LFec9EtdEQ)6cxByofLK4OP-1+|MJF=d^kf%23JAZgx%KcRM2Ju5gT zM8X|B1Th4-6n|553vuY43ccj$F7zT8fh;%Gxoeg%>BR>iMq{QQ-lygcknDTkIJ_9n zFWcc@go*eLtirDd`eLYWxu6q;*jtJ9?sjlUFYno$>@z5z$@uttWAYk6&8aSn07AZo^k@FuVi*#+Ye)VW3d6_yx4jCLd9LfMEv?*lT$U4KE_1 z;eFPUIixPKsgZw7clWD2Y4z))^cD7lFC<~zop5@+C%a2Q18H{z72mplLg!;p5ZD#F z!GgO}zx*9sva1yD2(DQqNxP%_er%)&Y-}$#Dkm7%@n$(^ofY~LV48=fsO)&|tI-So zqCokxLN`7GPT;xk5N`m~HTjzVD=Z?K-)Zzft(OVN6h_*f|6d^VpDE{WB8Wd~IymKX z{8Y7ngwLCvR9`kb+PXJ+oCZ>l$xXfXk5=v8wIQT^Wq-cT;D*Bk)t%>d+9lS>bTL2w z_v&={k}K0+`w7X_+sHwZ)IiS#!K9>tfe3W`>7VkfH5gW5d=iWow&qs5SrX9i(Bzav z%d%p|Vte2*+M(|_av|*J&>2@L5%z!13Kp39YWgzciXfOVLFLOT581|fAv||~{5Ynw z;fBSb^D5<58Qi?XejDRmMtRLB>q2E>yMr zLWsDP;Pu+eA)eyGE)8clt-8O{Kv>%4I~&6ja{O`%_=5cp>CXqO&+>Zj#_6frR1Xq5 zVx$tM4)KRePKf53Y&Y&o14@~(vj%g!n(|uhMkgmil zA!~n*N&?+TFtw+XZqI;Jf!xX#7wf2Ac;WD%`@WeOB7mjZmqY)o)4W&>#Q5@?ZgSO~ zA|l^wf==4pp8H&BvTa=fX?U7IRbMi`5-wtktKYb4@2?yD)O^QySGRW8a^lNfHFDkR zUSUsNV?DvCqUA=D;%0)0$Vi&lloq|DLw28_TRq*nY)o%XhG56~%iqyfvdx+7G-F%W z8E>+c=Y3qA^Xz=8;{-2o!nA9+2M;*Gt3XmiB)9FSKfKfbHP_H6|HVp9Z1eqpo{_h= zs&dtj{wxJ`hfOjuai~x|VD})rgfZshA$=c*;s95i>bf!ef!X{R3UOu0fr7PR%$X%YEF#qIn{4 z4ZFAL5OYw*#DD>xXLISbthBvbw6l^Ez3`R8kYke>L61cLGKmFCXk#hgGo-wb_i66pN>7ExoO0*}vZE;2K zZC94N@1CA17OxH9^WHbUoNc4*fCyv0J+P5R3pzw!By`MyM4-+2TZLat4wC?FS>-3W zFPzY{t3o&eV~Txp$-V(xNmRmRBhi*pf-U;II9!%ATdgDw=+Ic$*unNo{Y298-?FF! zPN%v}lx=xeg0I?a=Wf497bLT=l0|=^>QYGL-+v1*-OMD#a~>A8XGpb?i3AsFJQ5T5 z>~Mc88l-X!(^lef<8h*Pb>=?BCi#Z)x$0?ecmu#b4u@>J_lb+npM1)l5J2o3pT(U{ zafrAhu1cZUU_>AIZt)2Hf9$fii#jzdJRQXIw(~@iqfP8flxw; z6h)Boj8sTBnb{q$ec8|S~(SnN51IxKHNT-x9faKa0Ynp)nwgUg)57Rz9U52y& z<1(b{X%f?t(QDz#&`gImKueMDGXs4JodHoDueu7?GwHJLX}Fns5T*SaYx!sJLfq>| zfU2;jm(+_cZqafZS(Np~$^8}Zru(diI{F*?24|TpezgBwcovK&jD$M9`PEIp%b zeAQnPc(gyu-CyN)8_+Ma{fP$Zt3J+Ij*gMnSJasXQPjkXo9s7CDSK#`W&5_f&QygR zN&^Yi%4Y?|2?~-GS@nI(3f)58-YR<*@JeDPx&miT0NL1D&hM$^0OadO;zqCGlLxAT(#;Q76|WQ&A!5%gVWZyzWsrz>+)n}xsvdQmBs$4%sZ9~dF2|1 zFI3!s7lN!Iq%76%p(~=hnbF<@tR|)cfct!mK;!cpm9y(7f9A)o%D!Kv&8N&0iKF$D z=f#Yz5e(xE)a`@#%gG$RRijlpthL#bK^BG`3BCZE+=F=?%#}CnD772UV^Crglg6#5 zE!B8Z0+Hhv1H$74tMGrQ_-J)b*X7Fpye^+LSQm?cd&job7rPU=Pzl26ZwD#@fr%4F_4rtiyI24ootSgWg)B=QlrP@7^`Zc_=_ARlm0_93EHMb_bx8v~%rEa6$ z5j+VDn*0}JvP1t^lB2+qRPzPh0XIMHvIHt5K9wN|Fm+)y7=gU2*AY7;f^2yzN(v^2 zuzuo;;cmb*La>(0G< zeea_z#%a2H_`Ekct`nIu7LX8iSj>{xj~|=udHs10rOltvp7S;!vW&yQh?0ehpttie zc!bG}JI5wi@5u=zvk5GI`%SAUs4nriyZq)8xmgNYW^1XQX6=@;aDKYuf#v?q2bBpX zQY)yI_tT2(R)BZ{hsH4U$wLG!Wvx&ALviPU3AWC&th-Sy;A8~-Z1dSb%kJ@mT;GrF zO!Df0hEYC#(8CCAhH9ZUO|aAZ%o^)<=w2fNwlJ$cW~$6_r{2P7IMrP$;svHg|8W}W zJtYZzm-Xp4x~OOY1F`cPQyA1t6W1@5>yV?iOpCl&!^o@c>8&_HL3gC$g5u@%1r|R* zHm~Ki zRa`w`S{Tk+0kkmn;{b%1(ccJK1!%WLw5@@>t==W+xvdN##)qmcp_k8L>K`#BoP9dk zNd-IpLGD%X*fF7eVnr52?9}NaG@CJ;}5KNn@$7O zb$H|J(#bKewr?Go{c-TG00)0gV_K#jYQV1Qp*WoaN?-a_n{#=g5OmwwtGJsd*Q-%E zG%PjE(E#I7Lc#)BcRd|Ioj$>aOjem}FD@tNU^>RV(%}twUdSn&Dp6{2`CO^7A>2Jx z#Jf7qHBgL;z;LWt&Y@lNqrPSX}=nE2Dq!J8?8%|m zz--ALvvw^|y=3bbCq+x!Xtt+HCs-nr^rG)-aTpAI`2|dh#sG}u!s@F>>_GYTp^6Xm z%4#!6>vV>$H=a6M-jm4uq!z2jADHf_WxcT74(QI_MYBg#H~fY{W@FrRo&b7_eI|_u z?up6Tv}M?#0={1%AdXXg_UG7Rbzty6NGS*jR1Nb}nTlol)UwYTtQ(Z}4pI#Ja~>Ql zvWl!(uk2w?A_>%taIRZfheLpAG*{N7k2wSTl}ndyLBGGJsWPo`jzSaC$V^vOPTXg$ zohX%;ZFG|L`LeCM7^6kiu30>GH^FZw3U4+*je!Wy;W_tNn#qK%p1*v3((0RPNA+2rrJ6% zVS6gA6k#;wVjfXqeDD^Vi@4MbY)hcGQNyfjKEw*{+hvtuCsK~4d!M8+4}S0X`YL%8 zev)zV#!I;~c{T}E+F$Yy#la8JUlRZvfA4bW%s&hu;mK&2L3f7-*o6&+ts+GV#Z$Q+!R?41?@a#{sR{qS8w*viOtA^NV59cU{>wxv$z*Ps#2X09{$I$HoN{= z=DDo_;RWGF@jnlD0THO>%fvEzbyDv_-4b5mJS`Wg5}Ige-hp?SJ5fN>%lf=j4;Dx~ ztPOi}y)K`VbK6%=6eH>+IL2}BWXJNFcAk?TEPp9xw=dV8BpPCSpeI)ve~k|#-sP%3 zaMn1D?4uEOnXTw8cN_N;CD>`9?GheeS$#ISE{(>e$f!6s=R6#OGZM8Ne|C80KAgl( z34<*Coi;CjvLcH-`C=X80hvhU_DrNRh6^AFM$m066YlBG#f@6sudnxujmu8H-%8j^ z_Y)2@GzS4l%LfZaU`woKSeuv@TAS?-0w&{k`m3FO0#K?fEz!#%H*HI40DW!dWiLiN zh9h)CB2pxhWV#XsbRXwu1!u>#msd_vV0L!)pN^6SiT~9#@JcQToNm?n#B0&2v0h~t zifo+=lL@4tjY%wD@N`r6HCy`CkGWE8)W|adci+v&5f5qGE(kQNJ~h<3F!*I7n%8V2 zQ0KK4fM4TVbMY0CJ4|9ygW_sE7%ziT)2$Kfk;sVb<7DtBQLAjo%4q96TVNbUAnu2q zc>sJqnAJd~NDQYQQjf`vb6ei_w@nkc4ATx2Tj)F-D=A2fd0b$Vdw_oIaN)QDn%bEN zmaBP*-G7k_9{EYVtJpA zi=eK5_!C=5600=qdYFD!%{s3h(Cz`09tSriB-#tkCI@LUkF~d2w4$?RMz-1MjsX?; zO`l(dwA_!%=_hzMwiaOgzK67-x%Bul2;I>@`xSM*2K#ncjs-Atf))moib3cOSHJ@a zuxi{_$Ci;Hg@7WM(R9=V0CB0M8mvfGW?m`O^H#{!6hffXc|4(hR}9k#v}_Y7JE zxs5OBH$ERFaSy1Mcj zEhDksLyoN{ytAcqW)VHysuPtvsDzYbp#9Q7lBc^PAkuM#R=w9TTyVh79N zS9zXxRGcah)Xfao3R{l6@6C0T;5e;=kS7E*hgYedaXJ}cxLC*d9SzI9_83l&RaJRh zE6|VHwXCNeNs+Ot3%y|_JA1gE893JkUD(UR3G$&?sTsM`yN^4|?5j^j7Y4pe%RU<{ z2?&;M{MxV33NiD(qc7~dVpy4{!2nK?161xz>kogOE}G;XVLf|SEgOHa6_XrQddr#p zZ!S9>dR#gsSvpZQG(p46JqpL9vW&c2*L{`;(nmO!wixzbnDwICGBVgolP&;a*%JrQ z6K}MmGuR=}5;#DOD6wDaF=FHcT>`-Is$AG9Ye zNcs0JR{m$`BwS_yrvmp>!)Z|cwfMw}{#;E!B9O-PouB zESX!29;g2yeQ>$>JHUDgJHsse<~Je^y5J^Qgts+{yKdih+F|*)l95J_jF-wWzU+=4 z<-u1v%tF>nXx^;mScCkwy2RNZgoXNj`56<>fg1v}l~ytehwvVez?AXGpg4Mioz8Sv z+HL9UDGTQd=4^NVnopRUXCzE)o6B=EWVSk7xquP%r-9)&o0O-~H z&M=no?5}>CZlLS2)N^ywg6SOC*6CLVt5x3|7a(M>sis|F&-}h1@OhDLiDfKgC$VSP z>rl=(%Mry&zP&e4e{f;0y!L`mpHF#i*j%8y<6U6JL(oouj&vGfRHIzCw`u{ECBHD2 zz-qt`G-E`1lNSpul4%b|XGU1faF{xXTKviS=T7aji<3SJ&za?T0Rx^Tj6sK-*^u6) z(*9xa{v!%%77l{&)tE;ZY2U-`kC60j*Bo!uWNsk;#EzPr2cb}$VZeJvSuSj9Z||>HC_zRdQ8SD9q+uoK<{Rt-T=yo@u_xrYnyoBYNG;l7C!;t zPYQ8VX2%3HB7$Z`pz#xMP7Tb;J&(aRJV@=lAvJCMU;-bUf7lvguV2bdjU9Y#*lNZA zNF1oeyZ^oA5nCjggAFe#RyI{8S`5Ml5y&98EO&@X<{kS!7f9;_piP;UYv+e(joEVd z)|=Nne57Q7b8XqK*Wi>F;{Q(bw zh2EE@JEv`S4N4^kqD4%$0c^V3q|R;SZpbx(ZyII))lXk(%wv)(HNAsx9B!%bM96}& z4f{EAy#zJ3b5hp=8zUA4_B*zr61(4i!Mm~@ll7y;aDi`Tys+gX)hF_;S(hii)J{GZ zN4J=eFRhJjR0HDBomKf0?&{&yj)yc}IcrSrxFu!~({bl=OiCX}qow->gu!-|t3MAipRsWuuW+6hi>)ig{(7-p4!DAE;s@7E&!pp*W zIJWR1Pc@mP$N=Tkm*pkgFQawEV$!j=FVp_iVX zIFzQzEEjokx!)LB2W5)nmwq)BJmiw*v{bEmRDrcg%n3kNA8q6*4Hg=darj%47Je->>Y?y zet+MWxCHt=5xAjm-U^eR|4Jn#<+T}`0`llS=cJMhlxp+HA`-%uYWGb^b_t1*(#1T| z*zf}ReNsx=4nO*zeyTeF?@0aQ`*Y#}wRVGrT#{urdv&zp;7Orw96)Y$0nj)}znP%d zeoeP-MVnM62nQd>D7q|kPD9YT^54~`0HCeR3YqSL-L{oRQ3 zhtRD2URU@^x}>-Msf0=Epzy}e4>2vSdz*1QSVN{;5i-%5WtQE5zO9uA#bud;xhw8X z6||pGH@E!dANW-*q0lN2IRFYLat`cOG|EM=NEZ5e2emQpNVA17jzmkkT0_9r(MNU} zdK)hzel5G|qE|5YS|O}@C;SCxuR^^ZHZb1`CQwJ+@fK@O<*dfY2@t&J-@8cW5}a3+ z2d0{`RTOi~n*yTcV4pUKGkA%A8*rAO6!cH@>`c+C3}E&s?;3(0L>K$iw1 zQWtaA0@@zk>vtQZb@<_1o_6e3&Km*z3$tCG%ql=%`awfk%Jtt1;&<`D{>&eal{+4Q z2=+Ux+dS9&`Yh$fUw{Aa6q7&x&slDBY5Ur>ESmRPakt-QE2oJ(CI7YZeLA`N$9!E< zXWQj7rJ$EWSL#b_;N}iRO;rf8BQM&+BGmaD6M%!#5jQ9ESDq8w6xp^zN2RG(SAki9 zc5bWy;uGg+nKh(ii~R+FyIS=`PUk-I-ox_a8~X=DA|C$p1N!}`=Xm|m)0bQb<9BOa za8Pr|>aAT4N33qNX#bsP^@f6>>la!YpJY>m$$Wfiyc`9P|8)N)Y4+C-elP5g zt_CA@Q1QU@RqDxSz0pNgWXh;;Lt6WbKRJod)0+S6a@2Nm{sY6}U*D5NOPsWf_ zu|3%j&d`-M!@s@^`8QM-Z%8}?s2QjVj~rDHAdu$64rB!40Gc^as(m;UQu;>%?Y~p! zkUA!S91;BNQ1tB|TnPTNkYrZ|T;{LLSnqU7vpoSo3n}(Ev;65Q>A@m{nCnIQ zQ9@Q}aexopGn)?xAll!QgzE0%^+66Ij7c=|ED;0@;;M1Jy88F^(O+-w^uq5If!m&? zOIG~t+Vh8X&l~xfYx{pPW3(%C=+3dSmNNQ2 z{_BT(3S^MZB=8>S4}nM6&>pMXOFpN$a%qE+QXyhHT4uxZ{CO&>&3u-(|A*fphy!pe zdH~N#q5W%U`6=xW2>x`yPH?fnL;(_T!JI2Iq6`jF-n3M?>$<=jeM=E~ia^OQUVFUm zT@SBwyYYYYDvoXe<0^8xwpad-{`$Xth3{YhH#iK7fA{w(=wBYx-+%u9z5M?-AO3q` z{*wQt1@NC6;qPAR|6K?FcNhHsA_n*a=r3+abkxr$cokG@K$8mM&ErpxE;m3SE0YcG z9%*{YNGJasoXEbVau@PZBwAAAaeF*}@*Fb7fXZ6n-Y?f{F$3*7octHNul;?*`)?r> z*iM(jfK&Fvg0^ec@AG}+LQ1`-_7HxOEqT`|+EK}4#HK8PjZAfppTmkOK8orsQf+s8 z^vX}k*O&h@A^2<7^B*b18{lwEJN&(S{U36CwLP&mY=pc$s@oHX3nQ*Ga79J!%}qAs z%h@TeEnjXq8Kt-4j3kTBXLYNSQ;E($c|<%Fv<50-c_S2!k-$Kjm{t&oZAmnJ+aP>pEufGN42#5pPX_DDwTEKYG+kSzzut ztF<1O;R#anA(ri&e@{DSuY~!#C;0b2V>1HMjL&zo{(KM{##2X?TSCiuTxCv+(AkIE zq4+3|+w)FyCh*b8m?guXgP}7m4l{O+i7#aQ(tT%^8X9w3E*=xcjI4JI>{qh*ZJeKc zC3{Q8F#IgffY(T2tbRmQD7X6+UkkG4iG@kf4JAWk$reNYs;$<(`&POYD zT<1=3)=Uxpr#kq*KO6stz^U((RnwWmetP5bFRy;8bg$|>*>Rm6xT7b(Jt6I}oju)h z`RYyfCahL;M#fL&n5FIm)9kuqse7ix-Vl~8x|^486J&eTn#)dNtmBdx(MPQRZ=M0` z=`)};_#pfH8BD$XzJngN>g;%3=H6=bc_{V0G%ZO+V2twnq)RDhJO$*hJbB7Y2UBXe zG&MPqf9aAGoh-Y^Pl;yLU)(CKSlnH`GTrk_p_24}O;Z2&$Ch$FkhpLT@F2bZNI}_E z{kMh%+Vb&_k*t(e`J)*82but2Ptb4Qym|SH^;ufYSpJlXhVQ_W8Dt_^WBi~JFfetF z{$lz=jVmXepS9Ey#!~NE6kgn=ANgzQ`9FmK67@Io`)A;6Zj^ugrT-_YrNk*pK8DNm zWTgQZ{KJR&=AKW>5k>Ngbo5JKhIr3CBilanLE0mWLN)(}C;dfMyYV|xHyTkz4WAe$ zNq#Z@PuKg`QXtU=uKcE+*X>X713WneA2YYT#}`rSHxW-1*0Tbp0s{g=M;jZ!VT>Oh z$sk4*e)3GlCo^}QMBj~q_}W3s>B zsSn3}dz?e!$;zS~g{LUe!*f7=ovi6FRJ=*t^87n->vvj(*6$3MsFBB7NqTlkQ>fG{ z|5f75z~9j5H{N`$<&^*J>jx_p2JUG7*f*Xq6Z}t4tpjAV`z2@2@8zATG+zG@P2bo6 zi^Y54A=mGFA<;K)P74#7TJe8;a@&p0XAWc?Ks!=IU&CBZ(M(zJ<~IGs(u8C1cFgA6 z4sxfwWrqK?Ga}o_OA4m;E=*(fo?u}goSuS3IwD0rWQP-$la@3wW}GSccrI9GA3w<& znP+@$|E+&$H_^S<2j1SR5BL5ooTdPWYIjpcEj#XNd{BrXo9onWN*a0An_2mfmbX;q z%f-s?T3|!{M@$I6=bnJa%x>q;!+qvh+cS&fzUFaM_HA1|c^S1;@|wJtU_c%Dr|#4X z5Yl~Zk~1W}1Sn4TS}Il^H{5=ELACcb`P`0y-Ooa1?%yH&_dWdU&7C2c1PUd5@s0UE zWt!RiBjPW{L!^h*<|dSZL39rd)I#dRE^Jr8n`#n*BZ<@vQ-NEVw_1|Pxvv$o`I_Ny zjykcwt)Nd2tM$)s1nu{5KP9N^CGFl>c3uSY&w6q1#V+c{KKyP*@|wtNyp5wW1Qx7s z?e^M`>BnpQQ%DZm3;(beoOytP2WaR1d2`SGH4eqq(gh9~m_uur`S)~hsyi!x-Qauz zH4)M}?p@Jxvgn{t{;@n)48l!BwEbTt;rk2z z@0Fus#0Mi_GgGY2MVOLN9rmc63`!LTFaGbK`C9tBm)48qZ^|Jl!fvpxBu+7dwma8- z>ndHP&o+L-n8g`SL;MN4m6Kj!4X0ojrZs7B6dRF`3leh#F4$V$(Si~(XY2fo}K0u z7L2~Dhorr?|JPD|^JbW25!hqmS*A0ykcpt@w{i$YPdG${WkU8y#&sr>pOT)1o&Pys zzKUDG73T?p7bI4ys)6Yed`#0FmVpQj<4Km}&41WWz_%GE{r$V2ie~{Qlf40v^-I;C zv&)EEjwd*S!TF<6uBnDXdEY?xvhsZ!~Lwbmh{tN2u&O zjthF_&79{)45#ntnI)XmJQh2uGgQ6C!ws=V0Ggw@p;7o^z;JQ1B`-{7 zHUHZU{l4_&WO1;eJ*P&J?t_73mS?NvG^~;wAqZnGfRT_;S6QgiKJy57;cM{e+NN74 zAxPr~0JIfktdH0p#NZBd`dLKmI>&tj3#Lq-T4^!B4X43H;*M|My;F`%KQ^5&#Lx<- zdShQ;T^-0e$zLAw3WEnF2Z4#rIxDlP`EecXe?{Ye|JN`OZQxb6{yY=j&@WZ5D;_}b z;Gxmen9Tl=++p9MSGY->3)5!RpSl2vf<-u#FGy1~tpIat#WEY#vzkAd@7x;~;_(>` zK(&8|vQj)hflVCcsV9lGj+G`27O4h8AzuzB9Ykz<71j#BLmMoWPu_5&QX*EE<2*|?wqc*sEnMV9@|6Jke(IugxsS_*A9y&FBvMa| zC>yYkXr-vmZ=k>h5)yHq9`6CVfodKLX{e1?{Fg+_?Wx9=@6u+e?-A%dJvocM?znYt z7cOj7n92^B+2Uykih+vr9AzphSswq$ahd9QQR~Y?RShc**S-&X7O>aH*#?Tc$1lvu zfFgUj$s5wnMG*3wNO1f$_dNd}DDfW*fH?Wn8z<>{x^w(Jxt%v>RBT0R3wjTond@^n zfm$h(NSj!$Y2 z0u|>6e(AXEA6+tRAllSISD+5GQLnJZHg64HKUBe^LS?IW9NN=lz2zA?KQ+h?IgME2 zbkbxbfk(_T#aIGIk9_XN+)RILK)BRPYVARNtDk-f3k7v)=U5%<%j-~n(_(h_=S z^Ex(W^Fm*0$(TN`tAL?Q=r?`ttq&ulEw?#!iurbNy$V#J?h)dPx?)x&Jslg76IjnUwULWTFAivw@*-S_qO$yjSYik~2BSV&XtNl1tZ z$s?JRMZSb4Rlz9J#9g1OCktW^%8H*sf;#?w14k|a>6?~YLdumt;@*2aiJ22YHMca) zwsWZ12B&#B_96L?U;?f}e}~XCR)y6aM|m+p zx?}5Yhl=voqS5F0c37)>dp2NLddy@Nbjh<&}!mxPgwa`V0Y59t53`QK8_Q{B4S`~g2<~C-@IbK$5UXR4ZT@l!h^o#(*b zQ<__KZ3v~RQ4X($RAEJ$#1V&pY|VvHIcp5uLyN*!_~8)lIezEcW#hMSkF}4-p97~9 zA1G!&Tze^$+Z&j%CBGrzHuveMaURTNRP!eGp8n4QV`eyQB;dwi+;5}7z?YfX9jN3G zcm^Pcy8K#h`BH1vfwC7jiWpTP&xzM=DhNk9?udn&`5l@#6+qsWp;VJZ3YbH_vZBc9 z_LyF#CVF>of6F4Zn6*1No!QDH+1Sx0{!-oCH7@-%m?^1w^TK$YuQ3Pe6A#RZIK13; zN}1XRP5=0XESm)g)}HOH?2bty?ybptzg`+1^c&qNPG-ESSLXQ??xW>CUWr%n^|12k zmGi*CkMw<^*>Rky~eWMp9m;4WzLy29^-#kKTr~Ot*v` zDFdE|*OL`nk#(nUncS~o5q8;cF|A%-@Uuu!T$dQnRMqtoSsw3*$yg;_I`?F87a-)% zs0D9NB^?XXAe(fmSL;&byj>X3?L?-V?>Z%KBMO?XNDAMLU3E!LRB*RW7F?Fm6r-5?@XEU;9fD`;};V3&rq@?&gqk;}{vl@B#cw+#EN;(2Y$B}CEq725e>5JEC1W?f#MmhpWJ^Q|nW((FAO zTVihiZIx6RKXT?P=T4@QC%Qngz000`qKfm)J0W1Kc{qM*|tr(Ot(eIdt zBn&d0X(Y;GyTpOiesLM!M=RVj9^>l$i5Q}#fZPD4VK0*v+5`f&!DqDiVF5f>($~0* zkM=L0gRN-?i7?f-E z(F@N)Fk$sPq}OSq!@T_}bC<(p61Up&DekZ<7H$u18(*rPs}$TB;0>^oKO3Uv|9<@( z1TmqrXZXGj6JYs-=&#KIvrU+{NP2xV$2HnugGiC|HU$Kg?s#p@30Zp%2g=YCF<=Bt z8efVO@9liS$2YM8rXLA_1g$Y`vapgEVA~Zq4%j+Wom8Wh)9*W0jn{9O4}ITtOj0X5 zOlJax_y8o5FqbNYGw}@5JLYfhC`BNvnBk|n20-KX`)n32JdCB-R&OTUZdw7@U2QL0P_Lz-wk?B;p8gXKaoUsory6vB;ej zad!KKP+&mgN3^8QV|(ew@tzW;@YMU=#GpE{F%e!fmiD}o5YF;a`)U*K-FAOPrTLVk zFgaa@>z*R%SI;4Br24Udr1l)3O;hGDk1BH=HeD$gDBK8`F23Tj*lAk~$A7=otdJKR z6r}tu>+N>#a+$}8LN8!^1+uxw(vjWyV{@Tu(#$$;o9MYVPHrDRFI(#~Us!kDy`zyB z{19{Jw87mPay@El%G}wI^rw2QP1wl-eq5!H&#ud0+rgB_cPd(vKI`Q9ioOyA-xCrX*1G_WY> z!QXFTh#jxX8Ph>wBB$t7++jVTjW)D5C3uIO#v@a(GCK`+2jc)|c_)lXeDG6=cb`Yk zJxNz1lRK5_EYd$cue_af_n--mUDkkL=Gx;F-OhZj4*8{uIw$uUd-F3ijtZqhghE1b zPv-5tWF@MK@#I9qM$hFknP~oEz)jJxuLz=^;0)=yJy3Z;^)fq`HoJJz2i_mYl{*);LVh~7m&kO zZtnw(gN{~uAeQ`MdG&QC?(DV>p`xzjX%;Q+>Gh(TpJNXk4!(S8Cx$C6nHG)uBeF6N&$eW?39~l734*@ z9{p5-;7#L{widq08t@&&S0d2Et%Bay4Eq|N*XZh(MlgQe_d{|F0SjLk>B4GAz}h+5 zA~y{4(qU3s0#&FR#M+V4)ooV+oAA}#^z&-Tg3IZpX-pejuMW~d!S&No4hlIs4pbT_ zm#n<{6hX^;-8X&NW(x4JA9U4b*LkNju;(YXVl()yep1|94C}3V(x{#$Q@dd@wck*T znhzqdNcb+KK&>tH_ngEc(xu~Nf9832^w~6o1E$9@jk5eaVrD{hven=hUhKtNcyaXB zM&&<(Y85VW@Rom5(^tRkQ{vk6y6Fe39ma;Tri5 zT<^xTh?XZ_r}{dulxEcn@mO5*Eqk?0HClPtgei3c2+jkM!)669bgPNd_x;yh5-`O# zQAO9`KfE8u6@U>pB-1^(yFtA$R1HUkN7#j$HQy*GMh5B zt9o2F9gc1026A-Dh$A!yD?OO1!J{%?&YAFslVP#u^i%mKA_YCljG&V$RCo^QQ>BP> zezY!I;cONuz1$OpmGr{zzZwo{jLJD-Lb`p~npUeP$BS}8PJQb3rV|E*NG6C~;Z)C7 zmQ6r}6T<)^`Of96mZhV)Y}zp?V|Hq@K?YGvD7 zl00Efu2zR5@X@C{FM{gbSq73NJ@eH^_Em;GE&TK%Vzy&9EzSg1W^_F+6e-Al6hm=J zEOXHQEXIlYbcYFc*qj=;Za#kiiK}8T3ROf?W))gmnG`-s`M|ml_qBqCNmu`^l#Y@)x`*Y>JXZ0 z?rV!HGO+5%+nUF~PRiL0hY2#qBKB;*2G3k<%~1)VX)zCz@p4hdm*A|nYnIEDd|4b4 zEZbvLS3HtOc9baJHe$Y2I|}CECMU)U6fTe#FGHVnY#UxkDJe>ufI};}wNG}8SZMh_ z32g}<%n$PoRWXZi7EEm^5aH;%tTJxTtWhUp=i)1oPo(R7V%~vOn@C=S8}_czGRk(} zkjc8xohYln4H+++XKfzTuC;7$&4z3}ZrK8+=DaOCLcKI;-cl_jI58V1QmkX$X$qQd zlrIQ~;7+!z(&KZ%!{WvE-eGdDCZ0d`6MCeX3ca6UGu{^Qh$_Q6 zr!PCAed)7U0sZJ)JVd0M&!c2j$W2}g?gCxJ4%!RZG!2(6?-gr$5pE`n7nD*w^Su7k z`g{;<`R&KPjN>9ZicvO;`+-0VrY$&Y?7HB4*L#qS0sV;ej#=Y4b`m@%lyBnQm39*6 za7+|H7q++?(QltD6bY#1P+6sJZa0kVeeQh>)!1?H$@*!5P5Wf;p|zyofETsD2h?-1 zm-dNosrbz*7p}vrwr$G2W=~W``?N-L4&lDOUl;HmQRwO)VR&++`naR?jv%qI+^xjD z>-ykgeU;TPZtRP%$uxRiuH7J{Q5d_s+Na-L=&zkFaNND@tp>x0f;Z!{RTKN5DWA4c z!>F)*#tMf=p@Csb0ZOnVuT-BZ^C)zhI-~T+jW=`qhWBFS)MZl-bDlrNynyy1ICI09 z?Htu8yJ3}{iC>;qw(OTwTIW*}6TfI@rtC0SCg)WR=$;B%XswJ7w~|B|B$qky+`|dOUIhVNCTeR6e>ho8b%g>CVA$1QRdsDgK>QPVMFP zioRNfwO6G|&#%4L@?T#Z>?CO5QD06J8df)dlrzjkV>55^)Y&A$HTekcZ+V|8te0Fv zV$y-K&_TsjkF(Crtcrf(Tid^2+JA4iH~?+F>Ou40!nFQpuJ6BUH#e+yZFm+Cqxc|j z@Jnse>mhqE=?LV$Pt0am&elyrdyc55{tu@bM)kPreJrc4S6j^B;ZHYNZ zi;SXvSMVjjV76qkj=R+Fh)p?rt-n+mZHG55YLwZeKggdv!xe{Du}`v^{jqbQ&q~2V z(nZYfoedBt>RNdwvhBWc=PGpAl%~a;mZ9w-oQ_4xeUaH^^~So+@!dD--STyTre2yr zcHmSN!4a^!B0O3?uwosB*`?oYUtCe9zX62V5<%&57Prj9zCFD?|Ds8^mz_TuHW!Cr zCi*H-LbZZhy8vSI#p+a2l~OcETiD;oI=_l#+>5;p6mx%yT%&zc^9r~9bohj-n5heGO)WQL2SheFTwLIH&nUPWS&7-`P zH@K%3c@OO~k?zMCIWmT2pC&`yzmd61QA`?gj~VHEHdb6Bvv2QTv1rv&jqrPV);nID zX2}h7&Z6i2YxgEMsfpLN2bzkaXFgY$l-+gKqBT`1oUC%5Yda73K{JEO%V|0WL(*yq zzVzEzr{%%QRK|Ok>CYjy4_U1;srcQhml%8jW4<+)jCYPEtt$?|I%X!F21MUvNtJ3u z`_8^69b{u9>R$lI``gilS*^cedAPMT7M44>V;)GcV;cA3GiTDm%0m-t`Oq9eL-((jv{ zmtlkT=Z(?(7lQV(fjz|}T2s0~ajA9J1h-PHxHP&0IC;N{0R(SX&T@m~ZY~VzVXmi+ zAvhz7%pvGUdwF}`r`-do9Jct-_(F4sR&!7E)Oen1Tr8~mWU-7kRk|V?tSg_Ah8rKu z!o+W`>O>fq6DJyFBj>$xDaW3XOr!Ig1mn+G*xNXWxzVi~gX>?$%s~5rXhJaF;qCSB zfO;vAUzDbyxi7SKlD>xiG)*ifb*RUs%cU`;k@2l0j)31`Vw7Cj6A*i*aVEjwHj4vo z)nAT*B26<98OZ6>kr;9*dO7Mzd%E!3Sgy2|`OcFB-Bj2hD5n%pEqcMrKczj#c~bXs z?wXDV(s1zD%)6vx?T+}}5UK8%w|&Z)^;|MZmrNV9R^uyIzt$=Z5P7xLL9AH)PD_npCIZ1y73}9g zl*B-m>R$83yrW=9`^Bm;Usi9@Y`S_=;9PUXL6QynwGxtlMacdquj7wQ~`Mpi)4jNQhgGw`({ zfic5g-$X&RIwyf6&6ugVNydaV;y#cZ?WtJ~FG6d1DIx6r_GTK`+1-6^E+Odp>tA1{4(M2|-B9SC!VMO(IK zkr*9a0;YGuO;_YQ=mu#)-X*hr1-@!rIRSvRox^y=BMz;T&XV(vaiaFtqqX+w{TCam z&UV%HF^GOHp`n1RrqghQ$9iiv^-4uT#%iV?f5!o+?L06*>@v519!?mORV27QPB$&t z$byKy_7MKD@_opuff7w#uuS3R?Cc!PCg~}G-EtCvo}9>H(S|=Z=PNmCsUnB=6m}g9 z=jIB8HTUvZu4YmyJ&Ql9*gDeC&c+Mozw_qkW1aZPI3eODXxE%B)bRe-9Ol-YpSI=m z)m=}^EsX;r_^ccP$011;&S{$EspJNf3-QqU8qIh=*05O8;BZFtSmDPce<|2AQ+kP; zx9r6nD!Sj5;+qY1HvD9X!b+eUa{KM5|rLi9SIC zzE_6Usa*#XxhFBI^XcXBkjmE!t6sI9%h5?6d=aAGkI&CPg+JE&{4tGO%Ak9-W%m(% z1xADVYOU8DZclZlI6h+N)y{M+nwv6NCGFRW($ije?|(Oq=K8!S95)68a#HM@ErUV? z+vwnz#S`2lPsl(rKwZ#2vu?BeXh@f?t4uiImIDLO>t-y83b zR;)x=EXo>IIt31vY99z~I`vby2fNm{Kyx?yP%@d_a_VdO)dNeTKNTN5u-pz>9!i|r zS$XZ_ei)>hb8g@g*_bK1A7 zNGNU$*Nai9hEUzkBI%pBo5801>^(V!`wO<`mq?iu0ZPM;SKc$o9@+31ZGjZF8zrDq_~9j`gKmc-+p3)O+~6cqC|&hz1nzG$Z8yOp^mrm z*1&FMgZR19e!_G3LLrAX!qlWck#lGaw%%(~*GA)LBobFt>hQ9)-_>VyXZ$^Th0o%f z`7%t&fDN<1S^wTyS?z)z)uC>XEW;ob|Uk z$mHfG3pu{0+7;`XSFS-%MsNL>nw@wq{M=)qXlnKsB4JWPmofI$NEAnIaS`FmzQN02b z$FdtvT6lCs)Lk*1pxb74W;B5)oL?TV(C(4NnAVu2`H4H5 zRUk4`#>cY1^pE?MOx)z84L5Ua)e;8WEf-|E_%6G@7sq76DT==y%9ZLrW$-Z@jC5p8 zT+chI<&sQU0}s^NKB}#H(ShJf!FN%Isj9T1qY=$@$CJJ)eJ|Un-~}aJT^LAia|q2A zVuCBiRPC5)p!?$1odLfcg51fmwbvF>z6OJ^OF;8<3{q>e6d_Eq`#+mK<1{1IiFYlC zs*jv(IQe6)Oty&cK3hnwn2l4n39`-Ad60jr2=D*oP#INUk!Drkh&%+N2e=!Q#`r6J zC6Q$=#@%8E|3wq}kF)UhCAM>{A1*uo{&Cwak}i(w`*Ex8$^ivzg(r91)ctQT^+@Ok zC(xgD2Mu*P<;km`^+L2tui1h}Y8A#b;H{~ej@c^^S|kirJLxF2Jlnq6-6ya21V%S$ zo2}qYot?m~ugx9b6ZyzxX1bIQrWS0o=8-gpvJP8|q)EW6bocKhI6mVr-gCNA9BuaP zA=|S8>IxyB6)wlS8_pSJ8FJ_F4wi?FxH6Z1Zmy^KK?Bk`C5fp%7t@cV%vu@`KlO{K z$%)2(G7k`C+a zn}ll*n<|dem2|_L#`YtdRPZOxr8trM5_Zn4Tn1uvv1aAgsTTnz*wQD;ZNG!NH-q&dgVW#r1FLQmx=l{TY>AnA zUnzI}Mp9|$gAjQdu1sx%kAu~fT?Z_*DD(_UY)Z6PU1i-5G9_xOU1_CD{q?1h|5IIIF09M}iuSOPDb8 z!xZeUrc$^g2i@VT^cTC6Lu~lq{s?aj-)5BPqrQkrI;f$}g);fP_~DDHzu($jH=k=i zzndJBJ}aezNLOr;aUy+}H(Y$zm({li4lSHT992DU(e&x4s@*L_W)G%T=e&~h>14n1 zi8ctceUma-{I^yNTu-iul^t%2hQ<}U^>%gjK#a1xCd{eNz=_> zaNS&W+UFh<@2Zm!_jrdcOL_4&edn5`|1UWHDcP zm+V~@y?iuvaja=>I*oLKMj23}bwj{M8+<|UoE0D@xr23;cGWZ>xO=oPst4iQCmw`u zlskthLX&cknQ^|Fozzr>*rjq1yOJ;0E3&sHn4rRxA0Rd5^&@KI3w#<{bYXiGGom7O ze#R-Oc`puzHI#`|IjiEX zu%?tty&tiOs5g+}cN5n;Uz<{PeA5)?KTNuC>@^RTDz<4v?nG&4cN`yrPk7f8j-{f? zYB#{e+06!3#skxDkF4tsz#~o(virxA+e|oVrDOM?Mq_Ak$Y4rIiL z@QRG*1(n$$r5-f|tbDDii3_{uW2j6J9hws&b)Y>&^S9q*JuNLWm*f)WYxprhTppsg zvf^-{mSBR(s?162hd-RS(*!ESL1j~h6^J=% ziF|-H+250p=krbO`1>FkgUpTCBp!id^kC4MmJN!3tnSw3heelWXph5`Omg*fKvj`(!DXlKBUpd!QTVi`ZIPGSyZhM#J1Jj-i> zXqo1s0MzJu1HYl}xz`G&yQg&T8sltiQV;7lFTT2fJ(=&@??nlhHO!+@LE8Dd2H z`-4KDJD|`a!-Uq+sG zt2!m%5>Ra&q$w0A_*h~=eOHARSZhBB5Re2b7C*NG_w9iCVcp}3N(uNgZ7sNsM2uds z5WiVwgS9g=d^}=vMoNNsV$Rkvdgs=_A=Nv_Lj^9>Y8T`3?>(8)2m?zO+}S*t^q0b0 zDVMSKS5tK9f<7naY-h@2t}K;QEl2uw#23c^ki2-7$QattTS4^H@#8C6*b5s5#R@(_ zZ36|^q0m_o-_*gB47j>|u@)mv{fHpTbA))@5c=iz_r(cjyFb|$K;6TNF#}!PZ`Pl0 zUH-l_6odb;r7>?>_5&;5RJb00iXo!*VOkkzUtsi&+_rgMYWtMjK{5zIiKXzP*0K&oOe7H?ghQnEY{JGr-I%pf;>qebi~#8Mq#U zOt~4h*@MVU+QK;ZWah8kP$m&@H=3uDuX3Wk*?sm^nCA;qyvyV3I%ChYV^&tKImt~J zvTZ|FMhEJf8b^XgS252!G=+blz1q#@X%h@_b=v_0pdr;@(|Li3l&mM8%&3AYLdxav zUj{>6MA)IJfZ{|uaKY<})?`88JZkM!g>7P7kbtFdUu66qMC-e;NZk8BMF#o%=GnTa zk?)$@F=&EHfWM-=L)YfnTo+;}fslU2iRRdd57#C%hsPcYd+Y{4?6tL1L#UA}Xdb^! zHMHLh^j)|A%x3$N7KqEaIKn8BY`e*fqq6f5+Et&84`YgEw)?~sa#ONEs-q}`dUsSB zG?UZ!9y1ZLD{1`D0>pAhC6IWEt;8_c-8t4R4?WZ5+wd0ptn`Y=)#;N5{2QJwkS*tq zryH5~4xc_I-VE#>$3VOJnkNuk&9LJez#VYk_01;2Y?VB;GufRwu3J47)jqv()l|Z~ zdI8AdbFWFaw$;=Lh@trV5z7W>FF3BA$kqjiY-Ehqk=n!*5}sF@Ta7(6j7cmKo=92- z;rT~Z)QCq36aVD#`VpS*1z@yzwdaE@zc9(+jGl~!r*2t(yyw#u3#l#&o0a%RFf4J9 z=2bI0Rq!_JkphKiV^i;K)jEw3=L>B6?D>#qxV;p?4hpG#UaS{npJXH8!illU#T=F< zcgiM3_94k-R|_Bfh?;}f2&v*8ic0e8aN5FiWkp4gTfR=i!w|s~`HW+Wc`Uu+5t-K| z01azW*W)wT&l(8bLNFvl^(a;}1%7$I^rxkM=5lC%COm;eXR~ zJiFoCe&lBG#fq4U&tw-1-;THPorf~oM839Sk&e`h%P0qb)!d^<5%*Byrxb-(m7P1q zP?TJ@y?zSR*n&nX3}^qNdywx8LLv8XJi2?C;Bj3~J}^{P%Q9OjCkUEk9T@CgVkbKs zG(KHvmfV32(XNwy9gV=Bu38ItJfeUg-%`~KbV4JFxw^<<#4HCIat_K>I5DigzP>H` zy?23Q;sy-FObeU9_jPj}*LEKLe3|!rZ+>Q(pcxJ;J7`E6S!F9C85DmBcbc4smi7k~ z`Dz76C)5n`h3MfUY9lr+{5cBnQ(blX$eO_pHvDuEiM5OBpZ-5e@Bf?e2r)ddYv2c}UAN3l(5kAA0+jNojumpJ|+;$NXs* z%Nt~CHe-&%NCD4#S3J*sctJgF`1L4Owea>N-nT@wrBy+ek#`Te%ZI#I@<)THYmRb1Z}2+zvY@Y4X)fh`niZ zMXw^H9cyxW@|ayfeFp zKjP~gO6=aaRk3n7nA|hF+5+7Hui2=~U1vk6i@IA8>A;*})vU9rDhEvT;LN!AoiE@B7P9ql97phVKOnPe zU@LD1hkPWZn?({Sn}8~!c>(ZKOpuzR>#&lROL--@ZE`uF(%Tg)2L7mG9m6u0(I6bJ z$U;=5jc;6?`$2C*?aNVx63F7#*G$oc^3O$Bh;7oX7#D(Z8dPhNHxC6>o1L#ngas`C zr5?J)-56}7YTX;!+4X&Ur`+5FO0>G(O5J8^ZsSj{hEju1<05`F4??Y{EzYQolkStN zB7Na6n+3=IciOBdxg6TUdUdU8E;CrF2Jeq#TUt8D@Sv#1$9(~gE zrtSGaV%vQfO54Y9wVs;sYu)ml4KvJn4sm*`n=hwhEAP=LZgQU8plp&q*)xeA7Yx1t zEi0J`i)SrZ^ID=@T})UmX7D(wLpJoC5@e#qk(hDtP^Y9ptJ<-8_{@S_iM7-|SPJ}+ z8cWi>j{-$D5L877N~8Pkq@vXGC^UgW=Jqqnd5vsGUyVny9rE2soNslS_&5@;@02JS zA0QHh@MG|%JDTaG^rx*B!D`Tx6OE=bFvnWwk)o;_Zt4>&AoDZW1xTRgp0&|UQ56KhutuR#0WrKXR;A1_zew>a+X z)@P`*9Pg2dy^Dr)C+-mgnvWXFtYMRN0jbd@RlM96$n|}tTR=*o(j&?BZN^e=ArE`C z_L$z(c79>$g3|xML>^3-4Ds8S;#IT)m~n170i~yZjZIA34C6`Ch!n$)eK-ioHyK8$ zYgmemSMyDK$;-;!K0Z&45xH|bAUyym5m1b#W@K=;+fIITfgpTYq0c$R@i=s3$iztv zW#Y2O!9l)o1r)I1;{?L7gZCet57wIZT*w!;q*zKk0Y##af>GpXt?a#n)hH$ZS>T0P z&*Ty<_x8(bpP+>|ONsZ+-?5gTosMqkuQB$Yg|a#72?+o0oO3c&Q6_dn(nte7@a>lm zm%}WsAv-PvngxRll+dPQh3d4+>PPj~3Z7F}nA!Qy9?{%1ra@7sy39*WG=H3CGx@bw zkJ=IpvlZbIVT{+#xk>G{#qBdmvGzSRJrDPYvtqF8h3o`N2W7+ zfkZyGDK~jv;^!wLsO=q6)szLw)+1Gs!*;`Ct-=~xbl14)pkD;@{b^5|=Ki#y`cv_C zkr<8alq{iEatmH^%n*z}4ZcAgIKG~cfoxk>m@oo2Ub-af-{Jh9T>xq$Qri$m@2(8Y zd^3!X?7u3;%yF@rZ0*|V31*pXQEfZkX(60WxL91Dz4z_Yd+vj5Lcv}}hES{GbAey& zIvyRE8*Z)uni);j=f%XG#(X^3MOjN-|1kIASsKQE-4E<$%uS0p>tUM%GF02&KLx^4 z%Lg_HYwX!hpVUv64~(At;Feh%?S&?rMG+k^*Nm{O7DD6mCIpUMs!YYg00MvSO0I*b z5Smhd(#IozbZ^IVwQkU?cf$vIx2d^PN0W5>rDa{-ZnZi`Z>o)V^@K@5oT1V)6y@pn zFsn{0tu^pl71guZAZg$R)fh@#?p87LMWN1sg_0pyaWbZ zcPPq_jF7eLfy31G6+5)C@D0@xpb@x|rrT@4o-yUcW} zz$(IT<+A_0%+r2|vkO{Jc~`&6V9bWG62n+vo9vKfrY1~S^9ckkXi04qjA8ny_30ND zL+uhYAbOMzN?UOS@}aWLQ@@~DX4@QhR(hdtATzE6Txn#-?ovR|a@+L)8@%?rSVo6u zC|*s|>1+p#A>sLG*uE~%J8J@{g2b4VKk4PA_sRr2f+XCaz=N1Z$(-9tUVeM~tZ%1<2oHxi4seZ~q;P;tnL6#Amku!3&o zqERZpd9wb<;bSJ8KYq1|*_QVF&0)O?h0RNh9iy24eCyikbZIQuv%l|m)9VI8Dr6H6 zt@^O{iiJJSyV@MLmd~cb(ElZi{Xtj4${A&-ALH=%K+58c@U;T!^L9q>#Ak#HcV!5_vsm%iZaAu{7!Rj40WvmxZx zz(jXpNM`K1f4tdqbxxRuXRsuv26VOad_l<8j`_7op%y0lL(Tn%#JB`x*wJhU z)CJY}uD1D#(Op_P15RH=wqCF6KX|_eO}@NhrYjjzj7c;lRi~dpL49_9&yl+Ktgt)dOza0+>@A9Y7^e@ zW=SIv!FSc{_Un{7e`bRq16ry~Qwj!k)Yry2sfFr?e{Mfc3n53|l3#df5;fu{+z z9@8e3&3QUg9>LfE?*V;OKS7aMMfs%-ZY5&14cRg|H0O-q#K^G3>}ul&@Rc=}%s~d6`u?JCP>{iV z`=*-u<>>~pHBxQubkI&N`%`mLSelzjmAKXcBj)h^oaXL2rak8~>9lH&=00W)G{94g zVqb*=Ns0_j1DZhq8){Kdx2RO=NLKAA3Jp^{j#IikM=ZaRmTfNMY#V`rJg9j- zzY5Yz5_GTAe2TvDjf1VbU0%e&Ngg}~)HYSkAER!KOYT!!fsS!jKJevSIXR@2x)on6 zdg~X1qAXt&Bqu4CCa@_8C$)}KFfU4L*oZYFEgM^>yF{<@Cs+j^dvf5JjHlc3tEyYb zN^J{OAH;Fe{1tx!b7dZ47PlOn$`SDuss2_!BW^EXbXSa<;UcEuXE!~~YV&^#wbAt~ zaC9{Y9L5~Bs1JXUa4zKJD)C>ra3l2HVQp}S%>%URTuW+@{e%7mGB}_Q40ZTc?jKNp z+_D5N8ynC|IqIuA66VuiQLTIWvwLEx*7#(Tew;sNdrZk}N0-T_w8a%mM-BtZ>}g26 zQnuGDy`lYLSMyy54Qi>FCI6aGtAIuU`9%ZiZY({=&3u4D%K+thzPj33GfI*?Wh=Y= zqJ?yNe76lLe7raM!IswJ%JWpELPg85e#t%3@ZUi&4WUOTA~q9rN7(;YXo7H2mwXE zG&e-+j0g3bcGA+EAg1Pb$2E~(nbfV#-r~_Fd;}&)q)o6zU24 z01?ryDp1Q^;XlLYHMuHoYLJ=`8NB8k`*cQDy7%}TjEI1(ghLm*NQmk&?D2kC2}@4v zpvNMQn90W%v*^l~pdP+{A4k`_!ECnYq%BQi&x*NAyaf%kYVKpr%M%k8L_lxWRSX`F zW%G322cBQrf>PZtF&XzK$oEhI>Nh!KiW=v~QKoEAeagM_IZkLtQ%+mOqFgLEjdlea za-3h`C{UT-S>WBhSdgSUbyU2o)saRJT=DFle<$rC1RywK906<6#~T=7%x0aj@^Ptz zbV~j84bI|ckpGnMkeV(A^V*qaS?3F%4YEHx%C+VZvu=EKR>Ja*g8S?h>7fj#%!^e* zt-c`#*89_~*??JG+H14+xmuU9n%*#lj-z``Tg_z=%Y_OFW}Fy4&yIusA4Ox|2u~Fi ze`93Kuf7udYrCp2@18pW(94esJ-Hnwc44xXZJ45TFqJv&X{K#Gy2QFnn(m761Sj#bD6GQ!}o@RF$*wy}K zpR*Mg&?d4FP3CCXh)sd@Z{9SVTEYy61Tat30BzH4Q6EkW6!v`AS2$mD-|;AQpmV=_ zd&MTDpsxuRxd{)cF&61)?`sp4PcAjBcyl(cn$%7Wg_qBI4YwX$T8rGwC~+qUeCE+H z3>;?-2Y3+T5$M-+u|U$))MMhMrUgug$5Ae&Hm9yfmZifDl2^+^i&2d~XP%PMQYnSc zF?JP}T|N>fO0lGmU1a>O#2=-h(&;o)69-=i9k zvcMnJ0UH$On>4KGZ}ZjX%~Vv}xdHYXe!6URoZb;LbWFP3l^Ztn%tz;yqWZ;np^yUd zSLQd6-WeGC-N5L}q8ZdCN{S0C21q0%;%rd;>f!k9)$SgHRBjl9IEC7fFtVyr=cap{ zVT-qYy%#PznPb*5A9tm(G_pRD7UJ!mRqZHo1CcniGKVS ziF#9ElU?U;n?x*bj6$!4eF-jIn{N16+mA$1lBB^CXxg={QZiv6V8a- zXrE>1$-YzvWKFX$&jW@&!Jp8Ls$_NVK_M}#qS~5izfjlSCi2&Z8!DFBki$tik9#bzb%AGE?*{Mi;iO2+EwhY{i}9u+>q!OtdT~<=y8^!bLLtn`D3b?! z90p&}8N?{iA=%WmXKOg0-T_|;%hvDxsuq90VyqOsT7J5#1*D_`%u;z$cHZ3Grp3US zOznun?S*pfjo}?ZQ>#59L+{~Xv>^r;-5GmW0saYk(uyCtCjm|;4yWazYT&H-eru}Z z_1>kuAU^nRb&yuKQ@EmB)Ty1|GMXR}NF%d|VoXO>+irVBKou$WtiVbXaEt22rZzRS>TQ0{;$yiZ_ z*cQL^E{KRG$hQ!fnpQjdvJtJIt@d}Fd{KtFx{3AtbSXb~Azi+PjF{kg8^y97S+1z# z(C@lqCsuTz_b19sQ*-Hq?}P;EEJMmZ{F(=#d+aMRd7fKe3OENgYW1r@pK|8Vs3fI? z$W08<#!OwY8pMJP=oPH?waa6p8HDc0iQ&rTRcTAR3h56x)%MXdPQ%wMI-VwFE4ynz zlRD_gl!EfjEqG5IS#3=oue>^!!kZXu;HyoTBYXV7iw1n$4c@`R%de~0b(@l zo9^@;IKo8MbWt?Zb(@Fh+|hp3+fe1tFsCE0l>Y3c0}2((T!0$yAe0+4n7v<<);$9I z;;97^Z-eGdl#j-{fIg)Mk0mC)_N9GEq6vizgiAa}c3z=?^B?&fdx!!ABJw(jYZ3EM zf)MBme&nUnJL~5y)gmm2pOYsB7k;u25RSC$xQ^>(WJ1d$ zH+kBaWqdqwlgj5b17TOmUOkC9lm<*lwp`n4^G^9~tN&JL7}GJ}WWV?PGy8tgD*P$w zUT@mKdEIBX^DZiAj;^F%>l5Vugb0+3yCOli9-L+IB|8Dao_4CtZa!y^SsiwT3j<#< zh~agy#y{c-wDOslm=pPm*>QehgQiAqB4L!)*8K^IT!pJ(f2daMIZ4;F>)6KI&4VWy zgyHUrI=`~4(TMm5uab`ATErF)aE=FMb3iN~KIPKW)4MN2wg~g>xceZf`JKbpilZ+6 zuy3TgJbw^Z;P{P?vd*TnF)DRw{V4{iN~FoitJotUS@;8sP_-J(ea+D3SThweM+RFL zVAEtS5O`(!p4ODC$h(AVNXWPq;}oED*9`RF5>Ib^f_oMeY~~~SrZyrO(cTq`_=mKI zY}J|DiBx>RE*q@9vW*ulRwW`xK*LvL>@n*y{13i>C^*_N>z&{+DD){#P-hX8x^qT~ zr3(FeQRyEM#nSpD(?fW+0;bIjxRxrLj)#+#=y}iMx6S^KOrUVmkCyEXX40Gj%Fef` z);XXIls);1EHEs-32q}Llsw{LxIbePQAT^LnmJlKB3Jji!B@rsK0j7f>~Vottn9x{>LCRlei zHxT6ZWAWxCv@)R$juATi{tl4@oN2pKsaoFNK1BNP_cdWCkvRTmDniLDuK8S0@5!SAM~#k|M#r_t~0(#H}?|8HcV{pSj27FH|D&fL(jWrrS;O@YqQw*Y2gF z4=XIjrsF)dq`Nl8XU0MM9Z-x=YyFzMSAqL%<8-A=$P`T*qs~$VlthFDCD7@4R7`=R>`jsB#D{o|D+kNSUDEhaa+YAfMuyjo`FT9X&@r( zvM=^@E88as8R7F|HM|3T$QO|FE;prv2gWao!Eyl^h-=_6_5P%7>t_9C+;5=)zv4lSzqk^~?^ zT5az-r&CGFySwkUaf;yv1yz|stsN>fJH8?Ln-c{8mD$=nYxL$7rwK{QtyInZ!C86> zpK5JIAIXT#<}yVLW!P`rHhZ;y(~`BlgrcZDH^Z~`A~7Yv{l6?U+D`pLbHCNUP5G~U z7H%zdg9?-W(2uLRDt5DvhL1{$|L#)PUYkxVEwLi^H@UM^N9dNX>ie7 zu(FhM1Mc?1%%@Qu(SaKN9EIHM#W;GRWT?4F=*3UVr5*tO5gN0nQ=n*L6|iu2}F2nB4%}0N@@2NsELBT zzCF?ZJ>XU`BQD$E{KJuldB=4gvzC^2qy5V7^g?OGHL%0Xm>BUA+uI=eZCZALB8$4$ zl#=3|PiU1k&+A-D?RUjZC8clPyt%Oq^v@_ZpX&h6SL zsDL{PRnPmWxnP9xkx>1M!_BcO&lkvPgCX9D=SEmh#PeqLI$i4L5yQqcZgFd)du$um z9dw!LB-LrvdIIoUvMbF1v56`DiHKI2Lt7$bxxegwpdCnz7VvlN%|4bL^AhtIHy<-j&960zmS zV^sqV{k8ZA#8O@~zK@=KH3klR8FKWv|FLC)4YKY@IGu`^r44K+s`z=KHzXPulh4$c z{f>D40p0>3CLreM_@P-uz5ldRC{=g5W(Z{65jo*vT=IyKN4B;KeNX#`{s$@b?!*{PY_;t1UPjjwKM%BIC|=KU!!*wG z%lXZ0b=vM(k6E1wTl&@UeC1OQVL3JvSkv@cxHYqW!5*G2(s8)m!f)OU^rubyqJBk| zZcix0IzIQzT5%2Zy&ORxJ)q{^X6z>)9#{#@TIL6vlFo%!?m<_T87Ti@smqzp=Lp-V zUHuyRzbwtkc46%! zGWVv|LE*}sZ9LZP?JfgaqzyGwxkyO>KpaoUQDbC(1C}|gXAR9Gz&8z8Rol<0I=>J7 zI{c;Daqb?dYAL&+-{0KEvqzn>{vZgqq`2N__Eiv6ZWr4S+`2sj+NNp4n>`#21lHP1 zgj*ti{nm(y;UPt8!te`<-*Spc3xM|GGk=KR%<b0hXY|Z^f zt)e{Ai?205t9@f1kk6$=rl)0k`5^Og&&9i#N77w4kMF}Z^XD?+Zw11$UG&YM=d(i3IIp`H*y>2byTa<7SeMNelF2xravyG^b@TEPVP-VPp?TaDEFS0@^cBmYc$oUYF-tOWWDxao6~GhM}Ek*htM_k-vT*{s}4QiO(5@oblG!9AqQ38{}3yHZUNv+G4<=?6aT(IgR}Q3 zP{sL9G2*=b*6-fef6@oX-al5RUwXwld-a>Bt{T^kSFiQiw~OjM9aCNAtt0U?9KNpO zJ~@rf)aPj6w!YIuK$eH0E2%LnOxMO~ltC!z`C$y#QW89=OmeMDRveXtgPiu4Xk2Th z-(p@&-O^v$OjwLv1kl%dt2&puTL(TI(}EBU(ZcrW$tbf0eTUA)uThHYwgnnbgIZdm z_vr?9(>I_1In>~hZDR*Bxo&P+RG%%$_>3-h>N>Rk{(vYwj*@jQ%J7xp4Kn(SFZ^yM z^?Ju?u$l?)r1N~v14nJ!&?EjJVz)-G^mx78Mf1GCk&i;q;K)r`I83MAdPo1$(7>zC zwgT5B-2e+S2AsCqUcq5|h(mAa%S;{}PPN%)6?N+uGIjok1GNj9)ckN4q6`}&5!;}C zO>RSy%#7mzoQunz3f}VSbxHi?lp-#gdBdCdQQ<+myeYm|T~H30;XHD9u8HXx0T=KE0 z41d~IrDdu@t|SVQLC!3;-J6(BzHj3cMO#>k95^_?;dRK?lKJ-FDc|*5yUsq|TFUrx z**j&QkN2oP7CRFVO+t+@i~7;@^hQ{g{iiTjQ)Tk{!K2#yxR3AagOA9~Ws8Hsb%p6c zYR%OEUutji8(rqFO@tO}CxV!rjShcC;|Ii2#QRtMU8jdW-6ft7 zUa}iIp~)PoAm6%qOBn#lsu59wX(7jv?sI>BI#-5U_wGJmnllb4GzohS;S7UQ1juqd z#0x1BcK3L*AaUa|*?P6c-_0MiE2e1BYu;5IOjw?NytGpDu=sZ+$j859A@j}8)k{2` zd70_Bb>r8-(ir2Lb?i+bfS7vf?4!q4HM%$7A=I@2>w0?-pCt8&!{W2^>y}OHfYD+6vu-B5ZnmZT62N-)LDUbgWh!{oY3Gp%Z$*1_!|B35w&tt_ zj&-g8t0ig%>MUX1rw}(y7u@=CLaB=_g&hq3saudoeH$Nx1Hg8MD%M6eY1e7X(6OL( zM*h0UfKruW+NzJ`J7WT5ECW8{)JSl;%c+*QFi?rV7r=y@s9E~TU5$T|$aW&W`Pihy zUfgsI-e#E#{a8{PdP8lqK~{4=83J#Jd=rL8%4OFh@VEDauZFm#kznnAG% zNup%0|0nZpsJ*JR$2ZwDYddj zM17>92Bya0YSYeoVnBNilnV5wU(QLNy^7qaxXP&iJ&1V2XHtc7(9jkDr}=Kxlj7n- zy|X0`lsCz5q6exJdP+dPK}`FIO+ zz0tVP=6#CgNlHpT9z{})GOmZ-D9h(qGbjDJWqy%QBiVFOV`d-FdR z^iq7dlwOH(+IMqL`+uXI+@KxZvE)%Jt6!LdeTlgH9HCVe8{N~^og1Z9?La;UDl6Lte%np`O_jeR`eTI-5!o$; zsE=prRZ_Z1?@^ELEl%<`oB40UIYx%>B%_-|qa{doiwfX=IQc9`?m3 z2e!5QIdkMB3Pz(etgUc#t;JfFSb*ri@yj5KO_yXhO%H>}+}`b=fi(e_^VGLe{eK=z@RKXGh@nJH3xm^>1*(O^nE8W{Z$hFd}{B_ZjM>>kveWN(pV?v*DCi-!c_;NIEa9T^- zP;SCKNy}K1ZSWgRs^gB@(N9=FKmY@-&9uFZx`EDr!kYWi29q+&LPKwNtZEm$p(UqAs34<`P3GMMJ$xKkKh+?6cw_k3%n1}n< zm%*p`qfnvlbMx`I`(BQxRHi!)(V{74#Lp zD>$N`(*h8JP{Mr~uW$C%o;E^8#iZM8X^Ng#$hB<}<9V;}4uH}hG~=N1=gzwNtqJ@) z%z0ITjpsxMV&~PXSNoofd4CN`dI8CR9ClLmMbl12fSR5|j8t11s_^HfK2;9vfUr$g z9cQr(e3jJr*&wR^%$1=_vwJm{Sp-+N9D00#3T{(S5_8FL2)%xAE_Mi93VpM#I8oxS zRN`FAcO%SAZxgL7(%!7uet479da|B3J%EV~5zJh1Ei8cWwdTOU_@!+FOe_D<=vIAF z+t71j`xPs0E2pA!E4!7+Ih&tIWcYCg(Oo)`7%^t1ZugIJfmnWa!CbEC{!BxY?DxYg zfhZLHUQGFJbE-J6aLeA+MikTLIl&dZlAhMkcnUe42WSJ|p?@yXzVu~J`Ptw>_z*+? zw()jF36wzcMv?68PdLr7@c9qXqKwzKnK1sEXU2GhI!($UrITwn-)@`@If_(x|N8ZB z;g@U_CJiw^wNfik3<v4^YHokHJ@A=-MJ|AYLnjycgrM zHiXg{%dC7*rabUp%tpfPr2wr{dQy$n-_V|n=&*l)BE|nmhBUKxN7Q+Rbrh(6DE-oKF5^ z62$msPyy?pO7}manqPV|NH#vjzVTwGLKbf`KwTuzSXEIHeDp3v4i?^GaBRzw`KvR$ z^fl&lio)IL0IS=w1bJq(S<XzRsrmLPV1!Bu-CXcl4^YKf(Oo-?)? z3;GCOy)!TPdjvsxDD=Vaj=1AK;*+`dhRlhDQf&}(5Z@i2cfdL&Yb^}W&A)G3CY)aC zPq_&uUvtb{@vdJRo(PjZ;>KlrtgCxz*vcbLX@$TfOs=^_3^(*r@Z*pVPW!%V_57n1 zka-JO+_Wv3nK&?9uK=DFm(Ex?8aXuBW-4T()$f{tZ4lwQ#plhn zv&hm|?ty8%$tV2bhq|@G<6-Xr2(AlvYKWN0gKU3}zBQa%u{!US11L5(+cO;YzY096 zaCK-=0+Kk$J15s+AUXB5JyQ_JK-?N}jI8!S+D@9%S;q>g6f-k$pdN_us)G4jkrzPS zBrQ<;Z=AF}XsN0A5$%WzNbvyAnkdw2Jkh58Mt5{SOXIGasQU&RE&4+-Llr`iyd#Rf z#rsSpa#?nwMxu9_pH+rTA}HJk%D#B0OPl1Oi{WrEP9!>$bIU?RuRPMXHJ9!iaMVF^ zz%vQ1D=(74&9RFu^XfeugV-!E?1uVxqWIAqth?^`OUis3IL&7g5_ff+DCm(7Gru)% z|H1>apUAX4JvwSJakNA8_?Uff2b46)3C!%yMMpRE3}S?M_4YW?0N(QLcl^*N^7y%Mi6!PHx208EGd%qbg%Wu z52lmjTU!FtvgeleVtxE0zhV=T?9kWXA$%*1Du3%r_79KHTL!Qe_bR;`+4o!d>31$x z-RdZ?_O{hoJsRglhde^@`uyr+Y%7aZPp>>xcNtoxkW$#4Zd&j0n>_gVbYB_nkx~d- zmql8(#2#&J@)sr2UWy$X&>WV+c$A?w!&|>jF1=g$2rboJ@?lI=Onhk%pzW{NCl&`k z+9?*;f--+D_~pzis_7XV^&bq{rTG+`}zX%?$k?buVz(j#*UR zI`RQTBnPlS)chO-29r)w`yiQl%x3b23?3JZLXkXB1r@UspPE>y3^4&P>}TtPj+;h% z>l}i6P5C&&=yqPqZ8IYj=g~I~(Z11+b`F)d>7Oo8KBJ%;t+`Roju4E8H#kg)G*Y_TQ&Q>1G?+9RDXHtGqyOG4SoCV7(rn zsTK&lE$krnWq*ozGrwxw(#<=~(iMpn{H(f}26TloLD+lEhTxX{{lM3g6GWjE|67XMr(WHC%5-zwI9mVf54~0E*zc71v5FdD7SLO0r!;ii_ z7}u4_!_1;`R(JCoWZ&TWgc(F@w!9RmxX}-MX2mFZMup}O{B}lXUeI`{?b-C>60_VF zN1$G;)VN<3rIO~Nm!wzx$5CaGKih2&VPR6h1w&O^-?9h(HQ2nnk2>~vrnPPDL}iiH z<)^Ehmg=_;df15OD2%#Ai2s2kI;I2*$g5U`EXS5in3Z)Oy z^qLd**?mF=ynW0jWIe`=xN-jBib8^(nHF`~UTDZN-25n{#56ZL>ygT)IJKgsq5p%2 z+tQlp1HJDW3n~4afB;pl%6j>1iCJ^$x>}>QnTPFl<^>LwH=~E;Mx_zslIGn}S5Ec) z;O8S8u+5UL*Kjg`ona}6kYiI*t!Ao!S_epHH*%27yV$UcFhzSqD)G1 zRmXu3ALra~X?&IpPV$usGLI)FSVp^G7-!+~Zia zSIJIhQ&C|%%^`5pmS)@1i6HJ7X)BY%Oeedv3anpi~EkKvju`6mBW^|K^Ojd5SZN{N~2r zev}$goEP?Io>;X0ax>;p(UCI&2AA{gtBaex*B)5updVPpn0j|@J>|P-^WGG#iJCW{_=xAqIov{OO<@ZrNhxw)>MpK!mTMZAzzSiC)Q$hiHp3|utj`0}KKLl446 z({w)j-cfKsLUWy)N>a#Q9xsbR;g^I(drQ?(Q^Mj#H_Q;~#qHxN710&z9V4wHKl3Se zy`A(i>ppA)hOlR~6+1bR2H{$oLOrM4mAvlg!R|)ipqo+LPwra1c3!{{6y%~nHd(IR z+p}Ta8)E(Pg;HSoJt6~r!A;o9M5qRczmrwc$M$t^FM#^mq%uN5nJt&@{XYvD{_`?i z5`Zz_FVo%swz3JgzxsI*2`)I!3SX9A)V^{_%E;pIvFS^4&H~a6markcyNS^CpS9w~ zpl`IHw<0;aj*mM!g$1yxm;(2V(UN-mOLQy+XQzBt zUVY0*+q)5U?%IQSuq(126CM6<8YynKXmASX(z@tPRy^FSBi338NSlEVXq&yACv)*AGfBP z9a^qXbhAJVPgYQ+ zEN&NRjk$~-?qwDLEEeLj+8GTYemdyS_}-4K)yJ3O<*)NdJeoCGc6??10xIq!VR_lB zoI!>vNiz3A#xAt)F0U1=>Z2UK7ft;)JFWMH0H|M+ob>K*VVBoBLssc#b5B%bpfSJg z6>*h*8e~20=eLUVp-C|A;eL`iZPJe!(buVb_XY==_;+2~^q{ z14`xxLHZEc$GW(#mj%yW`Xt(6S&@HjyMHgixHBYg)!?DnVZ5FBizqs+_;#_M3-|F% zuO=O9%v}Zk8?a9@QD;9rp7LW~M)tf{z&ZLDzD#e!VWdYO%P(qJA<)P~_d*L+wW!Dk>ckT55dbqu(G6Eq-bL-{# z>pzQ6!}+NJ_cKcfSEO!Ym~ir(Rln;jAf4HsVC!_dU`xNr8zs%jFCl$vvo%@DQl%|3 z?;l@n;QcieSf@^6nDtEqPj-u*$~=UAzikp}$fSe4KsGsOb$+K&&$=l&pkS%+_J7Cy ze=G<@D*?Wni#+4w_}9e--FY(7U-i6qO^Yyn5IGnW^_Ph zh>9B4zi;aAXSV$8l#_TDQ7#3?Lq{i?p}To^^;A{~Ivgr)98&&PBb5dC_&@I{MEv{5 z2kE^K3K3|T&yxE6e13f`wmBPtUn=}w)tex!edS?UaHif{;|c1wTlNTX3C3B@))KYz z9vG4L(spVtK_fOkWCvhhi{k%9<_n!--1{b&09&Z>ifr;TzG{Crw&#QTa~_%NUsY3F zZY#Xdhz6Xp9Ma>s|2=Mm;(@~fxuEz*@n8Foy?(NshC!1hAJ=QdT~xcw%35^i%F}%= z{7=V)45dqfH>Y#cFK|U5;f7LWkc^#s$rLw>8Ld(j#iFPwNF zbpI`Z1_4)?q4@mN-|xG40f=TuzdbRIBMPT~(zYzrZQdg0FPtb&`7!%qBkPr`BCl!1 zk%Vh!>{WEr3%~Bun{>dV>@K{wGY|-#o!4EvR?vGV@SzoS?#S_BE|Au0Yf< z^wN>^hd&O#G*^mltNNFpp5~ylJ(1&J%l6~ne^DSrfz?BB`8apRf3)txA5(CmVRAa> zB(0vr>vV6%B+-ebhdeA&2{B1#RX0|-*<=0LK2S(ok5~Mrm`>akMiuSF_r#4!+ zu^i(tx;qo|i2-)Hin-3}whK(3dX5+b;>tV3?beQqRv_h1g9>4ea$q|L4H|FO=>7 znx|g>EzzCpa{t>BD}3SFXXhnznS+XvKhrLN<1dR{>SM@DnicL(Q-?1qPT=G(z@_0j z+Oko_>F$#5El`X>2cHu7Nqw|BEU7+ZA=<12@l6yZsu?PnaE+d=Gc$GT6E{Ib1#TQ)zmV zi7D;3-$IU-niRo0ws=s1M9f06OZ)6OO! zfZQIx=aK3xeO5?JKbdit-hZ{y`)`c@r+;+;>xrbr`_*i4!52sEaGj0O_<-Qv+IhyE z9K5K(@&iBE5y(7seJpi8t8&GsaWK#E!PD=iP0QH+MOt4Z5v)8E? z=(!8218&p#x8x6?gmn2+Z;SZOIAQj6u>KkT{{=K&OdtZtSQ#7s=vT$z7q&2o4i;bi zJZZj?6~w?Q2MzpCQRkvS_;7OO&w?+pJRK_mNc^_LY1TCVu#bQB$e(?$(*#FE{~}Cw z2G~==X^KV*E%E|Z&{vk3@$R+RM8lam!ZPTBPLOOUAA`&xEKkVfh8m?tZd??tiUw6w z?u)T*mFNU)v+ifnTjE3BBC>&;DO+0JyVO7y#>wggoq$v&<;)b^T z(u)5MpZ)WJ9M;_j^LfBo-TwEI{ttig(_sR@>!HI72>EO2<=>38Vh`Ys^9{KAbLQmO#1kI>J8lKd}7VV-}Q zFT2s5D!23$RWT+6$z< zqR|zVh<&{!br|$F<+y*8{1@QOXG=A!_9hJ}Sgv)_%qfF{!X0pqeh#;Q`Sz_;RNst4 zt=G9tmZW&?Fqs=D?w=Kc{zs{=6Zu}nX;EIn`=#RSBbZM$NLKCsuvk}lt*Ox+MYU$u z^y*%H;b-^Stv^lRzecH#>)v0Ciu;=?-HJJIiX2y$R(8L{RbQ-gi`C~`sc3F*iHz9rI7q`xc^FET$VA~!gkD83@+ zxlQX-{3C+?6YgpA6%eqK>{!@wel2Dg?b~y==lnGH zCHaOa7nJ3w%0eVT21wV)_?Cn9(WPZ0A~)TOMrs>Ye|jY=eOCqZhfp3 z>MtoVFbQ{cdzpVC3+H1D49wMtM%K4>N*;NN84{cm*&KgU3;uq9@dDlxyfQ`q>j%%! z?lGob1(GoKO1I&u`6uO{p!J3gCqP%PG7U?16a|?I^Z*{qzjscZwf8u`SGm~viyGdN zEWj=zm}P&{yZ6<6)_a%&dY1iO~p7`#HJSF%sIiRO2(VdtJ{2Vc%R@ zGH~@@=xKkl0Py$UE4)WV2x7k+9{g$#n-la*)b77xp!Uhw%zsl2{h#j83+Q-&W2K^G zf#;c*m8o7=FPW*SAkW7yAi-Z(AFE}vnUx8f3Hu6G_;ZX*mH(KS6;p=g#&Jj9_72Jh*?UAn+HOU?U^)KY-=`ZU|EXqRS8x z`aR*VY5(neATtbFv)FT&r%!*6*^l{2ZN%B=!^8XMQ~2*y)1CGdMpXBady&Vi@2 zKPMo1IH}_5D~Ixb8?!;|fuPfTJw+V)Z#bxtJt)mqW`6zi*kTdBSJj20Apf55*R=ok z{lzdmKRmI?2E&`*W3tOW(3f71PkVg-JW2Xs#xw%0`~NoP_cLHjsOuRl6UHbEnThjy}Y?~ zSn$It7OFXxUJg!-5<{Q#$iATPGZZrzufSLDdQ>*nAI_vvcr*Wmk?ePQuBZ4ryk^Hg z5$&&`^Dgy)@BZntv?srJdj1puv-_ILyO%G}Ea1V|-e)Bxs>eXbKbNa{&FXP%CQ+$i zObXVEi?*~l!575JFy>}TwTf}ZcCQw~k%RMDDLCiE3K3wz(%%U+(^B!x15)6J?`M9O z`YC7Ujd_Xtc$5IlrmOkVbz{{1mp9&kpXO4oP3{Ji^zr7eAl=_OF>^V1 zsTAA;uC2ji(3iVDKX@&(GdV}lzT2RuIV&27!vCETX0A*?zHN>DF|{{!Dv#$Em78no zWG>GpQquLV&yp$Ahg51o$41trXv6l%;*YhW#oT>f#3^qArsk0=kTksUY39Y ze`GVgZ>r)HLb4b5&I7aa_;$4?cF$fvr?Pm4fNLb8Vo7=1Ba1~478>@ZL>9L@kt=|$ zPBe8W>cO09dVko7|8DM)2(9_IAWq_kz03(1fq{;!w4C5Z6O$MJf^HX{-h(yGukH`9 zcEP%d7uMc=7eGwm^Z~m6te{h6NyNOkZ8VDwC)z|!2B=I8;~o7z4NXYs0xm`PF$(ds z_cQn-vE~Jk+H?RJ7`*~31RBh3-A;(Y;S;tPWcM#w%3{|)@g|>4z4iG-0(^{piSXTF zBWP0Awa0<~sbwmU>8DTC@WL^K{>Z1}gd&Z9je!8FqCpS<>W4t-Rp5TIBD6mE^I$bw ziG8~S0`5L)yI49M{w&GpV~P059CgEnq7>!lXI^LCpNa!n@L4~u-5&{l zht6Uwt$OZ&=yO=G#KOEMv0`r=!)y6;VI4w>GR}yN_q!IL5!?vy8TO262!1#Jf==fn zi4+>jws1n~?xsK~{FM;a4pqx<6m5mw&mluO!)!%E5&QMwKR@|qt=iYtdq1$)EHo@pLK*~K4Bta=v-$z%A9rU50sulLvvBZMCcQ3qLA!Ps z5Jmm@Ig3dV>zObR@+Kp!)@il2!%^q3%Q;(TFIgoOO@+PMlNbOraat3W&3zsXUpQa# zrOs}fGKbso^yy`d#A+_Hd%dkY(7l?+{j~QZB(U)F3jw!aOfCmu6cbw@a0J9k@eCPp zs(wWm_b$8sY)`)P#`Cp+w^J0i2}XA+eFT~lMbH!K<2j7%0UiD;RH6)ak5ZH4L60bM zdnzdM+G2_>#6x-@wQ=97xZi%=jJ2b7N=(Z_HbmWdZEj=a*-P-yZu7h<<*OB5p38M& ziLb8^E^?#<{18{toUE5Mc7tW*Oy)t2H1yg_LZ&0>rDJssYH!%B2(%mR6Tgzn2u8M} z9JEnt^ddm1ao(m&`Lm@Q1!A=X_&3-E+k>6RXAUk;@VAC6@p#;hQL`kHzmkU%2s7xn zes&kJ$6vxilfO9cVq$h~*P7ayAxF-2Q%~AoqY)C@vYa~LqSY={QQ4iyz+CO$oSK(E zPatf+N+eB19DVp#T+GVvd$gU5N1#Mu`-IU+#1yn%M7dI0JMRl`E#RC_G&bd93Y5~ z<7kwcLd=J&#o(yo_#&A4EXGen03pQuC?*Z<_f`5oZr$%9Fah+0X(>o8cY8*lL{748 zxii^%H7qFF%X5;R0QD6!CPX7De$0kvVh7htCgl(uPyFW+#sN>$K5N&ThktqNRkD@)Tw*k9d;DRdVg1JRKpQoi_EI`?_ZKnOYFG~H&2GA?eU{X%yoR)LoC~H)1^8D&EBCX~oZHIs}dvNzP3;5Ok_p8K|Lu!pFxK zeAl00WH2zGYFU(uZ)SSi30zxLhV91g^c zO;N0;yIFZqs^6YDHRApHqt?imF8Nvi$!sHaL=rF=rsay0B0RLf0grVOKlRly!e7Yr zP&TjjCTV5a{VojhELLSrMWZ_@tiu&v+Bt|X-4DhaNHisAw?6nQg;WuY>-7@DSn56t z4=Pi#Z}09lVv}3UsQQPSMP$2;CB4dzJyi1wOx{6QGQv4ADFx1M|U83MzSm zmTh@`juXW>T#2lChP7}{zg47q7G+zAq!)aeRowPQn;XxeSn7G<8{xEuhyEK!wa6E! z^`$}lH-~NIbj4ybgSoOGF{dm#S3lhxbCsWPK4 zDHZB(A$l$J6_$4b_Jvgz8VZfPkWuEB4U!9kX)UAQhS+Spd0g$XWC*9#rD7oSguFIU z`iKuD4^TtYqr}FZ;&ha;IXSkRY)@?NFOu5rULs#E_y|Wy!&~%3)6)9s?Va}NEXKPU zzyY^Qk93V(G=mi~--@0=`7#^nbh78G zpy{{+LSuW=)us}LlAf8_=p-%|jG^I#eB#fBbFxA=N8(BJ=&$-T93!1KM{?3iTj1iN z#Z;OVSp(PaHc~w|H@RG6XiPBlLbSMKhE>gF3}!K1mo+fUr^^^Fc~4k7G*T}eQY3)u zitTZDe4#_f(x?VYzT6kVi&gPgs>SobDzwr;OTLI?`{YTB4r2fS=P?kUn@Q3wJgsRm}Vt zX=8)~JlGef0-wsQu0W-l?K?|kHgO$?@#+mB@%8>tqupgM2NeykSr8_f!i?xW?`my= z_uWVFwkc1j_T5h@X`ydJ5LGRj%?ybx8eFD`pwx_!iXwHzv7S~uCl1!u<~1)VSQ$k5 z3(hr%3L|#QE3y7DDKdmMwK@r{(zvQ_AE}y7t_RvI`$>Du_3;J8q3SlD=S9+1_>}^7 z9o#5B{}__0Vh`?)*4ueY-EY~(6$uiLea{@p>~ses!*AWC#D~{Dg<;JP9M=o8j39ks zdIUENv6wL!h=0zKZm!LovePz?E^D_fii%43b@djTTH&_zJz;`h4(#$!C9w!RH1!7q z>J{Sg=Kyjzn|=6Vw1Zx3L8k#LVfsJhZW35GkwTMK8jiT1K)h$f>~F&om{kMzqt0)u zX*KgKuB=GfS`h@Adlrnh*d2aAa+W^?u(8&JM zcYP?Cu;nzgMY2-xJY57yVL%4df;1a=FC^^@llVHu#F?Ey$b20qP@vPcpHvDpENgOq z)Y{0^t4Md!%Mi5Kh%xH<5T^NZ{)IlWe~S@UWx@%XJd1>VQ*rw)U*TJsGM;PyZdzr) zc-J9J{Gt4jL?7*zL()vVj1CvsQN8=Lh>M#l)yZXA#MTMKyl9xYPP_)!0ci2nu5QXz z@1u}xR2eVe%(S9BG8cGF;k)OWuad>qEbuG#Kac zt|tq1bxdFpjDsu)NzlnGHL0^J0FBqC=DpHyb;K&@Ugxj~8rRl{`(RKk>$=(>-lH?n zt&%I~xc4Jk`q7i)teAP$>M7}kz@g(CZ(@UFpdiL@;WI_hxaau>Iv{UrI&nA1cKP8} z9B&RIqIOaij(gI!YNpXkiG3h@`&-&bBTm~>?$S7F+qGUSJ`e74-pRy^a%1x1 z`O^0tIO+OFHWsU@#X1PZ&Gd#U;PV4YheEY!{b$Cg{S(F7Ij$rmQOqU^-JeN#$Z&Q2 z3;ovh`z_mkRKB+*rjSXB($D-##{1Rq^0aqtcs`ag+Eb!CfhDA6uB{d-8GDP9;+2Ys zYCA=#oWD;vJwdWM3~~bn^=LmOPC;cG7^30>?^Uq)&fJ{oD0^-OnbgV#2|TmUDGN?r?s0hZNr| zRl6gp>lv{_sAtOhdEkcl?s#JiPj1gB$1V!}MR>@T_aJC>cMIq|uD9OTim%8{uM4Up zB(xj+gQQ7{Th~x-dnr%O=e?aBOPF&*a(*|LXe=BS*H{6+w zabS9RX`F~dej7n@s|X%Eyw>&gN5sF5nRaAy_L7a}BzufFFixP~7HVjp(X~t=yKxx3 z-F(`fCiQGYQSiXcm9xgoN9`N8qt|)ck5w}xMO!g=C^c(`xd3le2|4pvu*$BqD!pmt zcn-uvi}KE5hVD{tlWHmo0+uD@^X5exD5^xWx%0y6r(P9;JN-s5Ro0owf=+FTQE#vaOCy*P1i`1wq~jAnTJL#=e2v= z2amnrMLn+2T>o{Bt0Xgv;5j4TGgBg4ho?v5@ z14|$-$Bm{I&o+c7#DR_pN=pfyx%ql2)hQ#MZH+CCd;Vy;BRP`No5PvefJq6jzbOIj zcC&na*oYi+yDN&)CoP#_>$AmfSYq{NiW90Ny*b&8q}ls&_f=>n!O~@$vIXgUBGS?d zNdw!n)*udzh#S%ORL)?5qnLBTCIA`IM)MRD{OwGXH2b^{*eu%|=haHv9L4#vg4)KC zUAwmz2Ucf4U+MFBL{SQH9=LBl;_wG<$i2x$`^Ib0#Ogi9_uubQ_FeNa&TlF#t5HqmNV-wyC~6iF`h^D^_i2^A2*WlwoC`)oC^{u>|#y2uzl{-qpVL5XHaO^(JEJ zW7R1oVza$p?!MiRddFRN`iKNfnwlP1@QB2%wY2^1Z#KHX`cN0$%ta%r|J}}sN%0T$ z8VZ?|bownfvP8q~IIpVPs-dEq_5RskW#)} zxXa0wz{Sz3a@pwi5?RlO!Ox3~5St5(I!7o`SSq_8oo++=_~P4NR z<<_L)qi9YvzFApQwie4Z(1Z~^d#mkwnoSxYP4umHIPIpsOE++q!xyn>Ev|HVGiS_9 z`JjOYjW5{qm?R8)x7@HN-ymMQ+G1XSR18ji|1F8HK_y?F$fGvVCm?pS)=may%Ui*6 z-U@^U+@r=+((>_ZaEi*eACN%V?n1S)5*rW2bEAf z4+DU(bL8{Bl4YXtzWxBF`aBTN1b0(oSBe%8=Na!{^ut#$^Fsw?OFZun&^Exg;_XA> z6Y>#f)gS0WLo}k;98t>YDg>G67F-^>b0&@*sOYaI=x(edCmNoKAmV=0(Vm_pG#$7V zI-dQ}R;DxF`|Pz<$!6^+>krojc-lLYhHO3EDMp2}H=2p-WOJ@xYNy?BZqKv1 zg5$V@m|~TAO$zC&{w?GRb)>g8Br5Nj&IwVCI)CNu1nu#XbFTW$Il_{BPqwaN9!6X5 zp|&)@v}BSvCFkk2$XV&ZrE%U|f=+wAxaQ$)>29xD@C6kuk4=y#7QR~_S5Gc9rKvBw zD1m}+A|F*ysO}pVc+5_hL3CQD!3VKRE5TECyVJDlMc41Fb=O(@UYo`AO2o9;&o^aM z*j+rg&xlYU`N++8xis?p_<~|AiY9~_i4m7hTl@;upExUDX{j`kIajO=ab2}+w19^M z>@e~mwV8pugf9#Gr78-&uKy50DFRT)jIYo9Rcw~{vTX8bM?+-2%v2Q4X7|CFn)~V%%AhegTV1PA1ZOr^9^|k#SE}D zT2`2k^9Oz($#pj+opX~Z)~Xjk!YxLFIi;Q^SqHZ znOt?>Yew?8s%#C_R*p$as21tuY%US*p6IJr9=2_*M*a#c!gvZUa8qh02;~Ey2&%WSJ6b&_9o{XiHJ8lX-!)H+# zsd_o2{CzLaUI5;5dj#{q5bzOl+lKz(vZKS?$1csz5!rm&s*`%!o$SX+=ENzDl0u$X z`Ektq$#FTB*7FnRgJCk4TbXxB^q#y-524BjWa)FN$eT?*0ix1|I)bH-&eq4Mec@ju zswazt#11c{c`+8-uA?y9Z#^xyu}GI|x~{m)S@BzcQ;2j%Z*kqTfrPmc@nDXJ1YFSyK+FgRNTjyVri>ny+URdU{;$1Og~8g#~|v&VhJw|SoqQLKpRm##xg`% zE`u!j#VcKrR=yp!DB_W?6xx!fZ`D4X^C$7A&EsVpH_fdFghb*oB1UXpT5W7tBc?FO zrdq^QIlw$=OnnS*eb<-h*t@ljX@7TyFm_XztzJb-avif<|6%c~avAS{dv1wwa7i*` z1rE{G!~)d)g3$N8QK}0yIH`zX-9_2d3o*HO<6JQwWKX(~}1|1&gv0wmn0cciH*@VX7XVinNu@*!c&5&_^e zBj3rL+>>tuhMJDoGDtXgnMzFH!zbxY?)5>))6Yh2Kxnru`hFX&c}%S9d>kKg{o>}n zT*>b)A6B?`=m6@f*ky@>tTRK-z^zC|N1_K`8l4QL-rj1rhj6_^VneyB1UxjH@YWqF zP%99f)0L=}ufjMl*Wfx&a%sIfV1o_Ts~ls~HB(v9!eX7|(8)#qD=Ql2Lr35Jhdob* zpvGr$ixLaCsSR`!pAu9)yp+{dK^)NHliXdillxjxwjL418PwGjnn3w-c|0uVM9kxh zx}j}!_K>uwNeC%(f6CIG`Fu)>4l@Um5 zo)Pgy{6om%y0<56n#glvR!!tUY>c4_GbqWM)$>)|M$10!a*#S*_8C9*8#q)Vxmo(g zY_kc=PPfy~I_0AF-Sn-8#seuC^Ht`!p6v4^J;U%3iDoLaxcUYA^r9Lh+^YJeE*+;9 zc}@rVOYjFy6^>Be=UgAt%?J3%L-B>|x8v5prY8fV1+skVtafi7u$WD&fpUaT5m4vD z601IogZ5MXa@^ry2fL#S4As4^oi}^vx8>6WA8#)Cg;adJc`{w@*hurfcvlj~L)K#{ z(1Gl=pc7d#lbA#P%_fD-EX@Mo;vu8NNoF_9r`j_f;&iPXNTSR@%htg%uNqFOJu<{q zFE>iFXy)<^#bYqIT*(_bzPM|Mvc4Tsb;_pKs`-4(csHTjeAUoycKf;M5l{(Q+F*9Z zaZfIgUFLRRxx=J$5SE6K=yF2cf6Wz#3_T)XQW1A15?aRGW-^1(Z-Tl|8AKfD2k-bXy8vcH#tk zW^1Fa0#Mge6f50LwgJxM1bSy-zVJXIl`hOdLZ)LNsh&ThU0QjjH1=Xi1W_(u>DEpm zu-*}gq@Qoe010J>!7dQZp*@^X5>$+xZ%gi~aI58nEJMu=O5yPIQb9mHz5MQ-e3&IE zw}jfdmKJIJp`|$6lc`20g3h83{KmE3laz!T$;t8uSlY5xG)I0}syQ!H97A#alw6Y- z2qMsbNFPAjoOo@psGRaBq*H^WTguT2YIH|) zw1^JD^OD%&Ku7=%t-@!(HX~%_ajPuT_KSt7#)%5!@7iutrsc6+GV5C(K&hGSGBE6s-zk zk7VkOx|DP3rhUQc*@le|H!v8keWjH4p`)rV*DlCPkE1a)3HsQ{lXE-~E}^a%RsV#Hmn=rhP9+@6Hj-;$xR>lo9%y zsb%XU@|`c7EZPu{$kRaN$oey)7z=0KdFkg_4aJm4R;ODr z_u^A6GGEf1$i{p7qwFm1V2xdt5KXW&uAW|xpNc8kFjaLb*)(rd-9=h{p(G*TahN0F{h zh!$rdIcI`Ia`R23V;qOsJ!M%Ea;G%$8#}tOAF4|lT~NL$)V|~nsmV9u z+D?1NwYd&MvSZ9*hlZV!93!jDJ&G8jf=+Y=NsSopnpuwxPX{uUvS!y$2ft_ej2_?H za&-gOPV`R4IV#Q_BO@DkOv;9@QRcS`)G~z}_A*NXP-zY=3^Q+bY!kgt*_vx-4`h-z z4opUX9zn(Ww(AN@Fptgc%~Wu-AMd1S;QH!`MZ15Teu?gSfOeo zIw99-fYdeEDC?XQvP^$1`rdwPIN4*WMjS`^rUtPufg@TUvDiXlfiK)~cP?_=@(scT zm2y7+(R#l^usFT1Zi6{`^m%TLMn5`e;buhEbhzT(NNin>Y^p-jOX9A&op|>##1Ww8 z=X;RqN!a#;+#J~1pi~96|K%wlcOanV2aRUa1b-YV3#0O%K-#SobkZ3K%~Pb|X0SEJ zrdEwigjHL(>-chGq>n;}ipL!lL!;d&U`7#x;d3WZND`rGo#gB`Sh4*k@ToRLuxcAltMN&uVavF zNJ<3)5ZKslVR)$L=@bPPyLqk%?iqAGCx49Z%Y|}*yVP<=cvbH5hbMd#LB50Rj6TJa zDt=x}-V9>g;B81V^)j9|2w^aEOQUCb1YW~_1B7$;#<(n=M+u1_v}CYeIcF}Ped9wIHGZKT#G)<7N4naw*B_WTDxGAXDZ)ddK|5T70MS{J6P$8 z_JO%{kiTNXXttcMjBU7UIxcF{2@-=B&1*AyY4KsUHX?~>z<93rU~5xAB3XQ7eR$|S zg@KJ4KWcXhONheC@R+vd5a{e(=>ZhE4IMv~JOckCd%)a7(pD20`d2e~m{K|P&r~WR z5oYuVKbH^Gj%Ty$i#Zt>ca-pI9$|pc8u)LyM=Nit*Rbs3 zP%iuDRaUb4J|2U#iXxMlqo)kuHXOO7uv@=Ge9rKh_bpz-W|XY;Gc3H!;Kh?1QKxMV zLNf3d>O@Pm4Icfsx{C)?*^Cy>gsd(Y+BHt~mw8~N9I+1vv_az(EyP4sS2+M(XFS16G=${V{`I`9MdYg-c_5TQm1SmY)k^o#Qk zk2EJX<>q)>&pdoY4tgs&+2K@)RI`!$1<;IFR8BUSxiRdsb%_5PgBC?OeX)Xp7sG{!;%DjAg4G%l9Ys5-7X>!{Pob*zJW2 zI$jC$r_-dF?a}mUB@1b3D!ej?zc)t_961d>bh4BB6C_3>9r89+BP5VawW9dS!f_xLYE9++?{~9BKLE7%J^gs zTk=ciKG{nulpOVyFYH5bC=+L*B7pAMJzb$sPc|?;CSTkK1O!bd?ad*jrQAsC3qjuW zYT#>Lbol0U1vO=d3DG1y#~JLABFL;5LrhcKA>!uM!pg(Wclq$ux03c%yTtPrSz?Ny`$-*|N9 z&g{4tQHFH7#w$1}4_VjQD>f-8Dnki@8pB0YPS)h6`qY~^nm?Y;n0q9hbGxrm=PS@=}8eztufH{%N^?4B1q*>dBsQrnvhV1z@jd53!yvk6+-2r(jhhj@(f$0?< zGj|}JE`+2U=xag7>Y3Q!S*T9WiVeCES;Ge+4KCG6FL5=_dW_;n0iFREv-Cnvc~<;o zs~}AREn~Hm?~kG(?PnXfw_Gr?2#d3&Ez=j9%a(18lr0@n>7QN64{q7Qw&c4af;S5| zlKhb!@UW@=Dsr!ae8GeSOf(*eE1TPMT_kw?(dNk)CT*02iI3rhIG0pW`EcV3iBeY% zh}hi6cf6dl!%5XVY;Us@J0XRJjhP>ZM(r(iWVVx#oPoa{*hzHEh#9zGZ53)P~+ zY{D>M)ra*tUufANnRbOvB6Z%lWluQa!8;e@H!J8Z08z9Mu;A zD!Z8B)IQ{=9TchXc0JXYuNBgy(yR)dfuTNC(*;zz&DO3F8{r#cF;~l^6Zm2aeDyQs zfT6{eM!|PWrfR3lHqY^y1z3q0KRg)AJK}38nl$5fjaawZ_)a9`^-WL6r}tGezLCKb zfW#3Wf%v7m${kKbBKCpMHTqj?7K`aJ#h2%oJp*|Tp!{(wph565VtGz@7A(S2Pn|~E zr4osaY&K@CZC2Inm#fjRIuH1A1>CnG74uck$`=hv&zisfU9`yVRfc;~dGdH2!$jb7ktTn%fQ_56)!# zOwyq0Wy#pbUOj;=@2SYAYET4PixoF5xnG?G{mr3choA$0k?x@u7SISdV zxZ$kKa~cxFNm)GE)sn7%Pfe#?D~g#D$11*xLv&DIVX4O+#(@+&l6s4VI_>NgSR$mUcEVi-G$S)M zK*m!1DB}IC9?_b8)XyV@4|km7`uaWf1YjE+0*qz`!-_zp!%hh!ea3e1*!F0cj!ElU zpckZGPO9|bbn*lI9Tc95$NxK$(rZKDUB_m!9qmfOl>VAUYH_wAxs13j)}55J?MjPz zTP}Y#jFEE{MkC?8AgM~2LculbhH5Zwhh4Xwk9gR|X~s;MIe)T;QuznuN+PkzFKu_; zTf-T~cb(UI$o_Jx*U<()#v^9%S>@CMyEwu@F4JY9VbA4zYpg5Pk6CF#WzKdbQBR?& zM@L}<;(5HcYSLVo2Y-0zm{Q516MkrOSD_3rzm+Yg(BGlTqHrkw&=<-$lORcot!c6! zLPcNse%Ye4Us>o#M$-uN#L!gaE(2_5NhAqQqu!JQD79FtIiZiC!pQn@T8fa%J~9j9 z;*b&kgs37q`v{jNbCFc4qbvgTxzNd$2T4eU;Q49~j(kvr6wz9=L-$F6zw}0@-eQx^ z`c0C^ZdSL_H^4%Lfz+o8dP7J%uB z(~KBGmpod}I41QIg+e1v^mtCQ2B+J*&$K&19u+AA9Zyic0Z|s64jO5x*6*nPs0DDD zUaR<2Br6y##V>utS51ewLPzc7B~X1DT7-7nP_p>wQHOY1qk}fh59&BRwrF}y3Ha_Z zlaL?5_3wQ`pZKHA1qQDlfk~K8L{mkn`3^h86S(_UqRGtGo?@}rM98xQ996s=Z}vM< zgwk0c&wQL}2*IIau&~;p@tq_K!Hqu2?$h9hLm^b0DOnhM4MubY8L86feg(`zfD-Gl z&7`yjsGk1qC~VGtD%6~4^1l0*`9LS%G$8qNqkQaGoa6~1wQ$$3 z{#&K__AeD&RZEkp6ee?A^k>9D$nbrMqih5=7x}51XFS2i&g%xND(#C;u*0&$Wl_DeP>tz}iiY@R5jTbujtw1*U1lu0||(DP9=7?Yi`k;^N? zex3K;IjK#5V%d7OYh)pn183s5zc;72+a-27>B$C#n=Xe2YWYAM^=auftpOHN>A_|> zO9o!QBM-UA{0cZ=6{kn@3Y8t%iCHK0lyd*@1^qen2#W1xl9)Xb?9SK4f(qh`%f~9& zy8ti$=(jtZrc|$%9vdIc6eLAb*+wbOIoNI?RJ;D+$U-I=CpC_w#v16sXJ#}0eQs&J zBAJZ!u0!c+()J@NvKr@G=>h1Q*s>f34m&v}2$u8J7IyOHe8e6U!7*Fy-Hw`s`p~(f zeYg~99pWTUB^~@C;RTk679sp1Y50GqKUCib@hfiYX*IrvpiMnP=RPx(mV%m)xOMc; zepoKf&glCUsLOooQ|sX_x@p*>I{eAY$8g9O;eFm7ZYsWVq;n#xW!foWeP_)$SG@{+ zBjC)Zy47Q7;Vnd1tUWvrQv7fk3N@&AcXg1PW;R(VJrN0x&IIh#^4Qit#=snlOW8m? zPOluj|8Iz7H6wD0}l^Hmh%f>LIjB zXAXB&zV9_;GpmHlX^OO8fwzf)#sr^r^{p9&a;o|$byJf3SAnK`0HwDdWTMRhZttPz z?j*8i)IdthX=fC8C}*gZPP;)+NHora!D67NOy1Q!MFqb2WUa8+SQ@^#QuC5^eE8yU zE&U;cb7*Xv&13=S4r)2s%35~K`?5eR?i>)w{Owk4K z0uz$Se%w)>EBO*FPQwolO3OWhajx;G8b<%@+Ob@~&r|V1U~4L`a;#4DaX;c>u};0f zqkzZH>`gW7+LalJ4%vSNWf+2!*d7((<7XKA>I+XS*+TFJ91jG;}C|Nh;>f0^@Z9(MKF#&|rZ!A^N{ z>qr%5C<4i~Ur21OOW(B3j-Jq?jYRUW)f+w3TLL4gM8Qf^m00WBv@>`<&pk)ecwP+h z2JVYC26u64^|DNtPzaNtQ`P{NBXoFg)^&4#D5JvtN``4*@7`m>!bh9)s+p8200%TZ z+@IXo!|vK-`9gSVs>~33sbD4Q^$^@Jll#nR@&?dxs4V>SsN>nay1v@b?ZC@>HC$%zst$23&mIMt8HZ#S>o#UDkxA($0TGG<)Ug4n{O{zaWS+PpNC8P9 zT$NzW9m%S(p=eV`vn@WU7Z=i+>dke*77uR(TGC_qh){HN7r4-GKyX@D+V2?cmBSJZ zYuAB~ySX2l+*Gcf|JMI{D$|`Uk;VWyo@Eh_epH?W3`5I&Oh(9gNs-L!8h*=d-`^|= zR?*6?us+=iM`nE90|@iFB^;t=$9~*^mls=e)#9opbz6ndwEzzYuod5P3b$7Us1;r} z!&Ic(M9A%iY6rXRAVlbvGE(uhP`x#e?M@+9hdhEIPa!8WC_^fdcC3v9Z3uA@NGiZh z9Mk_PwMz0L=>O>ST3KH=E~l8|wiPx{;&46%Plnpf^>W!a3nQ3Z2R^92yK(4H+217P zxkQdMdz-+F5QZhFmw|@I?6(KLI{VPOe|vZ{l$!$Kz9(R?9@lY(vsq{$PV+V-z|nL< zt5JysH2I|ewZdz@wh=b~5yp=Y)@!wrwb46}3Y$2_TD^r~|H|*Eto<-|ML*Y`E=QadD~*owzUjv@%e4~DbKHHxtQORqi@KM14t-EXiae)BVt z9Y~S03j9pxZQa?BMBJkY3|UOCte zczl%W#t|7ei3wCxM%92mwA=c!3l9LaoredQ`0DIO~os6@Ai z5gHEl0r_8f<2kyW*VC!vndn|Ev|H>6=oAj-5}#i*WQ;)+22?EyKy6z9`D)`Nto;hc zeVv#Br1Tadxuz`lQPsx}x)k=N(zBgFr-nX8?8qyk!x&=Exnma>O2!nY>3T9HzmjkK zXmNQxvX9JjDWh2dO$%mOmm#Jw*U?gMn-R1J#BFgBaLQEI$ofWrAzvOou0$-2DB%qA zgsLso?maDc{+X5$q^EQ?A-v^m8!Y-6;X~v^s_IXFtE!1?ru%%CnJ(D%+CV@_zi=W!EyJo%(>72=6hu0tL`pzfL_m-R>5vvF0g(pjTr`5Tq~rqW?k=UfySro2 z4d+Fj?Vg!E` zk|hMeTP!!EwfAB1tg)p&k%&dtdj@Z{%@bP3-}v5rBwug0EbXF3tBjYpWj}nghoMtN zTd)ZI(*O8!FiX;={F!u&dVjUe)PM1c3W z=HADatxhOR#;yhM$_tA6HMf0`Vn6Am^?cOMW0f8k<}4$`=aeVQrB6DKIAm-wPy0jChgSp z#C7lXzc+-FjA4EdLhMFAjGGpAr4{bk%fn@jd`36yaveJdO;ArWZMK@&*Gl5OIuBWV z2T1PMy*R6c6uQQUJjhasP@kw+{NRh(9xmgSbto-;@WUxi1UBxS+{`cJ;`VB&!CUA6 z7J|rL@7^ytfg$ltsWHl)w2Dd)cFK)Cu)*#$8tf_7w@6D|tDY`qRbf53z+R@*Z8DG# zQHj3DkB{+Od{BcK*Xk}5J;$i-AzS12QiuCqmyOA|AhlddsQzZ4^A$vbSSdHzo|ngI zBYt16%Zii2L!;^)J_>;r=E=rXd6)rG{%)aecbfiuuKsZz>DX$E?R-k7{n?^2u#pEi0uX1FLqP(Fd6ZIS4()LXOe_P9*7=PjgY%TUU}@Yg}~9*&W3d@!p9|#dG+5S|GDML*?OVi~Jf%Yg*Mn$@8 z1hW!x@>ze_V(E`7-z6N5d}4mb2|0cC%wCVDUgJS5mex21-}izz=QiwT`bK4)dnKG{_LEz>8YI~n^Rs<&!zcJEKn=qu%p+iWIpJHKX#ZpN(Jc}wb7vWOFR6>9|gb}Q`O z|0vh!+7FlqE_g$FAs9qTjQA8MyZ&3#`Tpm&0wfMtOF_>>F|GCj4lRAHv!h-C9He(d zn`rktq`$BV$HB0qjXK`V8eA8s-~iRe7DwC}&8@zqR%_-9_wjbC+@t3LjK_`_r~V@z z$Erh%N6PQh&_Jzs(#v+m_Kh|iF}e2lCc&l9Z= zrftFOiL)NW+~*N>>N`S#uS45-JhSGrmqU$ZKp$)}WJcFn;;r*eEMswF)rh*}7Y<+b zYJ16(hC3biB?((NfFK`5E!yZ2Kdeuc?1O0|e%*#8lN5ZLE8I$P`BgEU(n_u?pY9MrCgn>^Jz#A-aFt@T~4?^KynGY(o+ z^KXA(S2}d08EQR@Y5_fu1+~w6gIoR^`EORVfF6JOLGR9A>G7McTp;JalS#EGW{903 z#okA-PE_1X+8AblP18MH zR9l;`yW*hs)$mpdPmhJWB;l~kmg=lC$*{dWqg=%!dhDsGVmH%6?BNV!gt78H8}_&< ziz;C3pgrVHnxklIW6E$+^(-+NXPD1aycB8ld1L>sAE}2~r-PR#`?b*Jyk&~)NO&VM{@vz_ps{!Ke8PFqB>orP{>{;V7O-co;BFJD2|zJADR zNEFM;TCt&4AwNo_(vM#2=0_f3_;U#N1}hIXBx^596k5ru(Ju#9nfFI=58ko+o-1$U zqJe4~)o!J&OsB5HzKzYRQRa{@o$6|dyK&X0y(OC|-Ce^dHD8?|_0{~`*wB0$1;dQa z`;I(bKG^x^M(H>kf|!&NZ+GnS`WxmOf#e`kOQUyJdFLv=v?smZa5A*b8syc*sHJZ9-r1yd z3#RaqT`r+7YLsf8Eap4C3C=i3xFu>pbCaqmo(<+mo91%rCyprYZxD-+9uc>YhER%1 z)Zem~+g%VXk_*}axB{{;KfC>-7Z_G?s_c*BP0K{a)JV?>nIc#TDi2OlSv+9pa)oxV z!WRs-pxVt$XsOg-Du^bGk$(P#^R%!9?&afoBOYPYO>xjB%3@e+x!7G~*3TNLMB@B# zZ8*{WjmaaA<+=;6fWCd@SuLsTpxvI{`QW?5?V;2}w^7%$Nk3}j+uzDQTV%`OA3JG7 z#&s$)q&+SQLaB|Kx`jn?>LP@?DxR+u34dOdZz?ilYZe@`il=f}n%K2mtokK6{jQOh zmF1z4*Uu79A&_f?;DvdWS~J|o2Zx=eYpRjGK^i@<36)S!GxcW>LL$_vXzqh4cj|ZM z+EfAI%TR4f>G05|K%}5lXe;uLyYYR~zH;-;ln#S)E20dA>;k_^a*t(HY|`hox1^v? z`sRurXIIak0_(!UnR$sW34#z{Ei4zm z2UCP(wp*Y>I;_%{_0J)=KOP?i3BIy+V}Vi zvv~^mKG~IC0QecY$sRkH9DUeUz<}YD0H{c4w_e)(BXcGRS3@&oG*X67^xU>ZYxPiY z{Y`#lO-_m5i}%tnH4dBjzGSgCe)Cd7-BTC(2T{caxQ#wI$xGAE60(#FU+L*RRxA4; zPRZ z(0jAVF{l$iF=RjCT>EgATzIuo*&=N1ndSsuhRX6imQ8Q;xT`Ytnz`n=va`wpwa%!) z)x9yO?B%)Lr|JdM`}~Ki%r$anoK|y>DI|h;Q)m@)QtUT%HJ!7i4PJ#yM6T|y_V4#+ z3|h`D=ob?S?*osd*1M^)d_%ww`f}rgBI}(I%(}Ba2#ZSEt-FVHSBE?yB;Q@v-Yaxn z9pJJYSA2?UrlOS3d=vH>p7@yw*~Lz(=p(rlOX?`5Nh|3ZChdkK!MZ(d(4}3R-DR$6 z>Zj|WB!1Y_N(#bIUDmU`ryIuQhSLcdM^alcjZrx5VWDx7#9a0g5xG?MgX)J)cA5rJF$ETmqUeP6{aX>9NaSx)8pMzpgflcC@pQaVD)$ z;v~07DfL_f^+cffd3ZQo*o7uQYv)a}kIrMsj8OTMcV-N%Sg+0OKqKp?YI%$2kVD2d zJN+DC#=n9L1s$=(TO%YFWuT_y%%`(Aw_B-i3NwmkeJ=%39pBx!G5*_KFLVY6T3w8_jp~4QcZ42D2HLtoA8QT81(UD^hnXJ3%=iLJmj7 zvTSr_^Xveb!0Ft!fY`>VOdH6YuQfNxSHocMyInV()kv)a@tN;rW5J#9TyawDog6D4 ze#oxUXp^9F`er@yQK#E_1qx9C{`CWGQ)Cs$7>zG7X_XTeGS5}AFD}ls|8C2+MrwDA z5*z03M6JGjV1~012SsyqH+6040WoKj5i{tSYV_D_6Lr?85%OxZ83mCY(HNr_@9z=D z$E+=Y_=FGoEb)*xj5d^!Vp`X9vPKdC^+I&{vO>?Akc-}9xl_DeNa_`va3snh!}o0| z6`QT_Xodq2N!+5g=_cfGK*7ry(>O%bNRc;d`$9^jhItjf!DHF6p%yuAD~8w&DL3#s zNy6a15XAUyp)r^dy6pRM=4e;iKwgHe;*EUX=bMB^w@`il!rA$U`|}8ZHthg_(Vxgd zj|Hzi5+7X~da>#}HI>J02OoFB-=|7O~d&4Ao#T#eK|qTYWNJppw<#4Y8(a zNA_4zDzFd+aU+w*ON}KV_D4_JEd&v}QEa8uDv$1~w?QL|hpB#q0X<%+-MtP4ux@ft zi~7=<2EFYaiK{--iR{%)7B78RyYz0ou565ku%aFkHTZwT%$af6IYjCZ$;c5 z|NfE$R$&DX-5Y)^DoLjHe#-0Cbm50~Cl8nOG~ zXJ>3JYsjmrCG|?D5IFagwCLOW2cecYr183@XLJNsY@oxeX#Tn*fR1NNw_j8i;}R}Bja1n4K#zOD3DRw|)WiEVnyfCA z-^q@6XUo;McVay~UA0qtKoFIsm_E^;;5?8!UA-~u8;}gQqw9sGzk&o)3oBjO12*&c z{qw4$Zx~Trrun0kHwy!%Dc+RoJz`}EdrD)@(XFOpwlr~jcG1bT)6A~sWy5>K>-4&@ z`BV{6Tgwyy`BM*ARmyO#M*qy``PH#vl{QS2+%VeEl<-c1s4%-d_Qj(T?wjOBHH>+k zYm>#*qtrY=IOO^qh0bbIrt^O&yzL7gsGKFj$6?!2lt0nWZ~HKC^D-DGQ-65N{eAtO zkY|&|MIa3TR;t3;>z!2~F7QLnZ0*(PG1O2Bbz}~Mzwq)qhOA& zu_>L$UuG{FDOb>phL~tPP*}gZasD!v|A7ki=cn}Gv==`^NBEUjC#ITd*OI(nOshT{ zyE9=TiA*fRZ@U;?sIXNN1 z?p(>(7gq5`qjg-?G>hU9^r^1yAKMl(!gw;K+QQ5i$IA3A#!9b9^5ZXC__Sv^?v4B2 zXUz8La-Poje2OH!T~gPdDy`qtF}hzrmu&DN*~x5{s7AaQRlY_QpF+HvhTk?jPnCY) zrO5<3e}{~=UM_mJ;q9HfSd>WxI^uFyHPw&BmpUpG1)DFQGqBRazp-^T9Z#XKGMWyz zKOo|*{dCs+T?Fy!ct`t9jX4heeZxxYi;r2UvD>F8gz4g?aZj&6ybgJI738O-dOi`; z#cQMGYsJ_~OZ%HT$B}PK(Yb&Y3a(?Sm^I9j5pzDrU}d-rn(jK{NU_tm=}_PaNuDHROT`U*#vtkLvuNJ8Fj zcNs*MoKyupC0LN?VH^v6gjq=t}#S2##E2np13QsoQycj?*3A05BKV^XE~!l zCtGV0=#A94XAK~7<;N=scRWyJm-38<)!$TGCrUWURHnfE3^$DB(@i(mKzsVwIoXS8 zg@kl6^HNF^rM9pG<;>JjhdBcCtDD)%wV#}(wHfr}Pv6Tddc8!Fp(TC7G8WX>EuI-r zt#bhM_17&T6>7=O9MJxeu#KA?WH-rC%R{)XH^zD!PxT$Zw9Z|14KV4?$ z-&KJ^r#Z-y1^)2WTf%((cB#ltH=~Os@?>s@$z^AO5pjDOvx)#M=m}SzUG!ET!x~o| zt?_jMTr&ux>H$e&xxEER7Pq69ExS>yw;Q&P+zrJ?y=8BMdB_vtS9w;i@S@6S@eRnG z=hfE-qC9we9A}byYvk?!fkjHn$$jgWb&c+cD5#X5@)^&`9QN#>3-43(AU~RLj%qo8 z1sVnbZLTmsd@u9G4!+Sw;&bihfXx9BM$Zqchnp|<}Q(K$8UIa%m63fSY`gruICDd=Swr{b0eB1VZVcSW-%z#I=2-ARbhnwS? zS2>2pF4nigFTw+cy}g+pK>c+McWXz_v$&jgPtWvoOZ}U zWlvMlt!eWU<;*i5`WAo=C;BNJb-5_SW7N**`}_?@vZ$YUj$MLZD=$m#OxBowd#s!* zdm?Kt7bJmq{^2O@l9qFi$8lezinuwHFfis~%J2zAr82-GvPLjVfHTUw*V6<4SBMHf z%(5u~0LxqJ#YQCl6A;w90{&~^YdOY`!&AxL*mZ`@F9A3v$Ibg!g-W(Hr?PK*=^YLM z1jNv2v5@Zg8r{7CgqzZsFdV`K04EsY$2A(EUSyJOTK#1*s@VqOd}=*(Dtvh+{7uQo zGmaBrw|bj#kO{6~Fn!!v3nHAns|o^#r)c=(5v}Um&B*K0=~jfr$|KyC3-A?=gO(Mq zLGM@k>)G>OB!9*+&5;?s1B`mD|7c)-1NgTdM>Oh|{{Rk+qd=c*j1tm1Hc{h{83w}q z9%i-cHwQr%uJ?t)sv1a+9B$B%hZoiCS1Tqe)vc%ycNZ?Wnz!zlkGvVBvGUo5kE+i% zPJEv|eLO;neCy0I0l=%JSsiE{bCz2OWl-5#X?Y9cgD=qT=lw*%o4TLAZX8lt)kT<# zTE2bDRWp&Y&r@7!L3X}rFA>0p%K}ek@$Lh*?TKVbkh}L?PAG_J+$~J;Gix(+me*V{ z1n_3hGaSvINdMW8_@~F-j@Qu4mt5bFf33qeO}K6M*K!#35j4tf$Ijj}M3=VlWQeaR z&`+QGsnIsfL(#15W)BdL(NLu5kdaOywlYy|M1N zNfLd&pB%A~kcc&Q_$=d56Px1&_=2 zSR@;d_PvBy1Ov+==rLt4_XJZdR^0jr3+=SS1dmW&JBjouh? zw$9QFIZUcsv>B{^bN~(UfdUWH7q-8s==j-xCddFxQSt~Lp?}z-sffH;tJnDvUA&8@I{$mJm*F$j4 zW5Pxhw1A^?(`9tl)YMd0rgm`E4eX#Zf*pVpwr+#C=NkaeY3%+r_!<)%%Mk9SDF&=4 zyQdoq>=V5H^;vyFH6wWqN7+R{{ZH5QpB;kXYou-hwxVCKyRIUnH}c;ny={&sLC?Nh z1vyjnp4I5hRLWi5bT*xQW7vVmL=EsK7nX6ow3_@@6*VwM@k&_8^tI`(fA0q>c5}=P zSzs#!pnqsqTsiXs_mYp`dj1<&uu!lXCEWxp|4~lffRDKbqqeAeMfhvU$il9{sC|gt z-Fb_s{s0bFA`Q~rc0R9_c`Lnr?o3B&8;C=DztM-tSIgH>9yu>tW+z9ptM{`DJt20i zk(6bX&1)<~!ACb$O89xuexqYU-~xc?etV>xi10TnbOuCsX*c3O2K`O<`h+roGo^ojYl`q%v$2&D-1}MD?2UR2{f%maVF)${#sYxxYpjKz5dY1f z{^G4XX%GfI^Lusc*UCA51S`jFLHh&){f4Wa;EfU%f=LOw+yD6}{v#Ihd%QE%q&G;< z9V{~QU;X$&*A?CyTAeE&aP#kg=NbST{)e#pOQhR>K4yO&@BbF`|33?QdbKf$V2eFS z;I$JN%)7#}uMU5cT2>HV{uM88l#frJpuQL*nT!t!5e7yOrGkA#W;~o*pCB)9(&K6Y z7hbqWMGjpJ`0v@Gjo>QvtM*i%Zgxa6*Nz%hwK~@<^8a9@!F0c&+1c;`xBLz~y0sQ`6S|UG$d%rT#o%;(R|px z15}(mPfMfj{oukBL5A~S!KznVAOPe-HFsfC1k+oH^FIlQ{!bE;3BZYJV=8#>=MT^M zgZbH7(BdJ&AwD9zIi-Tt+qoe7U(IjkGoV(!YMC%#z`mX!2@@iq9znMGYzWA3jw!!6 z7&=yx`yu>C#*lw{;{ziv`cyRr{m&ovW&`t^@63Kn0{Bq8;cZPN1=&Kx|Eu}A3JbZA zoReE*mcP25plbsOSaha|mpM0Y!p}wD*v{VB40!VJzgl$OZ`C*WY>$L~QRj&@t@ekU=SaM%YgL3HlfsTV>4*IZ06M5Hg8l9%o) z7*!{#PvH=4CBHeE9*o^x`N`(;PfPSyr|lKoLJjBHQxe2&Woi4LEJ6S0{{QRV!V17F z8E^Ifrj`FMS)yEie9lD|Fnu}5umfYN8G{hQ@~N$2#h9HukMfZi$Y zAc^h2+tPnq9Y$~)j-Ec5{Qt(1ifF6oLQP2?zO@4hQH?Bd^k1E#p9}uy z-HF5b-`4+c>;K=Ly6a2uzgz#mTmN73q5WCl06FzPf&W|JyFSMPWIi8NYn{|gK7MS= zRw-SWnshK@4Ikk&3r-SkXl@pkO%WkYmrcgx;i(j)_UO+^zAXE8`Xnt0^$T z)AxSU?&G=TRdvXsg8p=ZJC66!Q8!JxTxtN+T_LgW=CK8CZp+|IknIbhJ0Jh$T>tY^ z*bE%N{|z@%ev0nx{DD8X-rw`kjtZbhC-Za}H zLdg152wkHd=F0cod}-pPI5lK!AedB*U`IiLFE}mRW_uD#=1TkTkAuh!xHU9R0Qkpr zS%sD2L#kFzkj0I0?8KA=z?Map_VqEpc|H;R(-L8R1gwr8W>7qepG)-ddP`%tA!DyE zm!kEb>h+Mk`-+Z;h!~`p{f3a$i&2y4iv{5%FNq19yWPqPwKg&bJe1FK#uUpfu-~Ds z_2+H;xBK*SjHdSfII!^^w8Yw$BUmtXBs|* z&^&>taBobi+-G?6zewP7;q$ ze5%`ZenaUB<`whnA{f9%3*QI7fS)CP820dQ4&Mz|_(wq!w`aT6%ihU*9(BjaoHM6N z^Cs%1{b|!17{R9RtUEm+__@ z;P*`8b%hgxajBFT1yjpW(w^tgW~NF+_91>>*WOQZ-{L?fCLo}x2!?$4kf&O1{*pnh z;?Z(9G$Mpbx^1@JYZ1_Zg@H^a{-zfBgR2RPZcqa%*TCkPyU-TdAEMg3wZ8JH5Yur! zWuZMx(lMAS@X+kp?}h)XqbFA*q>I$TL(-sAGvGw z=WDj;u8Ba2R97A?tftboms>19O!Czup5NYPJlS0gA$#eA7dCB$K9zMN_b6H9Gk-9p zB*EsTr2vs7+P!-rzPOA8ICGsORdH8)-8@2#$IQB2c$-=SX2=BV=vOI{(G;S6kAgMS zG*l(duDBRV6>yHpZkrPxF*7uMWB=O=3O}dFdsV##s`!Fc6}%V~ADd2~ek}Zx{}pLpbu|aF^TnwZJLI%zW2_jI zt3;BIml*H889Y12^k9tPiJ%GDW>Mt0CI|l0hX2bk@$QCO>o&c5z@_6CMB?aJBjfj5 zcaYRvOPOsO?;X+~VkaKH3d3Xg<`FivEZ+AcgTV|i8il0S`C39a$Ka$kWjMTk{YACH zQuJ_pnlYNqG#u#DmT-ENg|TA84D(a9K!1uU_}RU55>Bi?&;4Jc11{glX<^+=Zm__O z&gHt_$AXg*1Rb3G0uL95Tz7VSECSH`UMjf)5gUz7_xjcO`T1N}AFwM05oT&n*Jzci?Ta69Sc>oq z2oT%*0pS!Cg!-R%)VdFnjw0I+t@iLNppZ$`p-u92#sXGwXij!@BA@E` z(5aTavfiG`uCz=0L~<^$WA@~4cg4jEgfkrwgH3)9{h>Y(9)bIt%X2?=3Dd`g$_>T= zQ;+ApX}4KCoCG>gWOG#6$#0mVeQjn#BZ|9P&#hfJ+pcA+gE<^-*rm0>;AiK@iQaSm zagY9D2;UN^dC@O>tF~2H{&@Fa{27?fxB5G?Ky=cd>@7zCQ%=W`)VZ3F#yG@}V7`S4Eii$89W ztFY{KxYBOCT<`idR}V_q6Tahjt-+$7$zJ-=zyKi&7C4oOn2!9+WtOYKfHP8{=HgK0 z1PH~K2n*UA7#@M?c$*l!Gn%(^ygs&qEu^5^2=_=o0vs1xEt3Gov}7EQ82U^fvzG_W zF@#D9jsL_Osv#yCMyAP$jx+DOc;k=3{PX1|xrqo)=nb|{q`ytM28U}sc6pzP_ve=c zA5tPh*)87dPn6~FZ;mUOFSL+yb93JfmiH3QRV^nI@Inm(%h(Rw6m3T$_*rd}Ige&(WW5dChAliKa+ZEjw)d<&6+ccK)M8S=nLHrSE5X-D}V#&yGW1Q1mdI}F9# z{bRr+JAf)(y6PyL_D2v@L=X;*z(YFfS^6`hkylY{rVNbL`@u3O&w++Teai9HX{Cqs ze9rR{ecf3^j0zSjN_kdt-qorQ~A`=0*Z%vpp5&8{xiS0JB>EKnz`& ziy-q%dUi<>v5-Mi>0}zPDGCzn_O-S6_9_?m!U)%&-C=Hq$A~(XK%uH&~AR z2=K1r&=FB8DN?)Cf+}t5iMvEY#-ss{VIn`QCT?Yl7BYc|5)yCt9rJ;C{R)phQ06Rh zLZX>XUH|b4&8UHSP8}rM{BbZ`g$2OvNi>6&i~X27JOg?^nAd5}t1<_;98e*+6OtHrq zM^zqAJ<}0F;TSc-f(U2P?(2v|G^L?JUo~|<|E;io3(6l-;xR8eH|Z2hliweb27MZc z)aKv0sSl*rOM#k?%c!o}6iCcSAr|_Cf0Lnth4o^$o$;x56LMru9qdpPVblue*l4r> zpU;*kZ8&T;@F5q?+0k|c@c;1@CKxqp=!SFDu!sVHb|kWw4x#ja*Am2cesSS{1Je8T zW%AqaZbH$(wrLCGxqJ-hn}P`0OrCqp9Gnoc7(Aj^DS4JLGTyAMF{H3)g;ygU!BAj4 z&f2T6naB*}Rsg0k$Qn_?xhmzfEyf!*kt#n;jzG>pa^#QANn3;AGTz3o9|nvnx~-wF z7(LDYWTE|&Slv+v_D)spOJwxl;^1FI7g!2MP`kOfSvt*nx!MNf7Xe ztH#P1NJHww+(Qhi!$ohQTNL&@zk&O_Vz+7SNrf2Y`KQR z=FuK1XI#lkUoNfHoas%=Hx1WZJBfy`4G>X?hDV@p8h>NA@Hm%;)UZt*a-k{^#-meTt7 zHUH;FX!JmlCEQ*8vqeUV*zNoCxhp&gdI*W*SdsqXVuVJxi^~n_SMe?wB7tj|5p>FA zz?Kfhp;Z(dq#HU^$W-h^tSbfQwasFx2BOjhr0$8kK=w+|CUI8y&^kvjmdUBUr66UJ?${KJZRXF&0b?3vNxh{h%jpyI^b@xW4LoiY|x-v z*^65)xj-Urzqt7FF`cTM!c2F%0nENFjLYzA8vM57w586moNhhwleTb*kR-{?$Jy4X_`i_D(2VDFb#jOwk1Ha!ShnEzp2b7i#J!5LJ8W zBkz>KFM*m$|A*!ChwuRR2?H#kFp1`^KfTFxAE;UFr=SV{`eKnB-qynOFztzVEixLF z-RI=!k7630&?9EojGvjcHRSJQzXP*vb zb5-w!`>uu1S?WD^&e3H^BtPGI-W-(7vu&37ro*m>%`J& zr*XV%cb7O@4zz#rzweBqxP^cks#0o#AGSsQ;z@}+EQcQ!$(&LNbzuvE*~jrxEzcaU z5i5#LaoS~Nj@DeD=?^C)7EdKWUg%F|Bo~`7JU6+U@yg@gJHMCck4HcW0Ne2EOO-z~`%QRN;H21W)+zp4zA#=(H{g(1 zd_=|v8i@33W_l_JoxB;&Gn*GmBs*uDcErj)oH7|GGt*A|3aPd~_-d8$1ZW)~_A$}X zuWcCMD10Wui{`UcBDGb!J2e;@lvy`jX(J8VrYX;kh=Z1)4vM==o$DVyUHh9u#$#{B zr?yA)!3p+scCz@+_K+#dvw2sqxOf%&C30770ptLM0@DNozdpIO^+qG-ddjT+i6#w znCoCc7h178M(|hM?{%mYsXD!>{Js)z(QK0}|7nFw^=L-d@4N*8rEVPxknXa=blR_Q z6D-_d7$PJ8qzZq@MRWwPobxwg<$fFPk2rws^l2;HobShHKFN@9@LfCFVj!Hq7$VL6 z`AxAwxlYf`{BfEAcAYlt#IL8o+n#ucMTqjy`P@u@C@T_BWhnp}ThfZQ>Po;X;uHeN z9^24(&PjlQ;9F#8bbD6Kb0%91wzRUZLooyX*C~s}^01-8c0O!=gh=cxL{v#${()M= zE$6m))PPekmv{;&>2fN>KNLL)*?)Fw{ZDc{4*TDt%|Z=Zp*Z7QoQ#AOuy(1Icg zW%Ym88Oq(%Y{6THC9=YI{OTZRB8>@a0oFTz-^pJ$v*n|tia&xZVx0;d-9Kx}pNrCv z0B-2>hUI}@YXsKq6Q~q%{8}ES3QAubZz*(c$m#y>Qnabq`cU>_D;3nT-Ar8J zgv)9z6eC8X;N2&X>6zSQ%J2OKLO=$?IrQM;pMh2Fi+WrcjS5Tx5Vi{MO%xV?`}S>8 zr#3m_HVu3(?f2+%8>KxFga^JathYzWug>>!uR|N=@486?dDK;ruv<+1-VW#RWQ(1F zxPr{tmeL(Cj!=7|Sf;U6MhfvTY#yub0O)hl<*v98HRns}M4qdrxIFi2+64nC^Z(JRLEnE^QMwiE5licL0u~(0m*ot&|3U6$kOB z3kKo9hd2 zMIr88gDWF>>IzL14r~SFSPRqjUKH9*`)vt9*z)&B3r>Ws^gJNKQgIku0ji^gJ{m~V zh-Zmo=t_LYq~uL)wZZ}T0%u0N#0QX2iVV$)2^;cf5>m0%Z>tXPEp*Km_*eQbGpdH( zTML0VnXaHe*F zy4sWv-ty+eu={bx5EoTs)O1V7a;I+X5FMPhP^FWdoXl@t#?qTk*CaMUq!?5Yv5AQ_ ztk9T+?T8Pkk|!#xgo7W6&wW5SJq468B%|krY1;zMRy-l>P%#h>8Wb?M(b@3{E6|)* z)V^}ao({%rte-xTOWC)wi*e%VIXbQ)n2&u?iZb2Og)y6CAn*ZZ#ZqEnez3;EQj&!_ z&s=L`zSzca>-5ml%GSg9^q_A!q}y+^#-M{}tSB@%kcX0};^6a>gE81UboD?r=Rlp7 zAjREU>q6IaMEg_9z5<4WcGJJm_$hARs*8a7v^j&)K#|i0~ZJd!)|u z!9wVwqdWdm<(c$W%;$j4)mOc_gUoBolP?eQBtP!2KC;Pq&?FF4q4q9Udzt%TLvK0S z(7)?x*D_3C&?xQ3;Qm%TOdvBQEnyR4_@lT?qzJC}Zr$10S-Z>4TYl(v+&EEIY^&e) z2GZqpx?RRHFT%aAw3L%ioH5yd(9Mx*F$P3U78)jOyJbif}kEvbByrc@f> zhsWJ9vPk3Nv~g^P|J2gfLx5SQ?a|&+m!f9n1nSwCjjhJeRl)bMlKDVm4E}kJ*Gd&e zcA1^b3N-Btq8cjeIZYT39*|uO94U%;&)1Uy2^4(u;#5@4iYPA1(gl)Mv7|%eK`{Ev zwue`LA}yhPlH&%uR>R{+Zj12t3w!P9K>K!;H96dl)w4mvB$d*KOBwCuHA9=@bv!Km zFG6#VLkG9Ym*YZq=C^n@-fq3Far8tcW*@UHtDATaH9T`#!=f`^wni>k_rlMR&o*I! zq)Qpw)oLTy6|b&iY`K~M;}BPWupk+uuLSkfR1;wF@A54(Fa-8XT25bZpS}H zT7l#aRn)wWwgfZkvf6Z#Z?M%@<*Q*#Ve#2Qn40Bz&%=@Om*vrJCi#39xG`R`9r{vS zsg>g3n^IPcH(|sW?e1QI5Cm4Y;V-5Zc3Ld&Sf%fM2B|eSY<^YQcFNx zxJ(}`sZuJulE_y*CR=Ka;x%J^#C_}0;}ruB?oJddgTWw56!oV{{wE+q`tJK&7)C(< zQJG10)S3p0*WqCp`hn;z<{Gv5haX0br|DV;sBoW|yjD`Tt%)ky9#GjzJ1RlATZc=n z$~#s&&fk}ea`o`&9XhwKJ&E>GjHmgD4{CvzwEe^>sZYRmaX4UWj*KQ+I;~+)-l;w> z+KNCt5=7X~%sgruj;h-Dpkaw{%I9xVeMSgTur2T_f-D`WdB|S=-f4oTe&JHEE=}KrjT1mTWCwUsNhgM`sGL9FHK@u^GV% z=;Zj;T2dxx;IB3}@Pic;KCw}m!p~}$o@R7J*YhtXChgp7a%zYVm@*FS$q{*ed8K>$pSO-tcb@IN>2 zVYyBnA7{38GR1b>$lyZEbzjs~z%e3LNao4R7}(0-FU{Or$>Yp>$5`z3z`^A>@YDzCIn;A~*4HnRJR(5ncg81yMjfoTLPLvhY?7#K&Vz||v z;#u@1%)4-_`8}yDm+Y>U<2%;5x;>kzko``9wY0Riox}26rd#UiyMgCkK~>eUs1`@t z{gaxG0AqvmYNOou-9=L;kUn4!eYD!l}y_uW6s zi*EPgOzDoJsx>*CG+Hy?qFKptU!LUNFfC$K$|=e)pPpFE2h3?NxxS!Z?(JPzXD4-F zQDj?#d!l&NVZJ;xiHMda$cE^=uwsVKy_EU5mghipTK}>Fc=B#mP8PG{cD_2BA0?Lz zJnT>p@?u1>3Riz9s&sdIBq6rAm5|Ri)1o-O z+393A8F8@D%JUQehLoYmj-Qzqa~d6sL~?7V6pJ67SuTprMbH$VpYM$!-aMg?W)%aS zzS)QKR8q2NfqsQU2X3f#N?EVfoz9FcJm$SnE+o`*G^7=@rl|XCB1uaJZ#}9t8qp)l z-Gth&QEPK$+vDl9=(_vq*x<|#=cq6eO>e(DNW`_&P=MMAd6ss!Kr4@)Z}jK1M=*rT zr=R<6cJbLdXkUy~IUGrOa2E|$K!?4zofBXngCmmMrtn~EP^EJ!=xK6$ORskGKs{0B z!b-2U;_zgJb-2xVm3DUJo97nXulP+S9@$%|%e(HlnG4D|I1z6`a*dBk3QT6WL{b;M1HLi)_vBMB#ej1u#pg-q z&5%x&t2}iZ#x#c^d4-Q`R3MMr$$fApa~mgv@ij^9gv=HWZ79pbpr=Q|ti+rf7V8eL zoShXNQVOtJpi0%!o~=#OC23yh)_7iPhA~g9ztlTNz_wKo(lgt@_em&T}LqTuC8qN{I=>~`Br}Hab+=B9)aSJ zF8@E`j6ttKl#wd@MXS{h7af#B*?_deN02AT0*TFf#Puk=3i~CHYWGvHQ0~4uStbU> zJ)is@u*+IL=H`~dmk*mMow65yQ(~kqUp_pnnR~||Cwe$lpbvaS(RCKO2sAhYcfjIRxbozK?~*B@rk0w5bMM#2ryqKBQH{S6da$95RoN95 zne-2-;w-}|%Ig)-YaB`$`YIH}$$3-VI|aN@B1SEiqP$uP2Qynec8=Vqr;zH_4KE${ zAGc<5Tk!>IgMtav;Ut9psmg7UVTRc3{683sg+R1=6-klp53@wc zR+lMTiBuQi^fgrC)c)&fk#CXEnen(&#Qp3C?(0`)CZ5>9K^eI^+a6E0YaePVxa`Xp za5xwC=klc z$j#kK0nHeCFqlMKv^tU(A$Qdw9khq?c|q zs^h{6#x_n@#3zk{>>psniIAGA@&#=L$au^KI8F_Nkmc6c_6d5|2nO+O7Z}2eeIcd6 z$}28M`HyJ^i}r2NY^jdBnl{?X&Va!=b?_EB)RCi)Lt|P!);}n}n2)|gEw3SCUtNTY zB;LQH)oJ@+-T0gE8lS7g(X_Z_C6k7vV#54I6RAJpP*{Yz;}qw_|3lcDheO%-f8$px zsf?~JDPokQ2qF8HgjC2f_NA=Z%bI0O$dW?#EhPJ3#=bM6tf6dUAIdtGv5jTMjOQ~~ zUH5%|pZoiJe*blJ)ZuiVpXL32ZSTb1-;kaD>iD|J^=wnKZB^V)^@hWRbPT453jYCL zRC2>>hB87Fx-rek!Ikl@bjU)ThaDex7_+;Csh(T9FvZB;$oFKkIAW8T>kp^XkNB50 zbIh(8$uhSm7?HMD?QSdy_F#dc&gVK+|A{@uv>3Cx<&Ab?&t5=4F1?VVM`N75-B%V$ z?Hc4fEo1x$_%Y|uGrL&Gy9I*myhcu<3_Jpv2;~UIrq@1t)SygsBX2v}V)Ls8iA!iP z9drtEsMhowE*%PCb!re_aNGu)TnbN3{5PBOPer|bbkwc-z(pA zp3}|8XdG*yen~Susm_kD$XQ>&dMx@MHad06&WgBReS%qNWT3JK!5Pya{22r5D2_wQXjAPdc!b< zv2mZMT&e5J?x$-dl{@Ejg+^r-!J{U683~^8?n^@kD`M1AU7#m3FZ8)gPuPDMP0S#|?VQ`%-r8#s zIhvfJR@AsuhQMH@bjM+}rn7Q|-2ij-wBspb!HBGX%*cTUcBv7cM$YIw-cW}#5V8Bd=o3ZWnbml^tdLA z1OH9^Z0k7UpptGH=bia_+Rn|I?D7b<45gZVq2{LZ3Kpeco{aJR5TM6sP+V-vz0WIX za1XTHF~K3Q(=S}DNXuRuOF6jkLF%CiZqJ@IY(&Ekco9pQsN9K9GOcpyVy-BM(b9?G zztRKNdeY4wLtU#N-vWY_|6hq^oShz zo`0}yt9VNF^^=9dTBa6a$1i(G6x5h{-pTasbBw6 zrkqykH~Q{n0W7%>gFt-$r0p4l>XQW-eq6WwRz^iVyMJ0@Ks&4hRFS0Z6+XfGwSX}p z8mxU){rxonPsL%Vq%JETP?7*zz@2-SF3HVWt}?eb_*9fZ89`4kUDzvtVI21Q=G!B5 zkozsl)37hZ9f$avrIemGa|xuee4v>eX4zocds=a)VXsZW_7yfo1E zAYM%Gi)1Xi&5PXQXpbof0?M^}sfL2JqgD?6^WUZsCjPQB-l5AQ4Q-cvaka3Miz8h1 z$q|fkU~fpO>BMam6ol>d#4sXo^%v|MhZGR58+V!fzW?CfSm&X13jnRa@Fz z5RipjHyILoQSZ4U;4T=0-|5!#<(naXA$i{x zF}%LW+}`6`XATOdA#&&0%y+qKk8d-%w;UH#Z7>x52*2v1w9*!KE~=&Nm_-v)yuW z{P_(!n0xB0Ub@(a+;)z>Z)7*a#;$W1 zHkmLr7Bjd5V5y6yP3TA#Eg0LRke1NG9?SYPk=17`5H^LHD8pW_V$taXNLbg3sEID@ zxqFb0$)X;Bs!9*1tX?vzOAH z+@(Hd-)&k^|Kc^zSL>)A`4d2^`l6*oi#Sqd1LK~W1s9_(=<(Nh+oRkcLz=~%M@BO! zkOQOZq(e!G57X$IHFmvAy`A^^q#Q9Chi8a6sFsiJ_@RVMm6ju;q?e$a%Ey0B;yOBQ zk415(N7kB!k1hMwAdeqpAQ_z1w>j%4hw>kY@2$6`f1NlpSBoPN@kqMcyefH((2SDN z$5e&pvkJQ}A$9kt^0hvh2oUH0k24s3oC^{|{?!^aL@8OoxzqgQ?qle-h( zRfdgZ9IFK}9LuU2Zj&w)5B3l$b@cvgbER&@!W#J7ZMzNKBNdO2_hm9P`(z&UG{{B0 zn_{xK2Kax={-WZsA=J*K*8^TTgCYGYx@n&L``fC6NB-*R2SVR$`1~5y==p5FiR1_m zXphMD>QI&ui4_Z9W!Sv)o|!u}^7yH!&3OUVTsOZLEpi~J^umPv#m5d|8sHxgT zlWi+3d!x9T4HOj>VaHa;gRx*6Kn;{OhZ&nA*+mQ+sCLTFzO&VO5U$JbKf^?KmuA12 zDfu|8>W52uyd5Z9udi@2dpcMf?l4l@Jz7o*fg$g>q|qzvCg2C+HMt0Se&P@s7X4}C zlVWgZz1MVx-KNBRO`}Z=kJ>Xb(|1tA&4L+9v>Gnqlli<~Dt3i=8bfud?h4z2`pM`E z95)qar10O(a)czW{eH9nqlF{IQPo(GN9(!m;pl$D@(%tvo!Wo#L{VcoC1;Y(b5R@V?- zQ*XOT5O&mkQL*kV*l_LiQjq+uxNm`L+t@I#qSZMy*1;GlK^dD{WpvXkhs=GT5VluU zI13+)CjXKdb!%a~=V@Q#1LyCTyvT~sD0X|l!U~?|F&ep8obp8Z8KOCH4Y?GRiuEFU zFi)lB1&n*!4V9P3=rx470w^Ucxh3jJ@nt?tC3%^VW*IwEKjeOZJlVe|0r~_)9PWYLIT>Aa!Hm`_GLD zd_^>uideqUl9y`IWu@`c!6Luahqvp`71K?wDC}Y&`&@w`k)o|YGvMTWNny8|IKTXB zncsat6ITSEX1m{;CY#Cp3fBm)dzdN(J9c;eHRHRsw(Fpe*;p>?Wu5ku^%P%RN|g1f zJo5@yt6zBkPrGakHTfRnLcu&j^E9)1{+k8W#lzk;u!MK zz@1W2o@kD~QYBaWiN;6S_%hxt4i3`LSd{~n#+J{QfNv^bqDjh4_tS-KSG8Y-Igb0O zmxXUGjgC8|L$#SB-CsyJt0UHl>m`J+q0$SJ4*}-zmTfPo%R(6;sk%*pHx1ha-<5}o z8&FAvOLX&g1Z$pcg30Zn)hyn&n;#ZBnOWA)Wv>0-Bt2Z`F zHTpi)*XxRX{8!}yz~zkbHDi5I$d+6t2?(*$d*zBAPQ*airZk}U(2a)U#KI$T;{mcx z9>w6xj3_K#aD0=1aAx-JXA$+Wd$BfA-&8fZXekipde4+_lYFb7-6-5(Ee}*ekuH4~ zehCjJ&wwLPP$%!*K!Kiy9BF~83KmYJ-vRP%SJ;$i^Q#jAiW!qNVTnX!fx*%<1n_KT z0Xvfu1hgupzMI|R!uCTqXx<2uO2B8DrZ>UP%f&FZv^{fSfByCfkB~#7`K)Ijxh1kN z6RB6gYZgw{g5|5*Irg7~%Q3vO>S}3m@Sc#oxxbevXxTTD@0}ndNmmSHz&DD<%0)dv z=I_cZC1Rz>-Sv~?%hlImYxcSa8XGe`j%m$_071iO5oXDs3_TTgflH}lK2up@D?`s| z{8XHa#w0p6-6(Zl#l4X(HU5UlRrnFP5=QLwx~XiP^wGzjd2}+#4TH^g7Lev0N_F}T zgVKRfafbz)4Ht^{P;JdKkftx0c{q%eP%mXn{(LsO&=pWTUGdksa6L)39CqEQ(5o*1 z%Wk<}E%=twgDz+dXX}{Kpwn+lJ}LFhAivZ?bQW_k>bTtKwpgl@Z#dz=3o$%1iEpx1 zz#KmZ7#2VsdJLC6q4a7tBGM$G*Di@Wg<%+gH0$U5GY2!Zgv`_G_nHvXg~k3zA7t3w5q8wr#C?aO(<|vZAE3J5 zApD;#lf^cTM#^U)KbbZ&AbM?tNpjU&cKI*E#GBKOWrXTN`hT@1v9ub{kN7U8KsBUY ze(%=)e1zYN$DpYWx7I_O#L#%^u1IG78?tqG_YIJqe$vh*g%iacPv=3835bK4dEzw0 zRMu?8)@ze^wY6s13O=8lpGZ<8^$r}75(BneafEgH z*MC?1|6DQtyPWS51J=X`acZM(rv%rSk{rWg8?`u{P4mfGYS)eM>+Q*2aIzZFw%t)` z9BLoesd1kywsS!JK|J*A@NY&tRPXQ`*rCd7-nCVyQo4!mN{DOW2;F7~3y5D@3*P$G z_`k#+1Igft-M0h2162YiAU-@8GoP++8*QxC zIbJDam|UbgY2BN|_mk(L+xaOV`&`Me^=SiTK7>t9E2p;_I?f@yt4+MAbcAecxixQL zxe;qSBnArFiD#p+=xiW08WQq;@@=ZuvZrdZ&@n2gqUgBOx%Yv~W!CBm8yfMh5_SU% zLc&&fO37H@T7fQHP}ZARIm*JT^XQyFB9Gs7ZNo9QAktOzE%!WgjANIxa)sAKyCglj zV9a@9KgC<(a?)UkX(%r}0(f{}BUGi3ZCsH)FX;(8`4(WsiHoQEi_`?`A8EHnTyUfT za2QG?yNgmoaJcF*YPwIQ_p7C6sE_hC+o!JURThd~oeyf_)0+ikICFqTQS;h`_nvho zpmsM2dhA}Yx1HKT$Y`*uZy5FIo(BdJv_xYL@%74JO=>|axD_C)2N{-|y?Bp86zh9P z{bjvC#VZ!1@*f+rJs$5G%-wR{UYIRA1iim8)8WY9%yKZkS#`s>)_4_3CxUO#*G@wj zE8H1`a!PreM%?2u=*tpzd-T%s>pQvT5~;t&$v+ib_?9b62(HAb+lIucmuY``Gx<$^ zoZaXjJ>6PyOE)gm)O+0vNHQJ(#n*`%4s(a0!KoNtL0K;;dR)#IHZXX8Lm4;6;`2xn z`KZ#*#EeAn5lSU}ro+Tue=9IV*+}`Cg!!!R*XYgcRc32ZW*@S*+aY4v*0UC3N&8Wa z$amDAy_oY@J;1W#`x%YRHvc3(s5~g5c0AQtZm3st;Qa+x5#+T$4J5<^dNKh-B_VX$#WmH9|H^_LBxf8b+P| zuFv_~y#m(yNoz>6d|~ElUUHI^=2r<7&>9 z;L|a{YSAu#uy290;`iaAM@vWEIE8j4SW{(@HKI-Wrsu^RUOh~g7vy(_J%jcPqh1dN2{B$KoK$knqr`s6Ah?Tz% zJS)s5?(mtsUiQ{MQ-SyA)z0dmc92`T7KHw22Lyj9I={d1`NBL9SZvI(IZHUa>F7M= z4h=?bD-xgU|1sJhcpLwr0;<%_AOy^qzvO; zjq4d4B}_K(ao>-7E}>F|O3nV%MdZ$85CQ7v^W=}9hCNJS_lw&<&xf^Z8|!U_?hPLl*V0e1=1h(Sgs;@vRiijiJY?A_y3p{{19*x23(b0c&Be+5B6!se zvQvfc4o0-K@;-A{M;(8P| zrKd<3EXw?g-}u|~Znu$NLcRz=K5G_SkAM>B1T#I}Zf@&ZT>Eh=I2Ez_H>DNc#qkQu zChkZz$?2|@SEf*m=3Ylf^TO_rAlA%AO6;qMjgLLd5$rzA-%Nu5oJVtbtsO=+_FnQf!yLmf>kZGbN`UU}sX>5?HS?PiuIJ_|g(ysxz$>+Qz-jUee@9Cg zr}*9cX#RzvHSid)@syhO*G6%)WW7FnakS1*(q-eV6csh2_2ikn_>*VzKUWl9Gziw= zGx7ZzuhHI3kLOo0mN*<;p`zSVobvqyUZiARt_@3kO@t~`j>VVLna zD2`jz3xU!3iH#!h@5P5`v9G_;-ETaAZ7WW~XXZwuvoIX4rHjnsd zDRf#2H4b&vB+Z>`#^65*z`md}4uoOdQE zrA7@b5Q3;CLX30Y1>tgWScR@>uBrG7uHEnpY{JQmG=Y;!(t%Kj4`@)}_P8}8no zq;GcKbu9b(zN}p$DAL&_pQNL7-JtZ9v0l1*u?F1P8Sqmw4MLub%}e2649zW5DC22I zJ`$x_(2AFM;_5Vn265Gnv`CGrf4Tf0?y%qcmnoOFb#X=gK3XU@)h@k*N&C&A72)Td z4)#W=Ea@@UYHS8e+d1KDhy^qG|A?VV%nt=KOSUsVq{a9*`pVYe1>HD)In#r(|DR48e3?#NIq z5=?$PX19eq*vdL+L?U)>8=#5jwO z-TWnZ;S^#R=Bp9cda`p>JqRP?-u*X*3Ard-yE*42BRctN7y=TQUSQ9Q(B4%O7Z0s^ zKuBs=KYK#+41>Ij&&ctM%(18sC_WUh$ZNSh6O~h|<=rgnmE&s(fh4>~-YKSLYwE;M zxxa%k6g0PDDL};U1yX>nbZNbYp~a??X`My42m7($OB;|)1nlm=DPbHeBkFGnQh-}Y z&>1uMFwL8Ef$!EuU;$p`EA8B@HQx)JZgD%iewWE~M+Dyb@Pe3cmZ_U$g*`d#;l6it zMK9nV^mSc zfD;=72`;ergw5h=(7^R_AJA`7&hZsXLl4?^2z1fi$AwNbFIo7R*7dw-Z69}wby8l1H!2R%^uGwX1!trl5z z`^NNORZ*(bxv!wCIhsq;nj8_Vk@;wv%954AP&LX!d9akh)z%qhvBNh!wzD)F&3k*i zMlppR(?oUcmD_V^)(M%DlA(TC^Jg6UXejBk9(^=q925x&I|d@zk;wd(MjD=%*%(SX zbi{-mqmsKFEvc;>VsRv{w~hCdv+hOD;XPa!bO$Ozd+u4PC)6t zoE*=_mB~THCGSj4jHBmC?xn=PuEVYHdS?#g zwno~Y<98m?g)1)4`(TV-E@%i0YHy0Cv;)l@Ti0Ad;#&)H#Dwr%!S%TeVaif3dsu+m z>PT;G8n7G1ZaW(4J4RlVy$VHm9}phJw4K&V&wJlDjDS&2JAc*dpP9-CMYI)>+>Y61 z${o`^Q16zjSm{%_g9-}SFcNp>Mdbfr2727`K7mMM$hnz<@!Y1(ezqV?SF;=l|TZQGxt3UZCO6@yRdp4X!5-ip1V`sep9ZxoM7(rrB@4oR1+ zLX)~7B;(`dw=II+z98==y3%nB$uJTQ8+^6Ki`gz&noe5sMipGUsKfk$=AQ>nE6qX6 zmV$gKv*4=0HIgrI>?gUOH!U0E!$_m0A!6~^Kp69D?Ocs6Zf3bb$mN;NIBhp8KSzU) zIv4D3)o!c>?CrFST3{{;p1rDGzAb1_`Uw8E--6ah?br~L(=Au)_jun|wTxnmKXviG zlxf#b&)q(~>sDOh+=o4zg%^Ppy1Py{c&u3cY$bpC)==lEocG7aBi}YzZfM12vMaX0 zZRUp~+vU^>gWa zK~+Tutoi_rp~gpLDtB-YJ6LL?8{poBPc?~Y-yg{?ujh<~IDY-JUWw-M0Om+x>lue3w8pEB;vcE zbo9H_x}DL2q(7ccu=ZtrKe;W_c82|>330HhRD0r56EYZ;u*B)M_-CYQCJUPsym+;G zl@yKvfoW(M%*rWcMDs*59TJC&u@OurGR*h#=MM(yXr|xc4Z3pfJEqAa6yi}N1Q-TDfn^Wdv-%byPu5DYP~P|#=*XQ z9YyKh0%yHbVapodpzCIRt^LcNA13T{(3g9~3(>sv`Ix}zQzWkjw?Vy9uT8UUIX4lw zZ#S!WG*5x&R=c>4^l`B;(a_qb6UexPeI4JC&z__lQE zBd4zpIfJjiiZF_uBozD0X21A;+1gMqjWFDRX?Td6a3j$jz@a(vM@;K2D-6>3JQ(RY;KkxVLg95U1iv<9WardyVBabf*6zwP9vb_vM zxl!O82s$Adf~7N=pn@cmcj~E99R|#E9-^OBm`n*{u=A$#Q`&im3@A}*U6$r<-Co`k zG)4+$UDGXmn&<=)q4Z6C_n3Ve84(((!Xs%B_)x#eV~JxrIrQQe&kwFB4NyzH)t97? zS$0Rm7|z{bx_X24d+juT|DiP?mFhU+e#&bjZ%jSd^-cHM(3_rmQ1Epq5$lIODO2j$ zQx%+l?tgjngD}qn5;LMf;KIluMbD~SMdS|b(RUSCN@erwRg`^2ztcSI=H@8T9x1HK z+fhXNh^XSK>kq8(bCAIo&w?xu*rT2L8sz$%CUSk@41z|V5V4|Ce zr$`=IB(G$xnwhUr6Gd3*26==#n;vEV+x(!BN%4!63{PQloOCiASIS%bNNd1p}%;>Gl1y|I(e0rf)Q8vmL&%^C9=#ld@p7;a`~qtUP^iHT(Jl~C17^aDIfQSk8}}-;FMxY z<+ndqRhJt$$TH{{L3EV{l$)j@b|v*nX)viCAEu3kF&RTfA3}X}OznwZE_)6x-4k;^ zltn$vf9W`Ua%&}?xYzTM1pWTOEU?Fnz9zA~i?0G$SE;EID#px#|8rN~)qjHL5x2p(w=+F=jru1U;Re&);ooZ>b@D ze7Y#fV8AA_u`%d#Z?74NxO{~qvZF40-0)5IoOYbr_c%U0sv76J6LV44uk%KK+%z$W zVSNuCF>zh_7v>U5cc|_Dink{yLn7jSnz(lj)<#@<(Xjfl=F*oeW5o^4;T3G)tc>}8YoHd|BNDZuV!d;gyql=LJ4(*eDS=Nl$INd=s~ZsheZ8|c|F=upy1!gXyek{jW1 ztBs)k4K{HKi!oMYtin&xP8K$J!}AmhvfK~5pI4WWCg-Y7g0S1iFF^?&2&2xp6MIy3 z4a6oz%AoXzy)3U(q1?{T*Ve`d&&JczRtFlhbqcf;B1c=$DOHJe>i7#%mIsT=2dg9> zJ!U?G-7gI8JnN79sYE@~N=5Sa@$ikmQlj#<53mn$1WE znXIG}y1JVIJlJd6x9NVbGUFxyfT1GD!xoVpp=Y?Vp0orz>7A1U84RF+);)JapA#e~ zL?ZECW4<5$@dJiJ$DVAkd?{w>fr}%Z!JM=`_ncqy` ze;hYWF-K)2>7gTDo#-(1+w%im<59X*sZthc`!((%21R}oxmmEXtLiBd&ld9DUS#=m zcxdp4jrxfbVNhwmJHcm!lL^ zTh&CvjUcOoGOZJk3PmimKQF6y$u8|HQe>;iUM(M9Og&_sMlE?fM~A$p&$~B^#aZ>_ zKKQ!-4JcQ%&Cah`*Li+*p;+MW!k)<4-wX0;v+%nUH1AWu1YdX(S^MGf^5@9qk%~0r z<~PGmE^Gc-laqkIP2v?*P7QxT28do z!JA5kAL#`V20dRoj8*dY73t<1t6qHz<&$)sdt1(bn&}?n;RN)=yCrewH+sm}XhBa# z)flOYwu>GgyiH_%rc00F>I4mht=?Tc@O!x`MAuH66ccdb=+UF?^v9x zaz$byGPXU6lJ6I>YIKgcvSbmazU#UsiBRCuZfF}RRDY>~g;%YUTEAE=r8;*tc(}5q z?jw}AwmzeI8DBRhAWN6DIENOGmZy$d^ZqbW*ee?I?1+uDPMM8hUb5*nyMdu@n=$Ww zJLi2&B@)GM^dK&5_N&Lmz_soavFPJO`;@&1`NKz!WNY{O({8rsl?xOZ~TJjaQ^NnB-rZV22l=NtDN{wzHl8ir6W7QSk zwGqc>MVO=vIcz4H!e8odzZTw@iP5;1s6wR1q>#uZTv1WfA95t4a3QT+D0utSZ-_500b z(b6m|xG$B`WB)ay$Nw|;RENTDz_E?{+sh-TcQZ2EW_pUG#i+NK|Ljd3*<>{sQ>n% z@d8w811{;kAA7-bUZr2pEOwr4Ktzo_aLu{G*TfE6cy9HofN}j5i^+p~DC356t1hx- zKJ!&sT6@9*31z?bESPQj%ohT7@nWh$z?v%I{dbGM=~T;WQ%+2#{W3e_u2<;~@C9fo z)d`R~=eNJEa`qXPdH4-8-_=N&jZhGQ`sHY^JEvjQ2oT0C12+&A*a0OymJ{EDWGn&G z`lrax(vCNi&o1`oCV=u?6+~t~WZnftqbfjSX4c!F35C4-S$v;Wtn>j%;|ro0OTqZ~ zdwLJ@KpBx}aUR46yC0ss-@Mdz_vX0vV0v#p@47g!(8l4)QZ{4hYn51-?hPE)J#XqM ztWxuyi6&G`iHJI+Q3TtUc3C5AC2?(x&j zDa?bBKn~nF$B#&nmM}CO5~k!GA^%nQ?5`((2fjC*J*MkNX(*)PeaPRysgZg0VHv?~ zd?EXI-|m;=9@4Tus#h~3|4R4-uN&!NZy`<>*^66VX{GtX9=0F)0&tc!l~kxk=LVWj2$(?U?en&6ZN5RoSG zFxrze_321I7UZFH+N8s+1ISVCS1$o7#JLW_A#PI6+q`sw@1*m&wIhd)MrSq6+KbPS zy2o+|i-RG5C&%B?tS(c!19f}+=$&zmhxTfo;F+y=+}g)GqC{Qqc)p=~)3{9LUMJxJqD&3OvZg3eClxA0r8 z{`>3mVj?DJ4|m&%AEG`<(p>Q4ky*akwj zvU)#!x5fl9%*HHXo%0Dc&Z=J+G}@(8N0SL+V;-yA3jUw8avrLVG}x!t7=2`pt%`8{ z>5OP!-b+vEFs^ZJpAk|lQi_sO7NsXtf2uolh*lMQ40<;d!rtg*`>n7>bf5plmo#M_ zy%@MmYPv%=w8JiOBa`Q0MdZkHFUuw6%JIo-T)#2~?}+`tlFPVA^unwp<# zRvEI1=X!a&kEz;Sr9mBO6gr84y1bz?(UdJiGNunn&xlG0aRec5e;9uRn{qB_T)G6x zkZ2Jj-Htp`&r6WT&PkuH(`!7wlfvmc&Q?bEU%Ngu`L+u~^FO88kvQJHye&jknJlHh zm(6N=p0B%{5seWqkNis@5og+HgHuQ%kNhdJk}2m2l`T&jb)p=*ZXHtq_+p38t68hcDA){BJPQLc?PU? z>B70{c9uoPeD1+wJnI*3vs|{PcxaschCMx6>1%=?D6}S|{4QQ))S61{d=|o1hf1V$ zgR16&{M=5pOADNASpCpAFeJx!G$og z^P2hXWhZ&m1hgC$6cqH@9ZSx2whtimqtYqIErN{111 zJ8@;Zo+xA3pe7rdru|41k^N%BRWP2LmiDw;KQvoK*EJe*z-= z<9sZ(pX4)A5+fJiz`O5!S|j$l)g4z5)|ZeDl-kJ7B`X>r_!ZJNlD<+0g8B+2&{Y{# zCfPsEe#G>fR8hr&c%wMYF<2ufBIeX$zd?-5FLUpd=HUqDFd0ER)C)mXvQa-CBwt^wM=kgmYhDGqMOo%w!lI?z?RiPfW4>c}+eb$9+&qLoQlG_VKI8UTy*18o9hh0QMyR=!fgLAM9KW#rq`8kerB*|=z*=qDwk>M4A*2%qGckb_C5y+C z-0h?U;up4mZnMiT4pT7C5UH%NY{vre_mvYh^PDwaoAF5rvPQalYm_tUT}7|!JdarR zdVE=dvCPmD)+e+e^6c$o$Pr0#W}=>G@^ z9cg;RzIrd$judgUcDmaieqgplDYziDxf?-n-H(+ON8T2I6t%ux0-cv5DIeJiPSa?o ziP{cOH*sIP)^2OG`f%ZoeI%?P3)JJ*W2=Q{P1z<0w9zKTPg?Hk1z?%ajI+}>*YPR+ zd!)g?!u9@RWBFJAwU)){>mWU$iBmJJbyM16*Dtf_%Hb`riC8hIu4n-6PfEjb(3^xW zy*M*ro6c*Ju9lIS-%GyHp^og#>zpZ%&9Be8O|75%82HUG@z)#@P-JfZE-4^tAp`7u zjZ4S8(>PcQ-mG6{-Kx~}BQ}`un^v86_IWc$t-^#>ww2NbQAH-R6#Wuv-(&J$02QAK zh$aUZQ(vjo125n#OC@oD(NG}}XgnA(l*CH_(eRpsbWZod9>}^EbvegzsL(*|i#xj< z_CU;Gn6t_m7f%QIXxc4-pzfDG{&YD1QAR+%E@2?)+Cv&%B}ZNsp{1CdSX9XtJUX5E zhuZ(r6a43^lAwX8Hsucg8Ug$4K~ssL&t1N|IvuiNm1n*;8l0;EVW#cjPk))eZwH(N zAyH^)6yGQL&95STjc}jqAW>y^cQ^584cf_T9?eTfn;W59sDFvPy%O_>`&cs=U6Fi$ zL+YAYk)f#e2`Y#||1mqu4F+ilIet764aNc_0aMb6Q7{Hdwlk$Tfb7i%X4(kFEoRo9 z30Tu)&isE5i@j}3n^JHRR{8p$yYQ|W=T(L6#6q?T(>|%oz-xYHvG~;AE5-lrWm+&d zxThaho9G22z_5=QFnQ&y$>IcSrPBm24!)CmB4LSPU24sbFe&@})8@u(R~P8|W`W5e zTFlX~=Gk<>(E=kTWt(pbyTJYjjw;v{q{1$3{soArwFH+7EscMq&J(?8p+i<+l0%x8 z|5tj=|5&x5C&3dw&*&8UN6_z|cXHu&9mjnct#KK-?R!EXRD7ZF>8g$hj`aOU(dtm1 z406bEU+c*CrB&%?`r8~5mhp>@hGS)HB7YXpnwq$n@LypUOyY-uCdNp%Gn)GiZ9Kt7 zE1Pp-=i21tr1f`@y8f`;7eu*~I*bT|fTL=V_}O)Ik|@lcr9qq3^PD{tdueas)c4cP zbjOZ?a8tV30(4_Etw90Tqo=P`JP5 z*_;k4LU9FvvsH>K5tqE!1EAwoM$3fv{I%TRX63)i`s?}M6(24+0}@|^vn@MgRi)e( zI05_?2SB-Dy6l!YAW(Jz*#INJWwtz4r2(vCe1{Gl%J~zFnm~XO>+D^OHxb!js-p}N zcjUn!t{Wz`I$)I3Rgfe9+C#~ELJzDhNsIDd`pp0J>A^FH^3gsHcbCV%KF;vjXQ!H_ zk52-CB3y@iw^dLqv>~-L`xcFvG z?!zV_-R`rVCd{UK2wK;XPmch4L?G)3rzt!~bz(W(nf$=s*Q#-`0!6A%))EN>g`!OA zEW49MH(DU_?Z5=8c5vB5DB$?fX|ARxXZ~9)_fJLxU}$gG(;iRU&f{P6F76sV#`(7h zdp8fzV|nPImo(lBg52c!&&$8wgn#16&(=(GFr?0pr6oQMSz;`RC2c#}Pn+BW){lTR zD>%037%#p9Y6}IBO}SWVa`z;NlK`-i7N~yc!HqA~z&PeSR;ecDG=6Ofi%tOJSzfOy zJ_O4~h}jbU|Fdjp4oS8!2hi^R)1Mz|L#}1)DAUW8-`m#!Sx*Sp?cYP>Vvm`N&^xaf z>v;=;sJL=x+~m=5(1Z6rh`JnPy8-%Y5fs#2f-?0faASY7x9vWY9EgwGu4LcI-O1$Z zMH^N1yqW`T_;p}St*E{DQw2QWRm&s4dF}sriT@lgH2ky#3_6lK$24;EpV!W$ahMN7 zHPuBB8yc!qpw|V$&FnZMWwpMhwcTr`&9*u3xgGY4H;z5}<1#(^$O70)@d3w9cJ^z? z^8+t-Jb)QOj4qhc4YoL^UL6QuX-2$JrnVF{~HR}?6ecVe$Ok1{6PR+dmtGy z)4@~F3HV#{indw?U-0v00lX!yunP{tl;^-rs0y~@h#nXOmroDv2PUQ(*KAPGnFD1~ zI|U?Sm3k6PM{3;*78uq@I53Il3l1l*8!19l!jK6EF@S~*b)If6sde-euOv@B?=J^% z=_=~Z|NBNi1o;b$p7!U7Q~dV`%TQ0SGuPF77=P`!(AuV?G8^ujoR5GVj)zUGFp_k9 zTwhTy1na*GPKVyc za7lU4Q_~UDsFks*l(4m0VP6Mb|0-{-IuG{m%!~am3Z@W^zC+6|;9%RJ$CrQIFaIpp z|LZe=sNbX0XIJtwEdIT!RPTZA$!n3Dh6~KUnK7WHj+D? z;SDxVQxQ}}*1RpwBc)35V59~poI{7GBGzB8fx!_mHfdhUTiXXzvNJOt_qdq=K&P-F zvCPswu;R1TQYv|0rE8Yj^m4?(EQ?ef3|ITi5;( z0|W&X0qF*%6_IX`k`Nh?7U}M07%5RiK)R7`kZw>=kZy*Mp*w~eVyN$)bBy!8@AG;7 zg70s`%+18U*IsMwwXStt%G61J;+8A}<_EOD{Yd`L^->JeKv@r{p@sqF5!obx3VI7N z)=mszJLsB(bLa|e@9z@}Lk~7}`+ZPA;YG%h8jux50{cj~0ZW1 zqe6hbkvJd(X9LO4Bj}rs+LGaDkiN;4{@2(-V~)`@8(#`V{LyM*zy~(*c21}d?;p28 zN5tp6T7LEY0TQkpUHqKN+>`8txo>`$>4D9O8{sDB^G}~)c(d-oG+sh}W&hO#X+B^y zpE8(V+x!!nFk6V>9{4o5Ri}PU%%}Oxv)KW@U{DX1uZIL2Fj(f3c>!<82soOqx|pfo z5cwkTIpkLq`OkK8D@rUk5K0JvEnoZPR>0@T$pJ%A70}-M$56)b68SyVu`XgQbaHAa z0D*__hCso;BA67zWAtqp%+$%x-fW32 z1k1IQFh3&yrHkGELU`*xNfmzj?*F|R+%7co${#w45&Usk;eev#Faw6q|2GuE$B4;p zsfV)0Z9AObf~BQSH&X0XMUC1td;O#-9Igz|KaHVyyvZ&zoy6; zcL=2p{yY~BM12n#?Bkai{#oE2tPq1kV8Y)4ZLIN8&_d7q@D;TPjRe{3o}^5>ndZkd zCH~XSriWtOaU%Elt^V^5`dMCz1RocGWpL}yz4bA}WTsch^8e|$h26P?g@v`S6V-Vu zhYI!a;`SQ+oGpR(oCDMrKH!mXFub;Db*FY%>R|`^Ns$)A$&rB5bo|2MLxeel%s$0z zAr)cgXYBa53s3q8{_8g}xO|AXYj8Ry_}hwJnSyU}5WM^S>EB<4!O#cDPGI!g82I@3 zNLf!@mofVmXsfkH-kXDQ6Sm)>He(7W6zBc3bpQ8}&n4n%&-65OrKj%CH6ny-Br)0gQ0!*07<0U`a&oz*a`tC9V1hjqg#HijL%nQj6oq2`{@5%^nV}G z&|U%tMLRUc{KufOK|*;@&fT&7b3Me;h^8-gOL-}!yf|8O<`Tz8`w%ggrm|A?2L zB>``?=knWyPe0d#X8Y=vkIAec^WMJ{Zhl+X@7*qogYhFc&ilX5eKgbNAWR8w1UvlF zWehXB4EFB-H>;m!bx;r0tMO1QHSJt=$8Np(+o)|oaC0RLGWl&gPcX}tmmKywfGzYtE6s1tp5^l4KD%j`!Oz9`CeeifBSqXh0ladyaHZ()xk-7A1?j4 zehogKxs(#6p>IP2ga&b3`MHxWbLiKwR3Q)LFv1No`&LSLe!jjLifwxP^}UN;Z}JPj z{H@0VK8ArA_>uM%E60MLs}S}PEBVmJ#Migf@&9g#-&QC_^A^`_{NQoI`6vsw31@ll zV|4Heqkd7M!`!}gjoVDAaMM>qg3iwf-3{}{I(WfkqL6)RA+}HvohMfg)~K>(Kx4e4 zx9N-JVT z^1Z+T1q0xdxMiF2aG*@V^B!dO=5fIfmECNwhuW9|^5XT?2}LT;jOvEO4_lY%@R|dg zh#h~WC_eMxEH2`nWE(>K%Y~_Y9t7@COy1=W$?k7s_~RBC%r7XT5zAd;y92i0?>LH# zS6HxASO}3W{s47X`9N2+g360eYDH@5wa#XQS&C0ge>Bc9>*cDx4GndTQ2k!c8w=ObQjYuSek$ikx9Av9LYkQM@83RAopTBzVnY^z{FXl)nLHG6Z+-L z!F(MxEP6xm;Cp{MbJv*@wL+AeTAn%!XH-Xwy+tu|t70+-z8Y`5UUyO^TyiONkXHZmLD*}buPEHSKr1+M?OYPOO!&%(`L zPX7%A?9Gyn_SWAK6OOrAS!?M=zwh6FutZSzFap;#^n+4w1^_WEhUKfVFzX_07Fr?`i*Mmx+J{eK!uPGEQzo=diCA@4(K#vx-vCRM?u29_EHB6p`ulB zn$MW7x#HTT04(E`>(sW`1C;Q=IneE56g(Sa7w8gp?0BC2ZSBzs3m$QyUHlSXKXS6i zpn2;TJc4&HYPyiCK3TJ*nQ%ubyVB$+_APZg0*bv+8AAV_T$uTNGOV?$l%-X$D4%3G zXcwuRbPv=PoppxIPc|yMGapxUDDCcD#%@zi2GtF`@9ob_lhCF-Lp~=f5pA;LQe^Pp6de~KjkVu zrTU2_#d%Fl!{kp=D`Fk7hUZ zYvz`i$zjKmBKqh|b2tFZW#=83Fp2&$XrJepiWlA+tAB*lf#KW9V7#r%dF@NIr(t&NHtnhBclvMC59ZJ^pgP2+f0}gZ z%4OLK7NR5!=I;Qr^ZkyUtKcP%3jpqK#yRT5CFwYpDN|m=!god8^VDswRngsb+7kou zyREW3$@*RZs@9?QJBPLKQE+b)kG~8Wg+)ek4xzUuz)q`ROaf#n$ww&+E3^KQ4vvB-? zYoi&jZ|!o~B4;$f)6+Eu>~!WBoIcT}>_Cp$`jf)ap4?j*Xf)1<+Hp2yU8ahnIcdBY zXtd{VZlX^eOr6CtZ^}CtB0A#Lq#j-LoIZEZt^I^FVnbRTV3JOlLrE=qs~#7~?*aw& z$QA0~N)v@+8))~@2#N4DfX{YV>M-Xty!6}QT6z&~y#K`o0GzKqd)NN&#e`iJ+PwxBSw^v;%R)z<^ty~X8sbw8$dBuYGJSkXk;O|HEdz{H(9y1UHaENh7-c>? z1x5rilm`I&tu#@^RC!t^193?)$1IvUeNWZs-q z=%ufDcFQb~?{n6YQMQ_D?Ceqic>!B2$_KR0nxAOXI&!Ca3+dG72o)fCrN0j1Uaipf3*xeOXTVV~-U+gPGHil^0iTn>V<7*no-JX9n zdiHDEib-K5pZJ(awRJ!E<#+#Nk^g>V>N5{w*5s?dAU-FSnswe0;&%b7RWQXkJXnSo z#}?vDpqupL=~#(Cts&42K&CpN#+`H3E9#vr;2rivo?f}Q5yeka9omAn5+2!ph}$Ke zu%;uA!{D>KV!u2byy8ggMGLc@q^&KjpZh3@+R$n=keCbEO(*nE5jlQ#bQ;GxbiUN>aSl-Q;EUlL^DkMF5l4IlvScOdo1zg z<*#f0y&k`wqtV1;XhXjD*e9or;n2Uoq^rRShjaon`3Z9Ich_@*x_;7WvVWFrtD+p= zh)&aNGyoWtI=fub6hUa3n`idbGDJKQp7+-o8TZtEob#Drcc!ppW@V)~J`dSg)ZR5# ze*hnbM{URipMEa4ot%Y0 zh61-+K@xiBr06#LtfjIx)NOC(<%4Q{SIFnmO1M zy$b-X-^)|@Og_y60fAtkT3O8d5qas8%;aUjZ_!Zd+{S@9K8zKe^{1_{?pU2_?LVad zZA(^|eh#}_PLn6{0x9o` zVcGj!LK*qWg1C(hP5PRep^UjtDEa_0OJBSE*`k;JmutE~ z37prz+mtk9f2a8m@lznbV(%}f_xsrO_X}r4n3akyK2vH$FJe0y0T;^zZ~_F5$HX@e4Ed;z@-lD0@v6I@m$yv%z$lFAeKL1h74irDF+N2! zz#w(tU0=%&+RihoZ|T$2#rYVT?o~4YPJ@D;Z;AYSBk!d@Ij6~J+Lr)!>RNza&7FmSMn$_rFhvAMo4;u;V{av$OyA)z`U-Ib~@SBf>VJuVW<@ zlUfx3h0(y^@;}6{z+}&4dvaRE3k&0o4%t{Gd&PXcC zh3)Xr4lvqfJR#~iKc~6Nc$aIm+!@T}kYaq=Sxf79CrXIYilCZ2M}|>jlfq%U*_a!% z|GAxBK*xiROybTyuo%nBtwaqVqAP5cX-?4Z6O0kfnI;XYc`1s+B>=^%f_2ePigPe( z1f^1F;ZD?JYupN(D#g$^o3*YD<*+Cw)U&Oq-PI!yJ*|IA?AD)ML0-TW{EHNau5a}u0fj~ga`#!rnkyHEZhs}uPmXmd^MqO7EtSwhElIRYd^^D3JEwPQeY!{=2t)VAEFrMUD(KW;Yt zNU9s_R~z=by0!j$%R}k|Ibu~j>*Hnmt;)$Zit#FFVIqIHAL2FDo{I#an>hhAwjIsi zIu`(`o5uhZn4@NMwuD3&q%$SBr}}MqETZ+jBZ?b!Em`WdSy#opcC+3y+&xV8Dp^;! zO6d>Z5Ya!2@;Yeg9bqw`!X#7afvBipQtnt|^mD`pZ!>@^b{r>7xISU@n!F3T6%Q|CC+Di!T|6N}V zGOz!PXtafV^U9^~A`qV+tPqxaI(M6lI%1zXUEZ5X*Tn`d@kh1&vwdm-Iv@=jEpd7G zU1(KRf!=OyueQspW~)1A?|U@obs_USjzLz;0N}oaMeUqTXB_$k zTJ)|qlOk-E8dJVLO z>1!NJ2+I1YN6ub0FA0PR+oBodFmM7#P@aN_pOuqxZp+@^-rmM9+dL#N`OZN^$GJV) z^5ByzoU5~t^h=(@Nu1*nqa@lXjrEK6@u%p(12+YTZ<7Nm9*jFH0vcixy$G&dFw`WC zWVP0!uG*iez5qL~b2wUsw8!u&R~vY?Roc1ESO(3#q-dzlZJtff+#KX~k~* zUW;>Fw;k;GM5AAs(r>$aqwNE_uL>$EpTM*j^TALPHUZVw4g4`c$;%*rfwD*;lVx(% zqAK^9r(A~}OB+*1$lyK9oA_{H%Ji=p`yv*DY8IG8os;nyZ8*jK0{BZEwJ9Gy?4}a_I?0hm+@2uc`YW#Fe{15;4kC~Mf!=6UwJoSox@OK@4!;9H)2O&P@gQSPT>uW0;pfUXWXv4q(5{b8K zUVY?0zt`3a&>PyDj_6GNT}JzUqA0P2q*ntg&Rh%N-D}bBLKy7~j!ls_%2KJcN z#48y*(3@FLL}2!LSA5fRUo5{CE0w^BW7IN3jpM?#vL_lr`Xy*>&E3gyuFkz{y3&oF zZ6m8l%XqsOZo*$KTa3mYgDjy~^3nVGMa3T609_a(yKsJ~|Gj)KZ}_GS2{ipX@sTCN+;Y#Qdq9+^CsCL)yEidH(AztlSJ?)nIUlv7ew>~n z<55aU{H#0EKDXIo^>OoIm!=JoC!ZL9eHIM|MgldHW8$~_qquY^Hh97B>8~q(Tc+JT zF-EtThit&sWH04hqrQ6M9XMb5NHKs`+S_HLKBC|M;lscRY@|I|K;Tz5Ku!@fz#_&; z)0^5Ar-2kSMx$?gXGh*9uxFbt%;+*z4z3~)MScp2$7L8ZeA5{+1-v0Z_|lScn*?`G zezLxFL-rCDFr#of>7eJH)Pw-bgkkFys&QoYO;5+e@h8G*bXE2p-bJ*vF|SJy#5!W? z%+ALl*c4X-fk|9t3=}zcfpS3wu5~a#(n9PlX2NE7_IEBV*E=WQ*SJhUX{xn%Pz$+0NHb zWN?Y^(&TG!C~h?lTc2M~VsXb(q6JB*N{y6x-Wwvql-ZG2Hke$klDoQ7UBkYSbr zsg+CHd$J}bFoqEUxGOq#7~jn@F3&6>InP%2JM{E&8J**Yl%PtPzDxiGrnK9f(C0Hu zGYr%4YuqT+$WtpKuH%7q^G?YfP&GW6;4jNkm=vg()CzqJXcu|@Mi}*{w+*om_2$5@ zz}f+I#vvLlA2*Q60AXlqL!+b!%vG~P)L)$fX2hMdukAbVsR^U=Gd~k;`qu=}(d;kx zJ)ra7pSrI^!RLWG{6MCo%*yGepveSUMhdCC{y3QCMu^yx^UXEIRlMmvN4t$)a-)gL z4HM4m=Y=bn`Fs+6r*%(=o z&Uc&lKT&~97HpL8=7Y!WGeUO3f6i8N+MtFyAOhR_)%y+)6MK0k_6*z2q&t~epG|X7 zwH#XaXdqidxnDnnalWs@Q~7AWj`|5n2*nRV&eS%4A-#Rd11y2aXS3U-`ysEdtuk9r zlvidGS{OASM;NzujhP%Y**srU+eT+TF7D{j^<+>^{;|_E-PQL0Zs!e_S?CN!}n z?HCF5IEd__oC+DHAGy$}YRnRA{m6XaIa4qtQRc=uGL(IwK-$tXskljpHZ+FCT)FMr zcK|)he7u3aX;bS~lBU{#jsT*P(z*J@O4Z_{jTf(U?kup<|LBQ;u;#yl@#ku$*Q(vs z=V-2+VYk|Oj5mgcPnkST8m}-c2yrai`L-TF%R8Q zQFbwcjmh=>dp!NAlR4MWJmu1zm8EkjNlN+GsYaC%78B#pgW*d?Z% zD=C+hZs|(Vyv0!=5_y|=w-$7T+qG!!!)zl9l&@j`X=H|Ffa(Kt?x3{k&w3WkJv@dH zf`@?XC%@bp-mm5r+*KvN)HT z-OJTHUg2t+wRjxCRQ!GPDKfx)!{$J`o|hlu#R;1xYaYynibSe1Do9xhzJ)SENl2l*q^_@{Z68I@`d#tQJ={F> z7}!HGJ#g9c+_dVkNk@ha9ZkVCp$0mG@OW8il-_A-U=ZJ=bgLdME{i`DaDW1RI%jic zVtI55)9#WF#9t>aR?AJRyg*0*?B=*Ye-W5*$Mz)DcWLOXp zi~LAIT+2Frr77PP4OC?y0l5`&_RHz9!~PHcvCPep7rM{WNgjD;eh8)(9JqH4dB-0) zo*9ABdPnp$Ow$0;fJ7^_bs2?#*VkmxmxnLR$I@zqDaXEJx~9d2U~o2DDm%@~Xu^~q zYv!ug!&AbzhewiSW-Cj0nW~vVGdv0dyCSDnCD*98rzqL>bOHUhhqk;3IwxaCp7D5} zGE9+$X4wsn4ZXjTcG(efqkBTcr41edCK{$v4t<`S8%2*Hm2HC(V5L@SaCxV^Z;rhK(|YI`^4h4a%4 z!Ip97HRH}El0AXS-bnuR+uNw9TcJ^5V1|H?L;lk?5G784w=#q2WW2>*#ZTT};UUQU zgO_U~rv(V|P1SRk&h}^7jIKpGi@uqrDehW%>V>mTABIfdiNpY$*W>X9*IL&P;KcZN zyl3*t+1XjC(-`hpqjep{)07KGN5^tLCr3l)@1U6;6&8!O?$K;L$g&hz>W4&i7(K%K znB5tk6IRDnT_-yJt@u?;OiYGy%p-5t4#eLfK8+#lP7`m?z=giw;}n>b1_KqonRu!8 zS1#EuNmBy-nA;NlUeBBkGw{ef{q{ko@t69saLvopk^}-brb--n1;aGvaxHdNPWV$3 zzI8qX9_W4@Ay()(afIGlhQ*D)EpEV*s&>K<3^>U%J`r^F%TMq=*43)??%Xt54$CvV z;XTN?mLcYI;-YOSB}xL~605*^9yXMAe1tur@S6}FAR!bI#(^z%Y3Ayc6+!Z;aIF z-qBKW6r#aHZkfx`pS7lBE3x*kPhT84;gD31_rGrN1}@*@7TvTpKxCADBPA zI+x+yAh5p^T{!+-$ywaksHL`IN+D$ZTvv)z^Y~5l+N(1`VNW1=2-kV!dGu|`_;4## zsq<>HpgqItTfl}Yn-#phOGW!svX3A52b!CB43z)_cArT151dTwb0z?R{hgTpOtW8V zU~X591xjNeUm4~YCZJPgFiCJzok$Qi$9b-(?*mz=``uu`eQo*`K5LS&r@F5i zPXhpW%lr7|5zMGbfj(*FSml|oipDnr+v@BPISu(izoMLlIl_UIt?}A;SA7WH2vndb zuYN&o81{}B`aQSVWEr(r*ovIQQ=ehsEC~^HA{Vws=^w})C=#IGxmx$-ZA_FZLR|1$$tOLga<7idP%vtO41=)fmDt9{4t!KX)*72|7;dDv}pgS{2Hb;FbR^{VfRJXUNVcF+^%;k7PO z=W8M&h*ciCl=)}Hghj8nLKLW5z)a4ti&8v1cWX0pW)srqB?DWY|=ftJKP%5CpvU4!|Z)xf&q z2zb$C7;wk(WX?tJ^5KYOAtb6O{g%$@%k~GPd~UHT+k?vRh`79$Wjuga4sMNdERgPe zw4WsGeYPUFdu(FOoYuCLT&C^`4JmgFio6F2`0g}fg*sUAPZV%|+`zHqc*Hwsv;I=* zfWLlf81NLdpIyXeP+06zt=*e57ztU~9i4Ps&w~N*cw~B8XeT8q`J>p`iRz~pK6WkL z1_DJU%S)-=HaaTEsJ*r1GfG_K>=jF04d|QAwKzRwU#g@-Td#7E$X*yA{u~^KWQ2l2 znh%57z2Dy_@G3@)O9bEs93@b$JXX87=4#){T(-fo;Aq+#CQl^k-3A?+0&_=xE7ogG zjA&$D%4$VM@{vjuGNKN7ROpd}<`-CbAW_k+Hd1)G( zy$&?OGLJ+-lw&p?%$nK&(!RO$5lkUv*@lKvq8el()L_QF_-LUsCe*I19q=2R7o$w_ zD_3oI4d$LcJ!LO=R+r@tOVvMy+m7Caoi`|gsw^dIGms5a@IJIJ)NI`Qod(yUccfP2F9RbH1q%o z@kJ_ri-IogPkrU&d>@O;VfpU|1q7nif3hC5Ten{3~~(r9*`U}0`0CAPh4ygdAip9BZpr)ZhIoECSq z!g*A_dZ2k$(-74Iw~v@pQCH~|7PXI|sj@fqHH-5a1AuC%w}b`UbA)a6NQrj^QCPOD zB6Ad^{2KlvxyoWe^@z@tO%j1h^R5T!vP+%dd z)j36xfqF%~br8EU&j^cBsK66$(;lHCGjtbM?|V9FFX@zs#qRj=HK}1S z(PAL?WHU0=u#4onAVm2pj`#do+J#X9t%&FsQu`+8lnIy}249#_fQn68INhL0my$@VYqzM0VwMxI}BZ*BA zrVn}ka;xvAN&imfyB4U8%!2FCc%b3kI{g&aOqUHUCL=4|WrK=*)tc~N%F|@rERsqS ze&N-nT5rl@hD66)81?bJD!`Fy4|?}(S|N$oePK8CO)P5t)tS= zo=uxesDVaF63^4E$03yXR|B;>(>&Qm&;mF3E%rvanu^g_p{c9Ra)Hgs)^~QQQk9X( zX%{y>Tn9wo5jCr>--cE4A3Mhtv%FrVb)qTK2;Cg0Iu}DxI23slapK(Jw(zgr%5;nE zRxi|3(0%AEzE^W>fePa4M7bvv>Io_(3v84ZgyD{d+9TFb*P4A!GC*lq&QAih#zEz!UiISs;f-MleYHpK z9S>1t)ofjB6L!W!5)GBFN%AUqi7goRkQVLXjK=dhfRE8Q=(c_3o0VO@yd^%Tb1*w7 zO)6>-?v-0lfqhCr%J(rvD4pY_9nZ3(kb6g9=~h{9%fo&cxkvpVgzOQ{5&6a_jT3d+ zu}2msUw`GuB(wm!+>|?6>lEp-3XRWqbX5u1-J86!XQ{j_ey+HIvI)I=*_raZ$RClM zOx4#ET0dD-IGC0Kw^czNWd*D>*<`S0V04lla`5QVxusA%qe;-^kCtP&kc|UcG@Ysj z%(7r(W$D0?^HJ$kgm*3i7Ik!RssY?Ic={d00)G8Cu{3(E?PoP;Lv!4HZ}h6cna$e4 zVlS^1jo+1hJVGrviHuMtf_hE7X@9Nh!pQXM>DxI=y-wbQT=*z0#_T&a3RH_xv@lLl z{c^E9^{X|93cLTY928U*Ml=Ti(Rgoke5bk$;2$niH5ZpTs0HHAG z4Cdcx*JI!3O1M0c9n;BGYp8Ac1D&9e@n&}|r{bh7WhAc7ZDqw~_52bt@0{G9$^%78 zldldlHwpA6xHCB==h~~i+;lv0*-o7rh3x!D+p=*mj}9^JjDd1Mk_&DwA|on7yuI?R zElCTK+dMD~a{7MEO7XWg*HXMw2I>y#R7<}ans}lE=a7E+OZ%Ry6>ev~yPWF;BfZ0J+ro?v_oynnJHs5gu)-SE4 zCr5;w!z(Ip4z?M0YINK|@dr?4LHRV39o&k|tdC#ShN(5di!|r%Sw_2O(A5*lq@{pS zuw~_&lv02B7;%J+0>J%ZwqMJ?2E)=oFCt?3ZTiH|(pcEurCWp8@gcOA*hvL0&B$wL zycY7`3gRZlIYfKd&qA|dk)u*lZ`2++hR^N~dY`mE z)UC(s#{J`p{h%SK#*}F#EF^26P6QzK@RD0AJ{PeU#$dKbMSk{E(A~^VmibiIdf52| z^VGaAn&QWOorNoL+*$2|sH7Gb^_(Y2$GO)P>d&%epdzaxlYIcs(2&2pYNt`87lj*@ z$6tXbT;gza>z+nsu&^N0QV(bh?nbnBb{6Sa*`7?@q)ZzaZlM=wQ;(gp*FE*J9!Y~_ zRJ}zGr8Zo(SXuq>l*jH`=H{8HW$f;n8?C?L6h)i9`Ce%-6@Qo7boru8Hgcr4$k)Wf z-3=UxrYui|5GQMBJI>qo^=@fO8{-dL_njm zWP$k9=1i!y=gZUihBrh6nhwxdepP}l6)q&3hJgIUDOB24=R8E{e3q4I^wwt2&=csF zFNX)r3Ewua%usU1;@Z{$G-R*Had~aT_guS+B_4w4nu?bzO9m3z>UX(-R+|P1a?tkU zvsVx5&#G~8U&GaV-&xpI$d-iBB;hly!HF&Hyt>EsI%~OL%7rdy`!Crk?QcF9b`{)g z!B*%b?xrNEqtHhU!w2#;WVyKZr1f-bQDMl&^ZtT-eOFNx10xMdBa#t+z20qC)IyA2 z7|?f^_+l_wDSLD2FH@(KwtnHZkkdgQ+tWG{w8Y!f3G#u-T$7q6x|K|#;s z1l}Q~6^g550Tplo)XcPx09Kg51-WpDYY>%29Xe@Rlni$=_b<1cz%y(>rBt(O9^q}a=?s{>%05(J@`=Ur2yU8~qaEJI4UXQug@XZwL0e-TS zzInec%?-kK+4;#TWzM*7pL7q~UM~L_<_J&Wwj`VHiF)>J4H_|ArcToADUhrZ#lP+5 z2wR)G?9m!bI_H)s4L=$n6v7o_W& znxd{fSY|(YI3sM>`uRWsH{-2nzkLACN*D2r?T>1o{20oc`SD)9`6YcUbYB#?=oA*1 z?G`?QY2dj!#uXN(!j2Zb$fp8;8ffWLH()*^1V+oQ50?KV&N`5fk{Kh~NDR0_E} zNctXb8xuAl2J-NLln+S+0T@pRYN9N-M(69A8UE=;^WbxL`XU_()|rAjaq-urxdeoH+mf#iG~sYJ7lhv7ZO0v9=JTRG|zWcP%N_+BKdys+7gU{>Bn^WF31 zHtjHry(8hHh6JVi;XJ*Pmu;_m?RXM@9I|D2POY2OyRVPlrLDl^esn7?EHB7A}MsEu5Qve53Ta1j+nwIQq&D6 zODQS!tDDjF;uk(Ad3?HjY1%|ye$n4q&vdSbP@KwSRNUInlVB67T{x{byWaq2N47S2 z&1WuarRmqKDjOT5#8L{^6Np+oR?T!{@y`w|Yf~N-J#$$dP}HPl-1eORzQ40FXBm8_ zUs@NDEO^g9=6bP8BSqvIeVexAWs;*r_0HWQ;O(fo$K%V$da6xLr)9G z9~%s2^}g~(vaR-Kyt-ocG}`|TL*7)KZPTYxxn?p4(2gS_rkFJT_)fK3!fmpYISB}J zY504&SvR_$;#QtjP_myEtH2=>Bkb4IH-ydW_->rc$(lXhR+n(YwVf)-*K+@Cg-V%i zSBw{eJa=L~ADxns05b6HzPXeKZR?>6MvHW$04gE3tu0!u8oh7|d0Tg@+=J^`1PG_|OJO{Zu|eq}9YFI%$z1 zKIkix0{3)Xld15o3RtARa^>AGod53NeB@wE19%tAPc~pES1*;^2=G&#mr_ZYCbs9Vm zID!t=)koiMzjn@tYiwVNQxCn7y%TiRC zwX3ejTr3>6nJg#nO(4j+hR~eMe|Uk8^XveHBBG-)PeLTDjSf0m4U60gj{X*9Z{|$A zx;9ccm7(69diDaW?&Q3Ap@c8T-#YjnOG;FN8k*tR<<&SPLf!}7rXkl#N;Q?Bah&H4EOFNuqr>8Xx7L z)$h;bBu7H!zcoaBf3tb`g4qt!`{RLZWoD+DI3z8gB;UD1Y2DD!)IE1u;ELAp>M@f0 zU3)mtSXPhd%DGPe`Z{s%3z?1n7Whnctz20&oovc&u?tS;rrjT88f>4&YO)m<-=$&h zQx&j9%IV@LFYpW4q8(Ew{A4p545h=;N7izZfwz9JvDs%SCZ|m2b^tJ^#)ea zVmhZXA@7#sqkHX5CEPqX2A`|rj;u@uqnn)Wrez(>T_>-~X-sLw@3gxdINu)r_7Q4D z@Gq0Dc>0s$gY_E+A4b23)_zZF_JaBaEufZ?+_7*L?$?FgZjOK)4)7g)ps+{2&QGex z!FlB?N=(#y0%$$NnnW6;>yHQ&sgfQz*Ww^b&6A$9n{9;|bXYX;H@L$l$00$q-&Q0J zABaW~pT_d{g0rvv17&eXE}*|aY|@i_F-2&G`EH0$wvYL+6|90S+!B^K9CJUm*jXY0x?$T2)1D|$(w zsXu)JMs1f_h_K7_%#zIJ;JeT)(k@Yq45W4tpF%Xyi2Gimi%oWPUgN!WwxB?gdw?Sx zx;a&f;?%j>E^;$K-A3-!6^e{>*OBA~^k+Z_WE^H7NLD!8d7nkPua3R2NXpQVRVM$< z_0O)?#?*zo(`7=x%4>KYkWGaelhtmH6kqq^4Vx*?bezi6cIC65jih2bQ8R^69DFdq zb)~?YnN0m=9)}4t2Ff%-P+r{Q^d+*t|NcAdU_y-;&dtx%?PGrjs9yoe|ALR z5xK>wS*sKEF6|r5b_iBqcmWFuJDH$7Q}c=@(?nmbHMXQyL$COkTw0|mKtGef;@3e( zRaDE^sgYB^xK?4ltIiUVe4nqMO}4}-rE`CfXD*L>D&I+8li#-3P`S6H1{&}#x?C8# z&n8P~kL4@iy?XAHJtgUM9Z}<{5lZYWd>tiY=A)K*zv|26nbq(rPo6d3jC>TU@Od)7 zWx8iVkAD8hhS8_dRXA0W;jXajY3uc)SmF#+e=EYmynMlE~P=p3EPimoVu{; z9q}**K$EBei0>Est0j|V$U>KU`VZ$34cIIMsHKn}z5(ylYLn_*$iw4~ney#Zs^+dr z1`rFP8`j!St*JQICtx(P?C)aOG_I*gQ86H%zUqnXDc)Rej9GeOqfww!$XwE2rrg?o zUrY8!L{TEI#hbGq*^O@~P~v7+^(x*W6j5iY3?2uj=&muAq$8+!R{Bzs&|?q$)sW|% z)#&R9WC+H4logrM`HxtxzqiaxKOl*7J$?QD!E-B=<>Rax4^%v@VqlLh#9O0vihxR_ zvT1y>EWhwuaag_4<$2pjvXm%X;WBxv4CH?ZjIL_Yu3Iw;(aU6 zUP|-AeU0(sHn|>^gJi*xd%J7P|T>BafO)0V5&n&gf#JHBk8+?by<)Si^2--mR-k;&ac- zvxO%^sbY?MO9=B4OXir}bwbMa=U52yLay@JI7g3AEZx&+QqZ`zRNyjs%0dcdAlKfh-md=Z!g?28#)a?%%@co0Ao`(??X31ZUq{c`KT{2ZU>FI z)BD4Z_-^g0LZ4+TR`nZUqm?BXG?L0H1#iVa9SiZ_V!)5UlDuWRa)7fi7?3^bv6n^@ zlPIDYP$VOBfTeX$9W}^9E3K~f)*I&T5W3p@)bzlgqZMArk_FL6Y`A)O1iqbCZE&0;Z3Qcm>;v|iKx+rZ{I~ed{m-xhh{!877@=HMvz3^9> zR2QuanFLpi<4--v6R*y6OUP;BE8;|;(l2X}Qml^*dk#-va>s(I(2!8HG^@<4+o~nQ z#Rk#B=b5#*F@#G69evU&Xa=r^>&bhj4BjytX^rg=nOTLn(A8mPn-$|_LN3|z^$?GVw-HmmQ zwyl|oYfg< zU@yLI8u3Z5!GkgH`dd#`l4d~xTeUb%9OZ*1k$bTn_CuUcy>0!sJSHEureEb%4`Gt& zEV(?1bVW`Qw(p%R%Hiy|OZB=O=}93U)kYZC3S89<($F2d!t5W@tSyjyd6YwQ(Z??k zlJeNfxcW!Y#3#9L=QS?SFgJPE9kppGh(}L*eMlwxFwp;7?KE_JFD|d~n}Wuj4!Z5~ zb(ncqHr_$#oRP!5QD*V7+bW%pmVGy94<`bzWjl3iUr`F-gUzG`MVJeX?XONMSD?Tw z3snp$v4k6R7H6CNZSTZW!`3mgov+sCf=Uyc_uxL`3`a+Nt$m~sDmuwaqQvYF`<*0z z|6>miR#-qv{@Vw3nGd6)i@0wUJR5T=oX(SA3F-JKNBwZ@71QvVv8q7)=tq6q2i_|| zki5c&{8Fz6qy4vc`0|Extly*`=JVaOCRX}7yeli3nO)Bi$*Y|4nJ7)I#-zWo3u%^r zAhT}z_cry}lKDFaf;m-7|4x7YK7xPd!D91weNUME^3Q&L1;5lhND$9jL11RW8k$Uz z!|{*_Z#*QCBd=fjoe@OY`!2(F3cXA-PFzT3?OkgM@9IeV5UC=d$`TW$VmSKdLmtIF zqmK@g6>nCT@9f##f=#Q==QZAAT1zva^*`AE(Z4OLWFj6Sq+Z&?(tBO=X_-SU%FI;? zsus<$UUn;H8L-{avu3}0AiGSLlpN##j&>ulVv>+ns6x5wxn*T^uJ+v_>D4ck@Ne%( zhWVWsR>O^#o0S{zZIMrP`7DTC+H1}8-({iNIe7ZyGwbmwuj+Rnt&B%0kyI9n-w(^? zs6N&sY3oOsF4PMX*J`{C&^$CNT05{Oyj|8u+`QI%$x%=?lCC#N(~Bj*c=3d=FD<=G z(wDciaZPs5q^NhSkwYhQXC;pId{9_n7KIfMlv{i6`ecMj_5pusQcq>1UJ!A&G5@$L z1gCh4*+a8j^_5jbnWs{(r1z;F)Ym>29REQ%br8X%M8lLAtwJx)%Li=-&I@_wzmPd%S;H9BW;1o^zga zj4{W=`;2TdM#6PZf^tYQHtYjl8k=7UFUffbr!Mbx8cpcj-ijA4uHA>$FN){;NOHf> zz;i;wN!CbWo&Jn=KT-3O*Eq}Guo3=IAA2pZQn7)1_SCKn^LVTD0Td!Lv%=TD7C*L=uRpK;`LOI;naeJAg z^#W!GHrYhW3o&5QuDO~_YP{o!ikH~_RlPg&2n{nZSEk^>WqgZ~y)GF(0%*}VqETJ8 zCkjz7X?G;M^Gqf3#G|P^Sg7t1%T~uo8Lt+2lvHyYJ^C!@PJjDsE_on<+!o_yto#fS zU?2d$`Yp{hiLx=lGH+iiDuvCkK6Jj;G>j~AZ1;%MNU#~pV%+viiOFMLNnDNj)e`CE zo~6ws;M3URIKRBk+93ol*TZJfN!z7znmCPwFdVho@2S#r_)68ZErX#oWgyR7pgT5G zyE}H%$?W~fXG5hKU>J0OiP!nj`sBt7?v_(L2iqGZOXw^xzDUB+t_o}Bl|fTD(%id` zHycQN5%IbiUovZS<)mVWpXaKb$BUuAJ|_A%D0_!bD@y z7g;@8%;*1^%BKq1MUA)KJX~tZ_a%H={34C#hwANyU2jE>>nc{$9WFP4YqXFAuQ zJosRt(GkgRYlyjhfASmbVuK@&=lyV-a8Q=w<{o*0^%05ucd@s81i#Gfg zV1l=W@Xp&&`IbGC1kV_oLJEJZU&}>I`#2c(E5l%k3!HBBeHb3H1P-JM6Nmw!30R(I zZ*}UWKXH1lVU4`4wj6*9MkDhjE6YiHYd$Ub9mi6nYVvyYOeUEX1rCLT+V?^2^@~c+ z3{gl^llbHP6*3mRx>(#K(5z-Ml^df|a=L%tw}oQ9ayINCn(E0!e1RG#-|`i}m7ZOn zZDU>=4=26gbGyQP5JlQmxz$Gqn3?& zH>%v#$q?%^1u5Jz#i{>VRC``!q|tPp^iu(9?t`CJ-n&h` zolGdfG5OXpZ2Bzi23c~en~9KUeqEh-Cgb$GCx1eSZlFWC{;WQBq!4d`Moxk&dmsEe zz$;&tR-qB^qeO-*Ttj1%`zGQBy?p29Iim(!z>g9h1JkR&y#O>|B+IV21j(zP?S&Z) zenCun{-1$-8NaeTtqxy z6F?nNl992b)iVJH6kbbU<7wha!F9yFhF%#Wez;I9CdxfgmT&O(4pE^JclLLklC;`i zeP>Dl5EdE4Ee`?sn6V56{?Tk@O<>lJU-3JOZ`O~iK@0K)AE0_;X*fh8iDuDSaK>{! zC>(dJK3j=`DrFST_nf<<*i#fpslj#O_*GIp%yQ7k9r}z z@xw~z1sE8>691F+=Qwz#vfSCHCE|ftEQ*BwvfgPI9%XORHi35rh|u2@>oAZBMmT18 zU46x5G+P5vQbb0&?DNlQH}f+*u8#QOZ_DMtHpiK@o=2;A@4Tk^I$iMr$y7iiEMiyz zH7DF*uz}t#5X=vkz2K9+Z=v;>sgyx`+@60{N7k&0L!rIH)w4H}@mrXok;=c4!5vDk zBFcB+Nso3^xBRAia``aTgM^f(wH{wMx2CFI2UGP=I zgGIh^VA-(ovG9Ws`C$KMSl8L%-lRF5#c`{r{PlJyIhgWLQgw(>`78gG=;mBTQ^SKV zV3TS_CfLNu^77`qItn|1NgY0^`BpbtWd339rQ_agPcVfTvd^iZCx+=%u_o$Zat)5B z@jVmO7q9{K`2oLLE-~2QwGyp0$ZM_N__fW5Ji6yjtChqTQm9{xHK0P3GOG3+UaO@_ z6pd$@F0Q{37>^DZzGnTGFSsQup68(y%O9i#vjV)%|!EkxUx?Y5@(uw1&dvrBz7BFIb z<8$0VZxHx1~brTAa)yEU zDOWpoM0$-hm%*sb?cC~F@HtC*ZkPYVYX+T4EDUPtXPWG@J%#hmzx)$856Nm?+=k+@ z=(koH4nV=4l3NTF??#jKT(C74hI`Q0Dbpvc+}fM1f_OjTt#ne_INS$|$Udx4k}%xr zx^LH_oFj|+A(LOMbxCzyXcT^A*;!^jo^uTwUiztC8OG*7^UM|e zlEC7~fZ%O0s_lAr*ExH9Q2K6z^O5rO^wQ!ynd|uu-BJ>b&I0F9Hdo}I8guE*6Hm0m zm$S0>-&mb@4~!c85exTF?|*Se{`nxbQH)oJ{R5%TM__jki4{JWz4HHou?hFJ^44KB z1#6*7P_NYv3qJbRd>;;LcaC3e&K~b^FPl~yK5FXfS;-PU=8NyDUmYazuIGyM087Z8At|8Aw`@>X{p%|A#fU7z>;44HDPYW?8Mp)gft1|nZQ|@zrD=e_aSDZ};yhGYqT%jgd)7y)W3tFSkMH9&d! zHd#a=r#n308tI{&n-{V-!2IEL=+Pp2(tf*gk&t94{Aj;wD*{Bpc<~j*v^T zP0U)X+()cvZ(PRPLXY|&{%gvBIp@js@D%#&F5sB#W=|{ym!4|S#B4fW4_jrjq=bX} zF-*lU%Tw5`(OqmmHo~_d=1{xkW_wBm&H zaLLPqIob;P&7>&u^3p}u&(ntD(^Wj=i!1)eH{VHXSwILnl8|?iFqXUH9{Y+7O}l<+ z$^yl17eB6&HV5JZ&^zF`q)04MmooeV(bE0hC^;YP2kSq~3n6UOrrxar{GYDygoH5OQgH9EUx1 zy&h|^x)(F?)i<&OhQ$YRS04yD-xIs`=bmsPkntlEJ`jedyhhk?i(Wb!I3d@V#4$iu zU5XdYgV0a{(8SICDUvSH_D5e`&u>2MYyOok)p_=xZ-b!qTdcIPGB|3&h7;E*MUowa ztS$--qv`VQUD2WDKym@c%enUiTf-?O@gJ{LN>ngqSFN!p3st%iIDL-RE$5v4OXh!s zAe2hsG9!PEN)K5!9-wNM6j%^RX?z7cUu{0=KlQaQq<+Nid~CN8n6&& zNH5DPi^ktfSob_OFb_>#xkyH>|@rq4Cl_oxIztj3)E#AH@mef*+N8jxQb+$hEL#PfF=oKBd&HXd?! zx%`#X!4P1Zy-`l)q|hwQ1}NirBN6OWE#4UZgy*vR=$Bzj{&~`IbuSl3z2Ljs`>!#Y zIPFpxe-7gQasX)6pa7xlEcsXQ`&1RTVSXTUKMv=F{hOo)P%a^V0hb%Wv)QxlXeMDJ zN3F5!G(?;6oUnPh3{L1#BDH$*0d?Kx?+(RCP~G9gQ7TJfVH(?mNl;w&GbP6AZ-qm- zX=QQS$Aeq}a!I5aKQ%Z{XYIvcfk`zOu&j>8;cwqR1nOCNXvR@Vq$?~{9LD~8cKw3% zpHwYY6)~(E!znMC#Ru{;GFM{2BHbaFFM6{ZszP6hI#t@JGPLje5?eF2yV)HUlqmkB zg(a=Lx7&{pg?M{E>;24FlPn?7J_$6xz=v&}lcnR0dA*uK0fLohQyQleo__WPmugy# zC18usZ~Y*&GnG?gQdsrnn`VW>bY*Dmmb62dZ{iLk(`g<~V9>%S^1;P+$4?N5^ao zsyL+&g|XTPI-U#y5-Z7zH2w2`aG1%f%9$Mx zR4B4lvF%xC8TI<}v>~iI+VWK7IC0uJ?s529&i@G8S%ImqM7F;DpdLA?QgfHeZb_gP zmMvhNT5~UHgR|=B$(EJHD(ji8tSEs@NF!^ZXZk}Sc1vFwg*&nw19(p#pWf4(tr~7; zGFz_(Zk=N#c)0_s@gL6_bn7&v>$9;5C0uwMZ=|tK;ACrKpe?V96Nj;NbwM+gwv5;m zS^%nkGcz1RJ>J)_>5ERkQTV-cP7yl?H2RudJLfSW~SR;ksbx0pe$d%%6M(mQG(RK=sWa1>eTn=j-d?)y<( z2*T}1*uE?t$%)Ns{L{iKeS~sKcCOG`5q`N`k56l{AZAsK1nlm=ogWR4^or~*?%SJ{ ztBQ4zr22Rd;V!0MPpNgO81Ncx2Tp8V)q<-)^b9!NJm|(US;j1VVcV$KqKDc4@=2$z zg>AG73Y*x|Pbn{ZcQ-@|wWtaVQ0%s!mz8GEyA35mnAp4!!CC85Y*hcE)2wfJ(jW!& zWZ4E#OnXhoT)|BW)2m=HDWPG|$ok^!9;g6ggHz@{mPU=m=1i(}M`e&6Qn^X2EF0Bo zsQBG<+p29dEqdS3*;OHWzY#AYkL$u_o*{31^Ny(J=_mrR`HhoWy;ea?Z_T^0AEy-z7|wF4)#Hu&Cw2qbI+G*cEIwA^wO^zW^EiLIqY%F@6-V%@Ca6}rE1E9z z>vVHFED$gQRmb*M+HqJdDOv7Jdqpc_5QajOlAl;hHelSe!YX!=5=W|i5`3bXs_EX8 zHM!p;J11ORx$w4CA#zn_;c2$A5w_=fh5_9Ic2tceh}FTrkT3u(kj{%vQ?PThH*FY= z$@y@pA^-Fx0jo4j;?1{IczWG$h0S0~Ec& zo@RmcO_WNsa-C6>=klruI(CWf3*7rVrP?BrdM3krhbC@2*ZcH^D~F7J*MdQnp?QPR zTO2G$b!D=4u_%OJ9o>-`JFnA}JAv7pleQ}#X^JS>>w)1tV|+WkjOxuodGqxL%)+eo zB;SzhX|x<^-&H7oy9}elq%o~co^;whHH=54VW~?UPL)5hA33e$n)`Z{;^5vBnF0Tq zy7rmzLNdE`US!AUTklwEC4wqEA}vR{rQh(pnzG5NtyAR1?5N?3&anzq;n=>&u7yb6^>D?a5WwuG3-+TL~3>8vp1m!ph}N0l%4 zUi^%_!N4!$KN{RM;g|ZMMRzc6JY4)42I1@n>Py8;9D!{eFw!m_V84*x#eI2sz#$)z zap5*Q{~!VyNHM3;`TWN6XwmJ1EcxZ;jQ7}1WH&J@e}uThwuyUoRF6iLS&lIaWOtT! zV5!c$hHNOlxo`f&KTBv__Rq=5L`#aucVApu2CKpKQFdXI2Pu?Q`LxOHc}52+2wb%} zs4Foce#Bjm6yi1(eatSTwS32=t7s-x4ydu-&Z@m~a_Np3V^PzjT&vobr9IE2)`T>eA>4!@n zYqnKYwsmk%9}YRM+!6?7wK6@W#;S<@G4!aDHT6ea(9pJkJIz=}pV#iGa=i7)7`EWr z;I8&;Pe^#-&lLz4y&ONP!O3Ee`rJrp&aLzk5}VG{{)s=mY^nSvm%i|}Xhh7l%Wbpjwvo5<+ny@LN6l#!IeDPVPA7UI58dHbEteL^ zHZih-WcRQ(*-aJ}OzQyajOta&*OMxr+yU;AePchYCBC<|0kV*A;omu>?$u%Y2Rb}{ zK=7|Q53u~qfB+&0tN$F zp;7J%DQXJ{LQ{%cjm4Q3%4(y14YQO;4VFsL#Hld5Ac)TZqwNW~XalWu?wH=~U<~3h zukQu#aKobogm5{~asYc-wI2ftnOL>*n%2vgHPT|BxXqrSSJXiIrXzz2iy!Y20FtR> zu7%SRxF;9dE+h=-^5!128ip3!E240$useX`z6x>y+N=Id`z)l@3-bhji^zou)A4}v zgdwgFw7yUCPU1pu=y*8Ec=KYWQ4s@?oroK}z42lovMrAVyvYUGx@U2~ zQOnIOYES5uoPSO|r~P3g>`Qb&LoRWFP8l3Kq^(Ct+(L8J#MDl&Rh3)3RDN-(cD}#J zAQXo^dL8Hci5q8+e!=$U+R5Q)$CuMAmLebNK?}P*b|?0M-|JzaQmsWQQK@A$GI&XI zwa_Bm^JTIBahz3W?`UDfxrM_t*E;p>3+|dDDWI79jrHGRaC^?x*ikaY3t6}h?WUef zegt}UtNWS8#@~eOWj=aYE@zGHS{~=}=Xi!j>U!yXYabQ9hrhV5v`UKXwqp%nGhL2p z0LbqhBtJq>^4b2jV@J2lmdS}$v}I-=T4cAoINM1Y_!Fyy&(~-7TpNHcn2xE}s)nQq zOE{TfyWpO6{RERn*%gC)e;C{P788&xd9dmXFL@Z+BhPCl+r$lp9G+Ly6$!c2i*J01 z7HN-61tWCdA=~9yk7Y2YM*@A55$%A^YLS>3?TvHVLu7VX$EH6q-4W0twl_CE{v}|s zUSMJT_m;P`J6`|_FpWu-dYlUcia0s$d(H)Hz$y#Ga&=YX$uWNGX%==(EN#7Cwfjb- z;T!)XajA4xq%lAZ_CG>S4EU1XxA3&9S_2w7)m8mG5gy4A9+O_1)kk(=BSM6qY1NzO zEyfckNuo_E*Bg)?Z8G;qAD`Gs>xD+_T$vZB#0m$bZ528-G`^O6Ee#JqL0t4BT4H|j zj@82jt$lf%I(6KUdR|OZu2!c7ol`K~6LoAL*0UaNcPZ!85YKXZ<(4qOcsPEV9kS?u z5|Q#_hFUtAy#2z>5@7QK8*i+c0FxJv&RxeH&wGx*drkz6D2f*FH90`Zq&w$brr<1^ zyDt1x_rWZaTZJe^olPB8b*xJK*_AFwFzxfsIae?J3I8pC1wP~@Z547s&9ue&*YP6xqbq;4!{LrvR1?g^400 zhqW)$4pasTN>rvgLomDg9*uxqt60NLg7HeS4Lx3u%D-&6{2Lu8arWhyJ3zu?)6Sz2 z2tlZ%5{oWVuH_dy1vJ_y;Z%emD|qr;igAL55^b3by6K>~QuT&(UliGvAG-vf$vLd& z4@=N=bS(vp1)xE4L?OGIb<{W?)ji1(QBA60Fz)xWP5Y#OZ(CP-BC7r4O6{u2dFoHm zwQ-%>tZBbq&@TNhF8D5jDm~{k5ULyb5}w$9$DRAM4FS!?Y_{qSZ8F}FnGIV`jB_50 z&fNpj761VGRpZFQlP0dt!Kb-{Z{$?pI~5}_ks5UQR9>)NfO}%l(j$diw%IM0=)N~p zId%_XkLnB1YWiPoabIo`?Ry|-(@@H?BN3Z6W?rRi86%bO=I9!@>XOL~`G>@!m`cRa zr7gk90Z;`b0%q7q+*j_?!4A&qEA9EqH06dJyv;XK%)&<$3w8Vj$ghy8z<|DMBn;b= zz^nf@VN9(gRS*;P!j|l!rIg$Hc4cKR%bsyo@m90u;Dg)6Uc4cY(ug_26-=bR@%*)I zE2FM&^4Le-W^d;kjHb~Dx%8KX^-3aJII$UJ)^B(L7XlFQKP_30+}xhZi|APP9oN&& zKHJ1blEEwP_*W$+18kj(yin9nm~Hy22@>OB4~;en9oz|VwFs|9^!E4nH*TaM4};B|TmE|IYhtf?v*K?!8|h(kuN!qGQeLus)hS26+k0uTR6jUE zqD~Uhe$S*`5uU>aFvO2YUQwL(myI3l7Sk#jDeN9!$*JBtmtK#trk@7> zppAOsq>X@tuz_Mb`M>h%7lB^%U;$Amo_}rJ*FN$zHh19}1r#0x8S-TVlZX$0c%JU) z#!ug=0oAvPMBHmKtvXk1*|Zzn5ju8Wj~k^t&1!vwr+$Gsn&B}+`b_PhA#w(Qmy)kn zLg-DSrx6LqgBsDGW6-V8h#xJJlPgeYDkOXSivXi5mWUCO2b=|iLHyuf%U;h^iZxJx zQFQN5Hn_aDU=lzI0a>*q@8X#9FQpP_;p*(SeTO~VpD}v170-T$;-?LR$i2_^9VDC< z0?a_Vz)USK&v5U^?(T0defK5{;rT;XGu^xT>6QzY13TpMq(W;~9*Q6A8LLbtRVbz6 zXvvg%42I(Qcz8eEz+A4f$fUC_%8LLNDVaKpb95FxLyO*1Vw0Y+Vko3?>00_A#_;wX zmUXQ6h!%fwfO#~-fD}lXZz9mlMQgYD8qRL>=S$nQPIS{IKIh$*Tb(Kjxx>`gZd+`^ z_f__1Q;fYyJT35Ls`)q$M@B8OBCQSlH`hn=4KF5&6=XUySCH*CUuAxNQ2b7?F>w1_ zB#ML>7`l!poy6W@qm9KN|Lm(QN&Z-D4>lZegF}#LSVFLSPZSAEePEujxsA6FULDdC ztO!Ed>M&=<_%+*r3pB)L4GEW}FJ@jczZKE2o(DV3l)d&``=~H=SwcwmR_FUz{3}%j ztyo}IogW`hP2YzOoPbZ!ySJ%@mRRa=Ws5N`C~kW@I>i9t7Bss{P^A2w`;OZ#T;i<9 zmn-_>mVvH`Dz<4(bir$bDP{PCDfB&c|i3otYfcY;p7z~c)8h0j7O{bsm0;VH;gFZcE zzT4uLQZ4p(I^Ygj*?mtV^x*zU`E%;0)8~kbt6unHZJ~x?$}TBR#3yy%-~GbuAebx? zOkhard;nxSdabQwdrM#VovFcBmERhNc!BZLKJKug!Ia8$s=k8HZ@8rnU&|iZJ6X)b z0AnJnq{FgVcgCkZi2;FsYhZQJg#7@kAI?rVVJHO@&1LWezy;-h%u8r z`ruodjNItHz5*DBy*Xh%jvW#@zwiRxTIVBh=7n=tB5Mf%U6qZV)3pHU;O5PnnyHdx zV1g}=r!U0jZ~+#N)qyjW&sk?;`iCQu$!O!ZLP8pJfv$|AZ;+oZ%eY-Es=#y^V6xSV za;rsGwQARrch-mE%Kp!Xw;1O$Jv*rBez;cmb8huQEw9aV^lZODm+jGonkVf^VA~pq z8otbxjBQ=*5dUzIVZH*G0+TcH)p!!KOuSxtAE4b_%Tu;G%YB>QWGM&g6zE zxy1-F&~7QjBUfte)5%aGcnrFOv)UUE0KJl{_sj1AXV=5S<3k?^2~e-Q(6yZR+SV>)fDNHHdahI!tB+=|ubJNq0xX+% z;KhF7{R%HCi0>s+NGOQ?PPKJ15?H?;4W#elXz7NOQ6uJ0v1va_fV~bGY##}r8J!Su z=;+ctGCVF{ArteWHAB|yJ3ftnN5*3l;pKVbnFwVholyv2F_UI*p&0dqs|({RmQC~5 zcAMZ-lb>nzLF;#iBgE1~`Z*kVMzx0cjx_suq<$e>7Vn9~&pVCD$9e$U0FDqQ!B-OF z-IwOV)FV-Dzp?xtXx9Fsl8j|4fr=J71E7rnLXtta%(VtRS?4xPl6f|jXHP?`Rgn2E zC^kAe@17${Dv?RJ41DQ~x~7n8vvPfXId4sB2G|g1o-X6CZrH;^SvlQK+h^VWlNJy|h2ajm-bxK4-04?s15dy?3nP&0*o z%?|M(NLeqWEnkTeJ-(PW%`~)Mx8^60@#8;PoAbIIy=5lI0sM^S1z)~2`XlElr$i%Z zAq?0&vDf||avn;cw+&25Ylr`@zuWi-xZDe3GMZMG+cqwk>$5Ci1u{;~= zA#D@H)o+%<#|Fp#A@6(5cO_Ek0|bD^GF7~tY^CpL>2zs-==+&UPkgl;SL~hHrXoT2 zHJSM|gu#s$7C_l)ruqowr$>i~m-`)Kua{G`#>7Jnelbs8Vv7O_R;F{N*&`n#B_-fE zZYtI-?(`EqQ66jWuFxkpiIEmIg^&HY2xt^y4%oYe<&2bvZW#QT*aFHm=R4%0h-c?& zw9FxG{c(@@BV+sr8ud{3u0>9@^#1wKKxQ*Xn^641)NW55oKrQ%h3UY9(XLx-qrf|# z`$Z+eaB1w#HyhBY9hh+ugbs*5(%3o`@h*wpds`9 zJYQ2JA778F?pV2?_6(pq8^3KW1m!rzntT_``C4u{?=MhUEti52Qzvj5l6bx?j1VcrV&;U(;1pt9b)8D0ZUz4lFBk zlj}|sNz5*7m-QN>pWS>ku(7>Fz4|AWLY*{d3h30WF)jm_+f8O(hb2}6PN@cd9 z<3W}(V7@Zq<^kx0LJ@-KnoT0GN`{;!EmEhd#d9i<+n3MZD#Di;Q%1THzAB}Xt<7I1 zb^uiQ$rJJoRvxKwj_Z=oWy)5lfu@0BBs4M{is#zg1*OLQe27~nzMg*xY$5f{JzKO9lYF0lP$S#s+!d8 zgzR2MFsf}RCe4p8v>&ManYF({8=^;+D%-J<33De0zrkPT$Z5i?8?eC`;IJB#mWGef zQ(~6+__YsxY*O?&AW^8L@y};%9!S-e5pnp~D=^ij`KJVd=Yd|mXi%6zU-=)Eiucu% zfPzn$$>(1%Arm_MFu%!X(huCBB{VYWMxPic;b@h!IUPDt{pzWtiaIHc^6h@Ey{l~1 zDbukm2uY|ZW*F|T?CpQ0Zl3@RkH3V!U#eY1?I47dO26chS%cC3{sw3s>6o#_^noF+ zr$BdGtgrU(I`l`t2&UDlC$ZZY%J2NJ|DZ)yY`%NlTZWW85a+a5$nLR2%=^&=E$kAQ zUhYG(SPw5Xk>^RrT)01(Q^YCObXa4(ycV>mQ0{aO_xwFKjW&<72#cE-t$e|k-m%}3 z3WgEHB8H%1m*FIiWm1W^KrAcC-h&u#2!-2k|phqww zHk~rMC&u#UNuEXZ>Il3l^KnGFM=kUx-6SPVSbi+c91YISwV%OAld*MGu1QU!5tUI^ zWv|@>Ow|Ze$*=y?T?(fQGv!s|Dih01Udsk6Wyk4D#(ro(iUyrdA^r$eU-lqmeiBKf zK1fsXhO7P|ifHLYOtmCQ}cB*AM!YUMc*Zq^1-p*c`a0>w{2zlj;H|&hEIOS&0AG2qaVWsJR@IW@xvFYt-eN>M+ULEPY8Fvq0GR?04gM zQ9^2YuDui`{c{)A^pP{3GdI^KWT!yqJG(}}(hIh9zF?KcMaP?uRQX`cV)8xxVSB3GSj(KO3RrYs%Om-;=53e!tckqKrN^ z=ZK@#i)&a@+#tKF)~+qX ztMqLuhdPsbu=QvoK2|F>D&W{n>(@0!Sjr3y{Q>J?i$Zl{9EqA5P|F}Fg7QXbv@O)G z^3@~*+rJ%=KZn_%mc>Jw{SFIO6>l5HTViAPYua?Imy{(i zD})&dhe=UygG}$HoXk&eV@H8=RgVMk5y4Uosnk5mchS%b8oIS&3(j0d#S)LCjvn7= z^+&**doJx|O7zB~+X9aRyu&9XX}a0buNG*q!oD?jimRV_Kl|x^#r^Fj`f};75(M1@H;u6L(n`&DAA|!u~L?5 zJ&8+T`2|-GXpT`rznw=-m!`^K7{?fv0LXFZM#zYUI`EPhLk2ZiETa zC{6FYSI8DYALs$hGiKFsu(vJel?AO_e&kTKYd8`F6QS%)qsP)L^jb;0^X0ptl{*}l zd(ZQ6xk8GAih(<{-OZT1`|4BFSY0kS^o9v?6CwgwOz(3D#gXv2 zW(K%6;V+ROsmoI<@Ycm;mzz;gLA?IKoU^o|rypf+ zo1>}ahS`%Xgs$+kmlblv9P2`HnRS9IV2Pc1vC_uBPsXw&DhG6K;$P~`z>-FBzU*Oi zIJ8=-N1^A0twH~)P3*F8Oi7vX) z)%F^n7@!cVDaKaFcIN6-CuUeEKBs;&Z zl5T4?U?b4GC>5~!Q&e4SC&UVamMQ@808*7c;ysCItnPYe$U1Ep*&X90)B|TXsh9)F zXG%%(Zc`)r(?Dq)p7T;d+>&W2@r73m#of@AI!mm-SyFEUXi&Q|MNi2UQ($-Fux&d7 zAN?-=#oh(IOIQn4Z`NxR(H-Xm+L)Z+Wp0y35aFOn&s_(iXWPBBe=e~d777s6*o<;o!Tc>Op*%jHwPz8eFa9v{@v}aa zC&pCK;8aORA&M3%@Gn`y-qeRZq4*RIh}(*P^iC?VvF zj2x?nNuFYnfD1r*L5^*3YEk|?x0P@KfmSIOpM*N$@t7_djgnt?0b#Grr-){njLQnF zwty4I;v=iS(WDmpQ+Sggb*(bdcdU3_{JH2a@g4`lU1BpKU)ypth}R`kg_ugIicUL@ zHpW)!ks?%?X!BiUmEG(6d~A!Q#!nH~I_sAaZg#r_L+|Vdauir!1Xb#CWtyY@GFnq2JJ-Ye5<@@#GZ1dN5hZ6 z*GoVkU+JJmdlBHHC=JmAA=IQlo!5^~Fd)qC9Ip2_Zy~qKK2MD(B5*iOk0c58LhX}0xEU@D*4cy(suFY z_ofwi4FkSB5|c{@!#>VG_zJ-gnRmAuVP!WD; z!ToDNgV~<$d}Ii(Ch?yd2=W2J{TIK9XE{wE;Sda%6&mP{@7esABLJj7p~cgAKwQ%q z6b`gAghb(i@ZW)n{&{A<}@cseHkSn_nT2JaPJu&Z+wjg}7(%&u9E7(3!*A1gj zJxFbSLB7N5@K-A;xl+gM?SxV~A>{eqW%{`_;XGG&FbkhN7Fn6H88Il`&hR@csIX2> z)_eS55HQrfZVLSS7k#1cmK0q{og=?&QOzAk>7Fz{c~ggO8l)aWL7o@ zRF8>Dv?_nMZP!YX!fDJ!5WEd-nC0aA^Mo2-K!f(3a~iG=+j4@UU`F5kXj#oZ$bDZA zR%i)KwMEkWI(udEM$6?>KdUU4Mw zHve-^K4@geES?-=jBTLr2Z8%P!rCfNRp8~Y-7(s5Sw(uO_Vfpi+sL9eoI;WURn1wF=rV#Q(m>R&iP2L*y{LC z=RDww6?O1p@K`uOv;Cj9`7aUA$~x8Y<$-1u@*@V&EPZi8LJ7hLzLW$r7=sE?M4bK@ zokHxq7Wwa_>bZt7)h|l>3SPlzoUeW;EKVGl>;!Y*X3jlmN(J{ z)u}~&A7>e`zh;h48hd+{c}%{9Rzo2U~iw(grWY|o&FD{)PJ6Mmc)v|wbl7X-}~dM zs~5?|j_hN4oR4F31nwK~F=Ws%+HD)-1-e2%IAIwjK`wW=TCUgD(VT=cE7gOpUAeh2 z;DDdes?tOGS&~$IZsx1n zlB3AVP|mc?{h<{4z51958nX(exFpwt6cDnmFi4I6q>eu)mm(N&0*FnUPe=B*0`)&X z@xLx(;O;)iJl&SV#ok-3x*hD!aQx?yjdM`A&wUZ6C1D=5fAU6?DrS?u&K3=SX-h!^ zk0A>crNi&-T{G<0*;LNw=OM@5L7$?n^_50~=oQr+S`H_#+a-~zTnIG4zfI&Y-dD|y z*;kp!F+o3sp;Ocwau>Beg^`6iB!AJnbSrsjrn$lBWExc_DnY;4 ze*K)ae{H=4eSC6_3_ve^j0v(xTfL+(fb$MQ(R8sNG>{Zd^hJF;{}eo}%Fuv_?7}W^ zN(M#rd-u5C%#n9%?%SFVzv#Ts=bOwB3*tC-IC}+EUOL4BQF0NcyYHWc3`M>2$|q{` zg^XNO#@T;qa{u%8|6XK&KgJmZcjV;=Kp+QKW3?CYp%mxR~l6y$O_UV&=#F!uveN45n2*Kt?WM-MDIrNZnn zEr3_M_B%V@nS8O8iI%Pf%!K=}R?E_6h){DPm005KM9dv_nJ1G12bejos%QeX+DNdT z42S>?ihb>RwfC~xOvSkfS@stJRI>Le(33}ZlB_(il43np=yCb`W){dYkyc+!y+c4) z6GNZXZK-IRg_mo3~&(-V>bw)QQeG>hK_^;yUVoQq2m# z?1}+I@QmLR@%s2*5AlBpm;QcIz|W$Q!+-;i#DFJ-kucro>l%*7(jJfY0JNt~J3}}{B7-*60go$_Kg|-*v6`!){gy4}h+{Z+z%{M~+ z&INHqW?0e|5){rPd~VNGty=uQjFS6WBTjdoT9^BpJL|N4$F)}Y0K(1|kDv`hnmhJ=`@|T@qWrlanl>^fK0PDgm^)(R;yW7H;o5HZUu#7k zxN#8MWLx?FntcB%)&I51T7k;r#+f2z5YE#xvGb%S3>OzbB(O^TkFA#)eS%OyA(z(> zcQ|x%5gV6MK#_9<2s*d{+?iRo$@js-ts@HdjHj5UQQQR}H`Y_$8Z#jiUlKxY+!29> zK)B9y?Ox2x+cM#NtMZS#tQ8n-M>(DWveGR`}qx(?Sdt?{VJ zzEnOV6up&Guvzw6QdYBX13^IoD ziHN}Af{q(Vd&aHw4l0IXwufMRYB^A0zv6Cet>GMn(eZ6#Sfv#^KVD!4OURC0--qz2G`%>f;R3l~6R# zM4N~G0kR0zRm8n)1W+@o5g`>jQ^>z`BmeK+nUS-2}R zdE1qRr1N?gio!8R&%2H(eapCa8Y_cAl+WSZV|WP|cCWD`VV`|1Za^_m;lyb#>tN4+ zT|HYU{n9=S2(Q}4U9bKX8Gyqs>ir9uvX|*^{+!kS&yV@9PiUXEK@QLl+6Gv=k-DCn zy!%M91hl|Sd7Y3yTkDPh4wh6(zXnOffyLUK^GFob`Q@sJO*boVT4JmCDX6$yE?C{O zi-qH%VLnQUU+Xq`GCHm65t55XBA1QW-NocQ!uxmg4BH)nNh4Am*Wz*l9d$x3+gJ-bnl0)F07(K;XW(p| zB7VTN=6xG*Mjqy=-UqDD;9agU6oC_1Uo6#oQzC$jy|90{gFU`#c@8L^0*Ar(O45bp z@2tB3G5YTKV>_P(s(LrTa~YOn!YpuQTpcc}#=hR+z8Rs>SF_!G4Fv4=KyP{>$ZD8X z>n}!UQNK_R5N_batrDbeMFLl7^* zr~~C<#e?{0Kpg-|%)Iz3GNa>hM2MHm1496@Y6IN3ntzNRdh^(PjC?ax5BCstFnq%z z!|&IKzt%eFRt8juNM$S{e|rH`>Nek|7Z&NK!yFq>H`IdXrz_$vT+~6{}q4#Z|5Ib1oxfg2&;wK5b`rzG

VESh=7MsAFD@lZ5` zF_B>yTk=`Pp8XGNV&PEHLAT67?TOwZ%~P#-{%ja>0$RL z`mGATD1-pkYDgzyq3tUbe_C|0BILw|+1YQ#J6rE#k9%grg$j*lh#_qq`M>(silqFham5P-CxZYzfVIp zBEJ{qsI4vM|6`VMg8`o(i9N~MILekYt1md)$GC`_D99S$u>M_aQ~y(J4=H3a`Fll7 z%>XJ5GlyKxu<`e|rZcBMS3wru(0n2-laclzapX zAb^coX*WZNS>g7@aP(D;iJZzzMHqbf^7&4i5A>Na{A&A94*Y8R)Xv`Ci@ljr7WILv zb@yr%-Iy(UV2DTr1+anj2ZO@-Rj2>)whVC@%sPGikXZH6oLdib*5;5^9ema{t2cH7 zf4y1L5wA#f{44w!)xuS25M{|WZ!ti15+abNmELXfy2H{J?G0gdy1A65S{mf53h@QJ z@b--(@iF5GBSqpXV11pVm(oIw7K`rlyGe ztnTqNu2JC4XIP!7!JuV6_*LOZ`ow=ayf6OR-MYRVBmG|rx%V$2dwP0)ta%QAjL4*7 zwV^YesVEtAZv4*23hMb6CnvVfS}%ZMMk{`wi!99xne0ve7?6Ge97L*7} zEgOJ<`*SCj@}2(d$4@T&xpbZKX)MsMkO@di(6(}%qlJEYZ86Ykf-w%4>$KkHpHM;65C9S- z1_6+yvx#UuLn4AXrX19f{VAzgciYy_{c!9~`CqnaZpNQkR2%mpA3z@Qq+E1|Hxyv! z8?5REF6vISo(Kf3z&zIVXzt`x0yUxgD3s1fb@vy>R&BJQfuk|=YPE{bx0^sv?mI|f zlCKrX#GkdNK38a4(U`&tp$t7;@ghJ!H41|FXYTdG84nZklEgMcs~74fPpj67nC)2Mac^vulbt5TqLxgJ2r zCD~7a@~WY78!fkr;?ZDIOag+OPke2lrfv=ncSaw(*{L!Pi(#X=!V`+Dm#M425>WiD z84-{6Jx}n`)<_1)uaV4l;+J!^=(NT%rYq%jC|Q$j8p^!($Il9tt9V3xZg4E7N|9RK z9I$QAO-2@P#&Xp3>WI=%JO^@Zj}5*-(X^g};!qy%_M76j(@J;bGr1&X8;{RgUP8lR z#;W-8DRxM~v?I}1dI4O@=nS0r3S|vh7-Eu#Dzev~t3gY*Rk}7nJ@PEkv;5)=AR{MR z)Pq|dZn4S=Jy*)8!Lk&7Rq|!K<%y6)_*Pn4ldLR0I|v{g2XV-cQo^r7^F0D!Fi@&@ z%>P*BlmgWF{^08&^Nemsy;nw%ro?qCH1uAyJtj|&1pIC}3&q=z?yInSECy|4?=Mtz zl!?#j`$h$&PgTq9N13HWh|q+t^9}Al&Pd*!s85+7X-!cgy5Gvtd7XDMZ?kN`EjB_1 z(h!oLvCu@n#EY+uL^OnYsuz&mZbag~&Af^C4}X^!<}YEJ{l5u2QUj*l@#0Hu&|@9f zI8YNZkhEK>=mFqb374mFes2RL%T}f-Fo@;z)!Z(bHjxo%AwF5e1ypZ_Z@oLmz5~bx zl}2ltm|H`W)ec(fb@Lgj6?TuEqGNSh^3n02c{fLEKPttqI3dNFtyn;%96b;xdMD6R z`FQ|Rb;Gd2vqKKRJHc_E%%|U6H|Lj(Ay>elZ;NfQJ$t)8fKFAt0{(>P{@7F@C7m&D z(DMzjP=elheD&6_U;c}YmX$$r#NQhsoctwTISHdS^(=$yu_48f7=6ZK45Qw1iN^8C zcIghm!t&IC0U80*Zv&mHhDhr-Xp~iH;o}~PFzzq+-U)+q2*jUOc0|x$J7xw7AlcMc z1^(nGq@nt{t@VPBNCiLpLmTwlb-E3Y2O}jBRJL6Lr=-KNRu9la&qYS{+I(;n$ppi2 zP)SVYXGzRaJ!RWjqL`}2tMBfuoIW0RrhbJbprHuFgGPskREo}g8o{~?Atk>aeRQyF zP3~k|km7b>`i_dVdLdH4*zP?nPT0|=H2Sk}g_#$Q>mlt;F=pkAC@LL@{U_`c0aj-q zjV?plZ&2|glnhCrZ52DS9k>M-g3LW9jy zT)^}IzoWOecXfwJOPb&97Dpt)68rgp3=mr)Ykg;TYuI-uOjvbPjN*CThx(KaqS6Dz zu8EzNea*dA4iBPO%(5Ck1I2`&O2eZb?cpoKcpcum%+|Rv-g{izG&(F05%RbbM_gW_rFvQnrJM8}jve>HS?sMF#d(+Sj^M^5JgiWx9~i2f?} zq}LPW093MmXLibhs@kk z3!+K*CkcPl~Wf65n7evLxj4Jk>z(f%V!GIQo78Hpim5P1q3qv!YC*Vp6 z3rdKJB78hLFlM()Z5hnVO1#GBUmPBhTR0jTDNkH+I~y*yVELpGG%e&g%2WH$inqQs z%rK_Zd7X!CI(3LYKR&=tt271lXQVDziEQ%(b=tf&ck#?noXnS#_3k; zXni~Urd;56Y0sXc{I5qJici|3Q;yx%sL}_0w3|C@7SXlCw;*{LncTj<)b-UJ!Ht0g z2M#_}lKUNWuY!IpfDaiG5ivc%Oa{v2wcz4Caki*g9)ep%kBu>12H)hf+t&zjqXHE# zU73!S_Nejr9qm)l^yu^iZNdo)^0?7EwJJo*sJKDLp?o)C-#Y|^nyQVqdk^>vWlc{Y z4@ZpVGgT$YZ6;SpG^iLNz*B*=8tO)O7v@Y^;Tc^>4S3nU?r` zQqQ6ujpC2(Qj#{et};etVeWK`&dJ6Vt)&F~bCoEq50>Wq!!kjJ@XujZYoN(!p8poJ z&pGTmppPcoh1UO9!2>sZ_7{+Yv?!XI;&k`)!~lrM#%QK|%lhC6HZ_yv6@7q*^I+y=mb}@N^)jp(Np?e@{H`?6rlYt?fyQ z^U}gXs#F49wsB4H7CFhi(~3=&k6%E_LxoPsE8cl9XFQEcOgP&8PM+GUlr3eFdv5fV zJkd;s=@Ed$n8f$6ZUBhe#c7u%rTt^HW73k)_1PBVSc=UIu58{75{>TSIa9kqZ{&}y zo5UfW-Z?UGK{wfbKzQm-{Byrr9;Vljt6n}U6Q~lD$FuIx!V^E_BnL@y28M{@ndSwV zJNt2W^V`qp)%7i(zXci-l9BeHOk{x2D0~ZnevO(uHA2@z5tN`~JV$r$XE287&jGfy znb1WpLvJdjN=_E3E?QonmgEoXecR?9AzN6|fTJIcRPnSKk&j$Ca`S!Cj8P90Zjk$I&ilKwCf9BXSRNvl7l)u@TBE#6rEm%%(Vns#siyi{WeQIU~ zduy8-?S`yyb6ghJ{=);5TUlo1J}ChYHkr!g&&zJv?DjVmD^BTRp00{tQ-U7N*hmYt zzD1_|pq9|x*C8UggAstq^q^BjhXKC!KMoFxs7bFedskZySSVl({V&H)roS{uiCRTf z!-d4@<%oP+Tbo=3Hyd>Tn@DNatf7wcCQD5y zirX}D5Ygqlgl3& z&Gh!trKY9H5O6uL4jMP0EL-;Z_iYTW{y6q|YdjLZQjrI=icaZ8OTJyUzx~|CuxWET z=2v??=|ArROo}PByU8fE*Ryd;4M%Fbm&!KGR#);1h06UghE`WV>AtAQ$cE%m4f4Nd z!id<5EkM08NfGmsfPc(*F50$`+HzulVfEo^^;pCJOQ*21lqq=|w_(V*U_4xz@Pj79 zR1!-y>I_!yU8pm~37fwOxxp5r@G-2iT_oB2$_3=HR>p!iG(cVnqPu9NLe3|*%Wk{W zT<*Hq6&A^KM1uaUrrm`D`CC^?*+6OWh0wL?2;ZRH#PCv)=g>;LLxJMt9|qYY6=Dw4 zmg$;p^;u~&l-bGOp`-SrJ?R7xvcKTpyMaK50~99!#?Dk=aQ?aY`+xu2Ng8V3ju$&x z^bOF3qu-C$0h?z1X#WE>182bW%Jr82F)C_wTwI)ppkQ};$c0@DN}Kqo$#!Dv9m}## zvwdz+EM?06)!{>Y^OX#yGho8{1tKXeIHnJQ4>H|-^V|93p@W-AZ4U#_%;xWyhtDAS zD7swZg|`02bXv@wJclBVl6}b~&mq_=Y*05}FIObjeXrELD2DDf=XVzx^7_hQ zt;Ae+CH#z%p=Sb=M`G{Y+ED_uAkxJfE9Y>qK26#=w+6f3cO3$1k5(zZIRh-Oc8s435GM*Tec+_wK6}?CyB2x|GGO zn!P1cxSp#I+XJClEW<_%8?{tLEh4Fx9&)Ok8Y#Njehhr=))7@|Tgbwu--68td48UeYpZjej{C}KGkVH*F ztnKeEt+?zqOC94$LD+)eoMRi9Rnc0yuO_hHzHP{UYiQ-$1Qh!|=NZsAc5c(?e0w?m zu(^bdfZH{UPpkTeOkxlXNZydJ)fE_s?G3PVzm2or3ohs5C~Aa6dss+4%46sT?RMGt zT5VgOxIf&tbBUUK`!%tbR*SXCUvqLOG|;QP0H_K>yAAAUDG$R)xjWSo==6$8g4=Qb z$0;rm4Vh$3{X(r7x@%*-BAeJWpPc0HgSA6Qet)=MEI zL_{`^TWg+V^2?NRkbbBHKv#vIS38l!FIRw~9gVhr-=!@T6l}4&?zr|-)2`P#AcY16*2Wo=lpk6> z@q>W|kL=xB(tKJX$gG|M+^jAGZnZzx;UKfr(d zMR;m*Z~NmZIwq~^5YOY`9U==*WcHv@YD22lWzA`m6IH&Q?HB!gI|P}z%&`oQ#c_yA z^RhcJ|JZH&rw_~L4q9Wv;5SC2Cu?tIYEKaKB3e@-iMT_`MMcIcKiQbxx3o;lUXe|u(am4Lmn}9XAdPW>;u+9`Srwzh9OCvoeXnv9$AXQt{&G4#aay!vdt?1|6=6!f2byL}#zm)1Gf8q5K6mM2yu7hn3mFh0DAS;idCV zuI~AxDY7*tqm!rg7Iz%OY43;WGLui@mXHqH8nQKdBNy0}0GoT*K;&t}iP=%SQTW)y ze(~EOqI5!-)5hWc%uWUCYeDobm_Kv)-u>~+0D6ioULC} zJy4XZOou3AfhqBHW3a%uf7)5Khgpl;ci`zWJ7fVf=jiQ==oih;N!*ce$n8=8_LzvB zQ2OZkO*bRPJ5>J5D*PY*+0iae+yD^mX=qY9ySk`>XkiKmClIVR%X+KC)J@qvZuK~d zH_}a%^Dy30+CP7?)Q0@w^P{=r=*HD?KKMS9HRz{?$TE<`l;6=IK%>wxngZS1l;llD z;?B@!sEzkD_UoEZy$s(!h{&Lw^f24o81anS9SE3o`3u5iBBCfl{}E545b!ElI|Fs3 z$bT;W5AXaR3lU&BH|-pWWoLgP*PZQt;r7iJ`i)*~Iy0pK@?=|~mE#Bs>#>DcyodQ5 z?-td`<}lT;aI5n@gGMzd$5LW6@5#~3dl5Pa^`G7b<;E8%RCl- z_wSqjz5nS0!Sh)>cgp$xU4#p0YQ+ZP<6B-|TU$H*Y+QiC_x5dozP^54{a^yUa7IQ3 z!_^C1yVrneW7o^^i{?L{NKE$hN-*^6(0}|NI+Q+HepBP;FGBukpWkZeznpYudtAFp*%eq>LO*DLIw0cpgr7d zJwN+;*w(_};_95#8}xg;*ZH7OI=9;1cVqhe{iH@}pT$eR{8W9hhCjpf?-XCh0qlAY zpz_;2)2$bS#%5b+0!EQYz%;&$+Knif)E~Yd^b!%g{Qa%ZQThmh2bv&@|A6|lKzGpk z!ud_}-NyU=SGyg!Au2$?`k`ff%3_9!iS$E@F0jp4>O|9!Q?17;l{buIq zyF54;dVlXu4Xh%^*FZ5EI)di9mBXf#wkw65`)`h;6qJ-o_UH6c5)xx0UFcvP8Hqom z;G-@7Oz2vh|9hjr@44ng5|s+LFW5cXumAUDvHlXO>ETFBaeN%ZlT#Z%gvaHy(!|;3 z^zw2fAv-ucd)D}eFuO( z`I#*^^T!DLcT)dJL~*Ql-J+i5H6x(x`E3&O&}zI6jS8M(Mr6YD;17)660<(bGSDJ>u|F)`HC)X{?!r%o{% z(}4Kkuo+~a;X2^Ku+XT#<$!V>iRO*+$j^1$X!-X=AFqrKubtP!UKtFUf28=2N&PRl zg-@FB&-QDlrndS)?3yJv*dsOT~o|6&#|HSu8`3G3vG*l-mwzS6ovr(dtjbU zh!KJDnj{v4`p3)SLIGUAkmY+vpOxZ9V1 ze{B>k?9bQ3+MK10RR6s3e?ISf_Z-+?yZRLVRZJ64b_yaI`JYMKkxt?b3@d`#54gV{ z^pA1;FK_+-59|NiusW1K*(mVX3?JsIjSdk^pT_tjT99<3tQr}5k+OGUasUiWBm8sE zH)C_swiw!^?Jf5G`@MCABBy<^fp9l2NoGOb0&?U(N3Jv+kH5F35MU_&cXiy+F8a)i z{kEYmN9P}t!DmqfQ0j{;DUDdauQt5>Y0Yo?mYiHcNdN2L5SWLze~{Kfe_q5I;5U}r zACIjlS1j|^o(|^4vD@gDA-i2YNxG^hOjQ;a3K>FUa-fV29K(r3M@HCQscVa;$K4>t?ZDGuc#l zx8oZ02M5bWg+_itY(T2==Y4lN2d}G_6cp$V*4iMkE4uVm^B0 z0Vv>9+orz4JoVVuj4}#gLgTrv2A|Ef*ORvew2FV-5s^U9FVTpXlS?3B;f)vginALas?+GZzLE15zB3%q8V zcLk&`bP0r3Dpm?vola*-lyV{Rbw`H>_s4E85%rvP_cnv9DGJ3W^$(Yg8g}&VY^muX;BpLZ`22pNL#are2?|Q-WNYFYF zyLj#AP-XKT!rZNnmhoPTcI=yJInn*j%`nY)hA|1gCo~<@%hB zq_BE&m6A@C>{}2*Qllw*2@T%sfT#WE>2kH$swdj#2pUY?wgd?)nr@vDkFtoVMb;eurHBnMYcpLlx;2bt5J+YwPC6`FCxp^yNtdNm#3W`fR`lj`W#y0!J_nte z!8Jx@!KoeTBj-qq&xW|@d_o`r!)Mt^V4alndw08`1vnsmT^@b`naU6sp;U}rZ4R_O zVQF!5l|?&hIYi0##1tC*;P`oa=53U&8U8xb(yiKE7SoNfvQe?Nrk5k6B>hX(38Wf@ zkL+qL-5IHgL&UMB>9j3tWN^{7qIsclmY|e7Z9D8+Gw+Pq{zNF7(DBJ0bc7$e8M7hWbS-Ux9u_o$1UUK9(2oB};$Mv*O2A6xqm012Mflr{h$4e+TtH4KMrm6K^`Iv1Xqx$E zu2P~98;EC;{18v8W(pzl`rvnqSP0ZwH11on=1?T@n%x#g(&(T0(L)PgtmuWRG8&u{8W=0!{0OPEyX&yL`e8O6!R zbldm|3w5HP)akYuRxhn4c=k7qblMhUVMGXei6~F%=~^`RAF{AEs&wQQ@y5LXlGcm0 z0bskx6i!ZmQx0ZazjT`-=%uJxOP>1@ffE$HN`+-_^KfRIQd}ls|x@MSp{8?TS$dKzZO$wj7T|Y(;`%|E$^^GNqM1* z?ar0NfsBKviTbxABrzZ?Fr5>D7dY>BTNyz?U08^Hz>SyZ63mY9YPahBUO^Xaic3w- z!=9p~<>GOJOE{^rSDA%mv2ITw=`43E4rEw2bs%faPhVM~XWuVJf~m%;w}m4ee#XiF zJwTR)AsRT_I2jioe#E^HfU$|{VEI)MWmizoChB zXJ9My%clWB%zCkCl=9Ep2XKV0y)fIQ!-Zjv(DGT5^F4;4*2}1&*4Z_@wWn3ROW`&G z?Ge735Y=D~6coKnMFW2FZ4_s2$_TpL*vZ%|&}?&rB*@N+-C!-Yh$NJOf(3|g!_Q9I z%E3W^{u#6e7dE_sX&1woB;cCUrqwj=h(Z40DZ&>g0m~-y)RW9D`!u922p(X;70Est zkW<~L$?R+a0seyomB#l_HkndmyoWHYFFvYAP1)(_sP}4MW-*j4#V$oQzWw(bAZkzD zzxKIwy|VVp zc<;!Ai6%~<-pDA3kmUL4MLrmp)aQ&v1{B@e8> zh-$ulPqT5z31&mQXBGA)uMh1jMA6;c@?3_N_6=HTzgJ{G^alX#Z~)N$&IK z9N_=@t^1!&qLiM(rb=RP55K+GzLS@WKnCYNS|=qr`B-wz78MIi+(107G(e!p1OF16 z5ey0r&dbdeIXylmC(2Y7{454K2C^mdo{Ve}LvmoM<#7xN@`Gywy-kBB_P_K?-&z-I&!?Wc8KoP{G{}#QN2#M_ zPaoNtr+%f$;h$`W&Fj{(&%O_djxTa;`f=CUN476AMmD2bT1lz0+6D^``$g{B_4wk} zWntMZx*?{Phi_<_wwugd42Sy^#_L3ihrmECUdmO8(|Df>XN>^^zWVc!;=9}RZiBtp zFFG+Z<7x1bGP&qOEF`OJ@f#Om|PoXhqru`jG zZ`z^TMbxytAB|`2XH)1pd}HlaPrVhea_QylPCLEXX1wyVTk>B$7O26>30vl(hly1I60N^9X1I>X zWk)0xX0Y6>29osF-$$DCFk$)lYhPJrDlRi!vKve+PcrEtrw3;XLGu>!MOM6Yvy(30 zvsC8^PSk@y5#oh@g`PM%Vif;cxKhfj+3L4|UBE{lYGJ$Ai`mBL-|O&*HGaJa&!; zUk-PNMF3wX0PJGjU(>zP&pci@INms`)k1!7wKh(^My}hlvHV`0E45%E;k|h2*$*9D z_lHRp*Sk;!#8~jw@D5RF_d@ltrjW7!u`nN(^a9d0;pU24hX2Y{K3x0W1-~iWV*4A# zVDn=S6Gr#;E3Vd?Injf-DfDJGHenxZFOTb1jU+&Ma2*pttgA0u;Uz zuAG(tbt`m?Q8B&6e0+XNFVzU>kWK{D!(C2g^8?tT#tF=%9(O##T>(gJq6cElx2?yq zqo$jsoEQWIWp9<*vF|=W67R1=puuI4ZCH=KCs!nS$JY&Ye z14}r;8J5*jyeaBI5({Wk5CJhFu=egdJM`M zWz&1#o&F&^sOiaG17M6;u`s@r-iL)a+m&NERvOHYdL8aW%pnG_yZ? zC)7So;o)k&bIU7~Z0^jYOO&?`1ZwAZK&Q}C_kwun<*^ExwdHuBZ(Hr({kkpwo_EFqu z*>1 zM&C({EWjB^@NDEs``Wmmy-OxH>y-u_hA_|(@uWKI$PpZGWOCn8T-Su#r z6TNj!JiO7E?hAG&GNA|}0iD}brF#x|%;#vJwk+pOwv6V=k7Njw{v@oqh6)aJv_RoA za7)Y=n}S!DPwvHSRnzv+i+ZKqq_)wPd~Xy)4a3IWgX(e>WfI6MKQ-+XhQfB7TzC5+ zrV3DB5`_thg}IG-Fv8zji1&)V_I?02}()kCBiEp*l#A01*}jc}J>x`8sgnRRJwD&P)o ztYJ1M*Nn?LH13MU?^Q1z%%oPnO#e#eI()yAd*jJ@miOgHX*oYW`Sua)^p%o0V&9$N zAgXxnmLJcHjioC*{8b=49fULS&wvr=$?}BB`PWbyXEcCCaN7PDGx=Lr`@E6-3yT1< zg)inESnOx_~?A+Wtfh5^zvM=Kk0mM&g2U%9pWx2k8`Ei)J19LsH*N z4zrL}eND}WS63fCL4@-A_{a!X`Sot0!*Q?bRppO#<;P=2+bhi|RbblF{f{e5x6)u& z1K%yM4ykuW3wwQdyuSh1jW|NiI~9y>pp~0R>$5c0>g_e;ooHiOs~C36MVSf&oZP%R zo0$Y-F#pshxaNgQ++u>L$DRgEN-npooRZ8-GQBp9ISk*nn;n;1Pd$+ocQIcQkj>>h zs?ra$G^O@TH#+ZEdCt*#eZQ*bsD+$S`xy0o>T`&p(pcAu^HJAA0xwJo(xu4g>Zh6g z{nN_ua$R3x%D2;tyRc0+w*wWtW5jDLSogofXJEchS>&bxYI{V_3` zJmF(SD{E%qZfwV3!f$5_8YqfO>;}cTC-aMeTg#8J&CcrOqbS3gduUZ_Q4AE@c^W3A z1w#h?XPj5xE#eH(iVIZ1Lf^~kIZ(jG2SzJvQqTO|x@dfHynS7{egF0O^Dng%gd(hOrM zytju$mjhCogKiH1|M}8Cg47q{?6+2@mm&HHHV9h9)X*Udzf}I#9BurNpo3w^4ow!+ zxG=l){rFAk;bT@rJWK|=TI@gb(#3U=<&`Ju`Q4gFr~REOXIp5W$IBAGh5rAFotr?7I_ zkTEs6jPB~53a^6V@EPm5PXgaN15oBnM{nsJ+NW}AS?6{daeisL7)x=h`n?f`$(%B& zY+;Jh39>%1<|uwH2F#L&t+e;yyrp|>yXlM0da}$PrG%CekyFHpkjn+GRHSvdhtTCn zA>1CJuCly&8>Z@0kyytyX%jTn;bhZx*rIrjjPc8p$H;x5Ua>3QQ_f0&+>BbgSO)< z&TlOfBHeElsh7v|5mt!gEDPhXf>jTi#nux2P0jR)+3w;$(pxWDl*6q~IVGYlNk!q% z+z;#C0La<5@P9$h;Ga9)wM6H+ar}08pPXc$H-bQ@=FNMVZSub%5CLoAr%;WtDhv@i zx-{aUjisv!_No&Ao+*94ImN*1(H|7eoQOzosU@)(|D~A1Z&N>fY$OIlgo!Dt(|nan zCLWP|QExuyJCH~#?)7-{n*Ko_`ialRef6PzYbSgrV6L|5tU%{nCKG$87GOjv)&!<1 zROvsjH5t{_Z2Lt!EZv&yLzWDX_#7ngFZPy%yI+JVemwc5Tw;uP32>c-oO1|H+E%JO zxQjLG;QRvuCQ6YvCo*3@7Nj$%)gxFve0t&Hn(umhaOTcGUu96Bn1=LFm+9`|K4f5m zsuE|pj(%djHV;kh-3klwqzeHU*>f1im#*XjXSwt)jMbD=?=Ii8L!4QLS9#DxOiwN0 zb$FD%&nThy2AQlDUXx#P5n8)`eRFJRTbxKLKsWdmtE;A-W^96HJ+ft%+YKLUM~(VG z*e!A~*XBwjiPQ2OGN?|cw?LWziQjYK%Oh+W!PYE*+febztR>Mk zNu4?=_)1r~^e`iyCgVmLs`>5d-`2G0ViH=eSIAgvoEmRJcU>R#Xr zzn$k;`gkNIq&#kRQoLR_#RbP^ch)t z`tmz&a~}nL!|b%M&e@HRj<#l7Zj3LZ7z_+Sa}R?`=ov>zK>6msYXOMO_GV1$Nu1O& zm@yR?g$=yiEYL$5yl?r0Cmm$6Pf?*p&S-4G}V>jo-=jN=BslLw#{L9AR*h43o2cx+T+0f&LwFgWV>ZB#aol6 z5Px4mPVNY9Yr9r?HX0G9s)E}G9xwJ2B=};Ve);_}-HVim@vB4`pSGdl zdPJAB!_xvVp^NDjgtS&iB<%6Rd}I{;5}_iX?2gQAyLW4dya_^RgxuyMdR}#%P$9or zi1&l_!$kY6r+>bC9zJiM_D&)1^d3Tdj(!QX+6!FPpW6l{#4Nm^PSudh1`QKFBVK-gtLX-X?@<=7+&`9xY!vePU?kF{%Br{G2;;2HS zL+5ZsBx%RS5Wt~!Dj8MR0$tHwEiD42MOAOEkYv~pR?6qMl&$)> z^5tEqi1V$Fck@8O0A8MV`~REs={`0ysH#iA!mmx0+$MP z&gSY%2tWzt&WU_Vr22gbJ)zqm#Xm_$PrE>x1h5OFahPxa<|7G2kU#AL#X{Fl*%(Z8 zDUKLCr<)WumwW)jn(FZv{OEWeyKW72LQBZ>c?40DnL|+m@3}rlLy(5bT zGKCcODCbjM$mxMvDDv3D1^INf4FUUDo&59nsJEY&U+;Ib$J+pWD^{;k*7P-tm`bxQ zu42u+>bZE)eOyR7#_08hdLd+3t$ulLILwh$Rq?ctkfKOB_E~CL3zBFwj3N0ChArVi zG1ZA*J%O(O8SMGc?J7oi<7s0AL%)>omYT**oP;+cg{g*rQwl{t<(Csp8PhFoB~HZb zpE6o4enUb{xIr=M2S7xmoTzLMOeG0#-!KmhG|<>9`|qnhuOzdLBq6P(M2oxKs%Bq< zrd|%?H;J}Shii+mT!U4O&@-_4DbF86QyE*UfHmom_IXikyC{T*0CqL6^rhO9>TU{P zTrQNE8r-q)(G9iYWvpiqd&s2wlr~TeGs?B~-L4KMln0~ANPog=_g1d>$_6!L)Rl8~ z->;_cKhsxg=}3Kb21##YL|ZLG=+wRDFmh)q_QgX3b}}^bmR`RiV(T_%Jg!4lDA%EB z8u?X9DMrBRpES)=tggWmR6%I5$Eth5MSoFe^IiAKZJ~UsI7!`6%(+gO#*3Ql0qoR4 z5}c2k<5j)GG_08udcaBsg%t-~tr&{;1IXWcN}yzUtI=|Y7p_t4rcl>F=3TpF{Nw#F zPWLkJPl+rGXHWu@O-;d0FaEr1iYj!Zs!uD8NbFvvYe!;j`fHnsjwTa|VrU@4N|KJt znuW??sLa=(NM<@M!G@6aNV?7=-pJTl3oz)XJPn=hEDE_Aa`ggCG=uR4!5&v>UnV%n z-F&50h8Ekch+AjhI;erLT^iuKf<$o2aa8&@m*YY*U5+pw`j=BFzymG%=dyE{U9R+( z^H>$5txaH5tSYt3TWT>y9t>r#4#7zn;U)U-eT)_##V=JW4z<6QM^iGM-Z-7>Fz1Ff z_uxCoJTiS>gze&*d7^qihf?NTYx$1zU(EgRIhwOgaGNe%m^vb}B%ZdaW|)X)xt*NrhOjmLh zqP#tD=gUmuuE(Z+6OO1x9u&QZ#wMmjNMN73H3rWUh-g6hURCqac$YhrE8;^SuivYw zC?O9%iN{ayBh@~>))FUk$07Fe6c0BHTjUaRW_p2U?o)2U{Cm#Y>tLWeaO1gFhIDnl zy09ZqZZFv9LUU?MD0Y5-yv3`MHMGmXfxd8PC*ka-}xLCp4O?ECzad~(fj?yxV@HZI68?a>m!mW}7w6?*~CbWb_m ztbmnQ*cL!2;#fdlL^6|M1 z?PIv}Wl`weN!F&6TEaA}6}#O_w-C0T)`Scc-LM|bvl{+1L=y7SO&xEV>Q7}aVOCiY zuKy7VLizlc@>|a>;eh_55CP3-E&WW$xq$%B9^p4C+JO3mimKSUxR!R;Tt_Z=Nie3m zc;1%IvAA9G95|o#92H7!D1TBE^FOYts#4CmMhWL|6KL2#@V9F-&A;?kJ3aSzl#xJDAv2B%7+LTpLu@ z@ngu8@@q&;Sw_Uv{;)2$(9Bmf4HU^*FKerdIO$F{>YpJJT9`$_>_0A!lx4{#Vw;tp zhtL>ekZkZ)yosYrDLBi_pUJ(~ZT_qvEwd#olCW~nBIhv2>|$_rt19_eq!yyxUI+G*pC(U=;FiUr4!k3OM zA0?n`N;EK6i|piZtWuPh${VhvXV42-;YUP|X>^yo3`Zx+K>!0LrzUZ&t~3IH!7u|M z%`L!*Rf1SvRnfCsU+={Ap3H5-%VAlgtQ5zCjzE!m_)S!7@lUnyhpCimlLLK$G^!b= z+8UR?wEB1`QK#@5-q<8}&e)(9XEx$Ur_Z7YbUsYMkGW%Q7YSaJTVG_9>$a2pQW?av zRZ%dn$Q0vqTiGp#r&WJffcF8XQ)sU2(^{@F!{u&OKCTeQ9#_rB2MRWAn3-ziznC}; zti)6N+*gjS#Exp(*78d;L&e^S;XKrneg!<3T*ae zzKQx~eR-e%u5_~rYz<9>2MM$8%Z3{&dqw(T1g7}J+uYKYh9!clk|gToNeE;kwRuY? zC+(#5lithl>vOQJ#4a8+lyX;2LrxQ^{?4=wL7qYz5^B-L#j&b-^g8ES z3#OcWxMA6t+5(3`5{jmVd5UMur-*xoqU65MJ-OvY`d!>pmOOJMQ^ps#`y&WH_Trp@ zz|C$mO-psDf!{kfK@Z0Gtq9&OfjtJy50Nz&vX13X@a?ZQ_IFsHfU$ezp~P zlpr<4kW#URT}Ytek9UFx@c$BJZc2PY0I9BGdgA{{Q(8BEN;KFU@2>lOwGk8W&IQlp zU{_<+O?fRwc$A)Mi%1pQsTJdJM_A3O(o&7Us1xf1Ot^}o{po~sG6osb(Bad&uJW9suSd2Cz006*)j(DsGHE!zkSnM{ zYZePk%oB04i3q8Z{2kjZ3xdgdRS*9)-{C=4zWH+Z3#;tK*g}OYaq9%zRBZJR5<)wC z1fJaBhHkAgn4<7XLko^pNFBo)*Ah2LB?*f(_WwiKTZcutMeW0iAP7<-(jXwxpwbOe zQi60NUD92H(vs5BB{_t258d58LrOFB(0mV`<2mR3eedtPuJ6COo{8s~z4uyst$W?; z-VpI0ZY9h38u)H^ZNQQ7%9}C^@lL)-;B+1!KXn?+mw4QH0){QTOic&CsnU72@*Ev_ zjRi=>h$XWMwmx3PXr$JcXXg}NfTX|?fy?EWXWtMaeu1J<$SQ)*XrrHD@{v-@W+$o( z$CKxYU>~sJ^blvd5dv7=y{Ul{>F%LwjZeTx@4?@-nOW^7w(TV$h%Vk5MA2GQtxdF= z->@3MfyfoqFX=#(sd-ToY8M`La>F;YI^X-bQoWC}U2GdS?U$iQ&HySq=9}BoZQixZ z=M^<4qWC$r41Q`|p3gnFyahYzYb&iix@KCm@IO^GOr_-r>J92DX(sYkClj%1qebL) zC1fMdxQR|`aS$5lCQvt(q0LW+iRkq04)9fCc^rf@J>zP4riJ-p{A!CgXE55yqj)Tj zhM=<0SN-#~cJMtdX2t{8RK~JFVi6r`zTXY8Cme<+W~WdsX#~`Al>-t+hQ2zTx}kpJ z%9Q*V`?w&`F^dIvdru%UWFm}xyP8l%U?wl`oK#W6SVCcUN6_JvwCuFcoe5}TFmmyK zx`Lx(mKf{NVJ4Yk82v=otXnzwp$2)xI{O-3 z;ei;cFog@E*Sba)Yd1=8*+_hFbgYnr6rwNvjnx}X(j(bQE-@fP-tKC0m@UkzXkI-y zl@;;@nyu9jI-b9klX^ z!VC4zrAe$=Ukm8Y!-{D(@=Z{j!Mdql5@plK2k0~3QQq%TFl{Nkvs7vxY`KE7jx@j>hsM*e!0%2Nwo_xm z3aj3p8!*MN>L+I|VGNF9dv6%@9Q#ES+|$N)D?n28ST8I&`O zwzS##fU?cozQW-Bah~{8SF^j$Y?BD6nkjT33>2t6(b(N*m;L12d}5lRlRh4zr{@|D zG3hsBtsFEWm;Fs0M5&}bV&80gaAmZI|=9Plh<1d^n4`KcrU;K^q1>7 z=n0}FGzyHTj?9#e+qWyL%mMKvlxb#hRYcz4Ke4bVUp849Yn5IGz20U$3b%#)D1h)w(1dK& z!}6}@5h6v8ceSaUcr?@2lFQe>CCB4*T6R=Dz^zE~htP^HS6piOH3cl{4H<9S@h~ES zKcq3jF~a@!!7h4Pb}}dDGux5(h@mvQ>V&)fS}tE*1cuWlw(9*DkX))dN%#cRr$}qp z>&w5yseEU8NQO{DWYJR4CH`t`-cWL1F$i)fvW))#JsaF&n@%AjoHFwS8?3&(y5R&8 z8@t}6>fV`z{kc40XJXf_h3FkSZGse%&8=#*p?X1WzdOwqnuxpd42Sd8EsKSQh692u zB!^xvV02T)ZkFCn(0c%`cj`>>0f;0u#^jH}iOSukqODMm5$_U`!kx!o_x4m@KwjNd z4?{6UcTrAKe#vkyvpc8#jCVn5k_1H`lJhKWDvpQ4z+@{brll3iMHvKOB%)^^@_e{7bYYQv|ELoCQHepXd{n*WXn_ zFSifO63?9o0f7_4YR{85(Bl@yHTs|qOQJ^wXEz9*N}y#})vWg8wJ`t&6G&_L&lLpO z&XB|=yB{hy&Vs3ifW)Z8IK%Z7M%fOE6y3qO$a+P8B zIDb>JG2B!#L->J?`VYDmm67B)p0$+@qFgz6$>8+6Q?6$zSz%3^nQjw|^$e+Xfbh^k zx2}-!)R#nKFSGNc{I|ah*Uv9#v$bz^n20@TxYv3|K{^IQ+ssh@f)oW8R-DK~*;Drh z*O0vvZY!y|r`wI|W>B%RY1{7KjE$-Qm8!~Y^8B~1qU=n|+@`uW6J+-b4HPApHJ$oX zzV<13#1Z#lL4jJwZRcADR@%2VDs;tV%8f^m;W|Q0*EcG{H9&wWcPoK71P2Eyl}x+! z;}8U|8_l$|XuRrFJ|w(W61dHExy{d%a|&a>SS+;K2KHH>Joyk|%Qr>yZn01ea*G`B zLp2jBNdS@Kx-dWCE}q_JCxT=wPYBXcph{UEV&#J)zYTBv(DmEY}7 zplxEQbXwXVRP9HRjhoAt5&C+Jy!z_Ap%?%FP|Dgm4ND;f@G5KEHu-CY_NOmo5{cW- zjLncq3dh^(hn23RiAXXT6u7ow(_7xt2phG)PFe<6S;*6K2zZ>*>^wdP?h_^}E;d za}*ig>IhKpg2^|F9%@et<{@10)^sGHmYNZ~1z{OGHRVpyPKSf>s=8R`o;^zcy1x+Jm4H66PCmNRrF9O7Mi%*M8$sT!_T;qKEu*68Q0geK(L zZL3+o=TSrv&fhAEX2anb-xG0<2(dZWRF<7> z5z9`sY$_D~aVo5pLF8@;N)==)`~sE*XaC0KZC&4#uV>0OgNh{#hy7B(k~TBK*8{|T z2E#&zH&VkfodLQ}DrA8qV@ajs*F&3a#99!t`zJq^y{an90ZX_#es!*C(1_uv6rhvC zRiPkWzeQRD7@SJ?72uNDzoEEGD0J-N`y7rX+Jy@&-*$FVdF7@}1{b_7;!bs!0C7`ZDJs~LTRm`*|1^zO=B`;0Y+vY6mcJ5hp zD5zYgF37xa7PRCx9gHciJOm4~c$yxei^*B`9X3+h^R#m)@%XKHDtt5tx$;(MC{GH+ zG~55q76>}0o^Lz63vEx8uZ3p|yhgWFn1}$XY+0?K3oloy7XeWD*Fg&Yzs=dQFBqu# zrwOPh^Ji+&XU($1fFwfDrd$-+n*HSpe=-g)yqDj+2*!pvQ)v4cIo5 zm3#pxI!kc7d^m+DZ;JjP6o@&_ts1Q@7f!>C)o|Tx8pvbc8HcP>;~J5Oc&$5{Ae*gL zBZ~Gcz9n#|ZYjG^Nx_LH){EI%>c;tYw-qit!PfVS@ARFjN*v)u~+_;W-d&q60if1SlCoF*SOG601d&C~)ZH6XFz7_St zE0M>~9>cGle7*7=KW_2AVyN>J=waW9UjNis0GZ7fj6mT%$+3y)`Y*p3$6ow4)9E9&jw%}ji=2E8n%;QB>2krxliDsMGli}w&KWUJO#rzbz{?K8 zP-{?``PINMTBOz!FG!|4@~gov9UX;64&SZL7e^OAw}en5p&sTH<7ogdz+3Kj^vEPDa3`e~=g!5o1mK|z=~geqlgR5NO@`k~HIl;A0_5GUFy z%kA7E1TI(>axn6XfNk{c8PU%5O*w_*K^nuw1aIn~r1QS&R4mX+Wu|UA=On)C1Ekn) zcvZL&3{d;*VhB8qv%^C*@)(*ma=a|Nz)jWYp|l&13+G=#Cc_S!z-n65VYNV+BcM!0 z|C)tmLosz2i`6Z;4a+a+xMM$33pM6-pz>=Z>zY~J-rx=OVxndb(mI9haNlsJJPsM# zQWwZUS79=TsSEclu8Qicx106i@HDjzJr-r!{NY-cI}RK!gq^MRooyPR;c2@MQL50* zhJ}$qD-3sjer+^p8YEw#$wP>#?%@}`xlu^7krz^o(cO+x?cdwby*{^n+Q@$ysv28v znrEaSUdg7*BpLA|NN2(&wBOxtiudEX=mpmaSV2j6t^gt=x2g5MF9BNf?xkprwN8DZ zT)qPv+t5cBlaGz6NtK11aQjiA>BNp-H6(B}2S1m9$Zx6pb2%S*Q4Ne}171sg%B9(E ziGWYNk_kEktq5MbO&>)Xyq`)I_{4rN3J}UQUcZTU(SM|Pdj&sUcQ=GS%Gc1*NiIsM z@XQ+`Gv$-c+peV^TDO?ky3Rh88=yW&YN(f4(f79O8H6^m%|hx{MBVu=Jcr zY!rC0CwEuVT!mH)^Zg;!{qFzx31~~WUTS~)PkZ^y``=u<&^uk5I1Au(P(vj^ZOQ6V z-&f~-1pj2>`!vCN;Va}ftU zIeqMwVJym;%}A>O6N&=#GBZs;`!B*p-eQ8k@9jC|vy6u0q8s4_j54FGzgi1_tZ@m7 zZ9pSYGO4QG_yF5{5DpY~{gCQA#qp8`FGZ@b;Q8TbS!C?HoZwlwG|l1I zR2JhL{wP@2RE4tSSO%XF_6U$mF}iVbX6{`#XFOKx=nikN=qrA(yl}c$ONhC4Owjw` z$o=Mp?oYV<>yJN!ua$olOj&`0rU2d2gOa4N+hs+B} zO+Nr7JH3cff^DLUYnJtASo-q6b~-dMNi-TRC4ybmM(~s7rKc$52Bp76_j@%<=&zEg#tVdvR$E{@~ME*k!YyU1e&N0%t+E##c&7at#|7f3%#Kl--)Y}C>SC#)eWQ0jDB|`MW@r> z_esqG^+g|Q$d4o)R8TK3%{B9I3G<2SW5cF`PKE0x; z=~bBb%ag_rDPx;NwgsIpwq|pQLo{3tex6F=f7gNujW=wSZ$5zPujteAeR{`pa?aYN zm8+arfZP3LPko( zQDir^@`=tuo@wuqaSHL9>3npl|GJ>snBP!;yg{e&KcGI)d9iAL2lXG`?_B%mY7O=s z)X&ED)-_}1w{T*;*!}zBn_X5(JwrpPAJ0avBFuX|#^m-j@LOfX+jL2!z^Wx9Tb>o? z)W=H=*xh4PCbt%AT*sZ=A2d^;jj5RjhLV}J`!#_CeBwMI zF)RTAifK@Jr7+^+M*-QQ1ulC4vrr)-r=uTKfUAOB5z;|lzxN?Y@FA;WM~=n3QHNNWZ@T|}|cNN+IC?xSaYL3HnZ7{=%7grWjK zp6_J{GS&~P8D`K(*ZdfSNfD0pTl*hBFgJOAx}=VROoL*Hm5;K4wxu~V13e?kJ>6%k z($feL=3aQVWAK==-8bql3)KJ?zClw*Ad@dC7Kc5>{0_y7f1}#)L&Zr%QjV+qpp1fe zYJtLdY7h*+6Btskft`nLma#T(la~*RZrIcXTo#R$NUXck{S!V(Dpj`?*Xj4o$T5c0 zXOx%A6s_h{EoNKOHLnDquO{-AlXRxcTFY>XtZ-#O*Uvos-{VGDe7&MH$On4bSQq~e z-6!8mHDY63&n0V}^1lR>K1WxWf;s$@Rk7Dx6wz3bPCb;I$$p|YhFM-3T;eAls(g2* zYb^wr(2Du@v`X_BI-~jPO~%>R>nMN5I^wqGBkGh3%AyU`{PqjD$48Rg%zqK_+Wv}D z&PP!}V|@gW1SGe;!^Q`DYB-COzBMip3?sKvDa;>LvnYP75LG9T~i1C3s=uU5A>2v}B6}zTM90^X*t^BZ2AC5iXz#t5q ztwvj%i{N`t^*!$AQi{bdfhRrCaIG?~j~ZP@sa{+nLb-eG-%`2ki$)y<$1AnV-Y=Sd z;9JvpVa~vK&~Sw(W(Cwzt<^k~t#!8zC^DH#Xh8S|Yzn2jRX!+cK-_R5@ zflJP{=-cz{6-~}%290rNE0r*m@yZu~$KwS&9{8wr#C>b$u5Rh)OZG)(AOg zlG)BldyfI0wL8AjVF-Sx@{x6O%bl{T?4@8gUn-$xXMcRJ5*}aX(QOJud`Mo%s@O)= zuFl=&)KV!_7nJEYP+wBv4Q@ARh8 zOdW;U+VifOL<`X6P!_|TUvz)E{Hu@mKzQzBRCl;H;?}DmhSY`zHzd}ac%y(_j%+z1E{~=b`dfc@^n*b%8J9(Du zKFGN1RI!RLud|t?(>7ENI2s|VNUM4TvCTCbD(;N7L`_Yd?=Y3SarS7z*SXOhq0sj% zVpONna^!@B%M3V1v|_Iw)1+RDt0XrVUkQb%=^Bz&=eTX2YSUs&VSe!npGEuo>F#Ch zWbx(T(_WHe8l?DEz7 z?<+w)(gYKm6u|Y`GjA%qwqKqa3~VtIyagUhTy}b=lG?!wTnqE%`O@l5C16(SPfu1(ET*Fk1oS1rRC#9^T+ z3tSEi$VN}-(DLj;MORvo1~IwIX2KD=T;hU?w?j<#P7Q~So`~&SY>?zny)KZHI>jMM zPjL{xYY;~a7MN3$PcTTGD~NYn)Zvy4B5C1vhNJ_2^H8}X`qXc~_|+f)4>LzV)*k zGP%F1*X$c!j^j*8!Z-i%bw2JWUaO2C?tHM>PiHqhUf0uSRu=+MEDP_33vh;3E!8)S zT#sYh2bv^@EishVqg-Rv1ZoR1&2gCoJDr3)|R@KMjgSzcr>iww#~wz)e^1snt7vwyC2visQbmUcbx*O&~cku*YGV zQf+cU!$dGA55tk`sGhRd^bjAMUHj)5NKssDJF6MBf13Ckn;!=Fhm(QTw51oyUL{nmA#g@lM{~h^(LFqXPiG@|ElN@q_I(EPa0r= z7osg?<6#z@m2%zsF!c|}G?9K?E)8KX5k28%a#OjVA@HSUwUoFl$YoizVY{u7e+v7c zvDWxjj@OI+eVN7EgM~56S=UZg%m-p(R|>mGS0&Fec#npyPb(lR-dqDWjkCqvCsx@W z;3gVwNrF=b#2!H)=AB_~J-brbdaU?@;H4*AeUao#&VdQxl z--0hZM)V;8cTIf&CoX~fa71I)S(po%a2&!d(Dm@c6S0pACdvhNgIWtRic^6n>%MW6 zcoddEx#YEsQx+Nh&2x;?B-o0)E|W8?qj0>`bM33}7Q*%0OmNmEQ{(!#EjhhOiP|QL zf)#nLEIjj}CAte|8^4(VF=fgIQqwiv$U{I>0ysaa?VP|DueOi zx}i&^qAyYmON)`+oC}H7{E=L-0PLsht`~m`V4Sn%jD+!5r&{aA5UVohsWh!`$X$FK zpm6{V?G5>urCvT61>hya-&L7pB~jJ1mc4%A#`jbAfq6?AOkl zhqk!S)(D1wwf+)1yT$e%!(9HcaMvZ6kunNFo!O2~G?}-=X$33ep#Dxp|8kby)_2-R zSvwHHan%1oK)Xc-O2M~}dd(jETOs~_R|-zhpuEag%FPFAzKLqC9uF8(mnm$pb_pI|&;9~Wi={V@cuggjjCt9Dcc+&|;`_OrgeK4Pkqg=q(wFPw}p zk8pZnfw{A^o-cgi>w^0yk@5PQ$`m{276}E;B%9RrGSkVzd>2mFy?(A+g@n<;fPo5w zSnQKczUDj~^F>>J)YfXNIm#rE8V(fD+SD$X%E6{MevsJ{4*>A8Yw$Dedj+L_uGR@9QE__Uyy7W2ZN#=*y zr1%i|6dfm#XI8>sxivov6u0b>Q>B20 zKkS99hR*`>-e_M+#x8YcuY4Hd?Wq7$VYM;DhVs;ut15G2e_GsXb*O3_E?;amSs6XZ zFOo6X;yg+vt{MyA_eF2jG9Vp&&QB4qZ|u=#xc2#MW{$%6JN?h+{A$^{=hQWyYOqpG zqVc`nD7-w-8O54~KdL4PaObI*{j@_?i#@ycxT}6bn#o7`5oVHjhY@Z;lKWJi&Rp}l zn;Y!0d+d(e4g7H{>nsft#M1FTET3#N2vzEsQHSxe3)_z!H^VNQp={LpjHk|6njuZ7 zN|#M-+!@ct_0^St3a_2(qy}=xxjqUx6K1Q8;A!GTFighbK_Ed^FfF=uP_qDbfQYO0 z%D4IN6A_zfS+@s2c-wKX%NU1UJ;wvM} z4<~AFwjzA;MC1IkCdGUANkB>+IW0sdR~7i5 z&uyX=nFW7ZA#=!IKErnK>I+t2ZW5tMe5LkKNyJs=hV0@wIh6z8{I3O+CUdV>f0ii5 z&b&e!x_*0D-`acowE*5;!vaBg;{r#GxV{Q??z6EH!kbhft-fbdldg8Z!`^=UA zzxKi9!s6~eVy;}|&G~idYQz2;Y3`|g8-gJ=(zk~;KRHMUfwN9T&I%qBgRzB?Y0n1c zR_?i*9>5!)Kse`7jT3Jh*QAeToAP`41-r&L1v50X^yuVFgm_*IH4XJXNS+P9N>`{( zaPhr@tKC!wfh%c)i*6;5bmLEpuv*;cglDjKNl;@r+vIO9yRkDEu%j&21064~V0r@W z-Vacn1S@V~BK2M}ov|zdm*!O^oioWqtZgbs7P)Ua54dZyPE?q3lPTT8{^yGIFxZrGRP4NLyym zaX1c^9s_*)OF#9Mg}J{Slno?d~Ov{c;TUj=nCz7PV@uB>Bjp# zYq>l%8~G1ppmJ>{Zjm$NKD8**+b{OD9D*kvjLMmJ8Q!&d`#Z|JASwb)0+_If4yN<2Ik?Ft1- zb1DhqxkJfWlU8$msOV=ka~0P4gBTV@5qaeEPLl#Na$5HnYF48^DluHYD!wtqt! zT7oe<+B-%sZvF|In+G2RF0$&#QoEKqNHgtsnsa)@IV?P~>n~AOk?jw@>`mrY@5(^SVH94%$7Ju~r@!!T(Qptgt4y})7fM;C5u-?)^C?#;E!kXGuO*M_qCfp0<9#v#HKpn)iY{gf=Xos|3>he73_Q?fBOLJVh;QwL(eRIHhW{ zHrZ7DgWfsRR>3;NN4VGOGI$ZPFz!)xLZSjzGB(~)(*r@sdmTNWPV;nUV|04?Q{FqL|O@No0>BAz}rjC zch<*?zGdNhClGfy+XPARk&9(Lnzh5CR=#SOKzkc6&Iv zxOr5Xw^PH5HvfR8X!jJ5>^>~9)RO!w`X2nm#G(0n6RPjLBB20I`fj)WsT#mhg=&^>!;4n1laE@@vt$Rp0rG?EMsEwe@6#ys}DJ=HdzXf ze~HMDi}h26@q8g>xqY&)b2FgS8sFCri`g+Jt*n31h05YCyba!J4aEz4mpJcl2uJ z1?Z^-??u&otaIiQFD*G5cjH^Yw>}y7W>r6|cV9212-CV$=dbv2d%b#nsTW6h*h1Kb zn&%3=+8r`$6{L>UGwd#(-#I=}K|vNdLM!O6WY| z@v1Ef6*o-vIdN)`_YtprljU)B+T|gvVR9KRUViwf!&{X2Dlnj>_cWyUc#JdX*4p)W zW&@@JttJ*9$G+J_D^{H9|O7}Pp@!UEWyIat$O{SMeSQ)b* z#csjNsC7|ujJ@2$`cp8B01Cyo18qeNH%|1AKPYQipgj(}bKk#xHO@}~z&G0%d9-p< z{l~KhJqKL47ggmd_ zl^S+r{*a1Rh@s07eHOrXygSF`et1mZ2)d@|2qp53qLnc@C=1v|-*o)EUyo>GRq2g+ zre=a2w)@cY{4EyNMQpif9dwapXsPX1j4oA7l`aR3Un0zam%UwN(uGR)REG(N%UziR z$DGiQl?lm;-*$f|2kq1rjXoAFvExaLvJ5KLi6EZ70~QX=SH#hMWQRV7iA}=r5940T zAaD7DY|(hfFZEfOQDurE>Ka{90<2~gLwGzr_@Xka!nRQhNiiESUmds zzL+77jZ6;|Nd4wH`?=|?eg^eOIds0{3rvrEuwE@rHzMU%rWovCSkRkK1HOGR*>?X` zil#pir&_{B;gCl-u4Nt$uqw`oBR&7y_hKhP~jEY|DmVRcOsdWv#RJzcu^BBw;? zTxxbvbfzodxyRZR(ju_;%@wuM0f^kzJo+<4V7Z{Dl2;4MFcGEuQXt`+oV?5d(d(## zJo)Q%ipeEN+I7=N7J*(nY`bK+o?>7FlXSjozS+AkwF%$bTMxCH|yW)0A(~FYT?A)_{G5NCVFwHA(E?C3_3W79F`%9DpD=>yZ@U!7pzX zWEsiF(f44AKxFl1-UYO{)P#M1IU`^(Max)#@?2-huxho5AjT@;LK%`~Bj`l~vSl$x z;>+?*lVc{(w7P<~c^y>y+$Md#RsG&8b=&LqXz<2=vZGpKtsds%xUlGU+ zc-pBb&YSCo=j7e1s65b1uk2O3U3xHZ1kEkUhCF5o%sX3#$`-WVW9lZTNK+P%UE(+4 z?R%ZS(&_xi&VY#NlW6lZILS=i>FUt|VLp ze$}hoHCt!hDs_QQpGG>LJkWw0_B4rBS$_=c3Ossp+~zmw3jHbnC!G8H;onkyzLcOG zX)F8h_2Ea)(IRY+{&6p7qDVFA=Yi>^324+G!$e$*6?MJ76-U`OKC#QU@m-Zz&$1=Z zK|v^I&YpI*B!vEA#4NAk&(0@FV~kd&l6>47-||(F#xY&87s zT*-zH%co;Q6|Pe{KhCKAGNw9|B^vv*??Nu8F^MIfcGfg z(_7hBGhe%Ozr=GcA|jtG$IUy!KjM+?B@N@?-IT7#_tE{@Wr4a>ypYm^0#JxM($@j_!R ztb+i@4Xfl+)W+rGpwZc_S?`$SH-pCfpmNJqDUzi4cL)SxsjIEx?n{_qWY#wZ< zeQ^cZzR)p1?M>_X$~z@X#n?GVb(h?14vXukAXcKwlzzO6BU=L_?kkrLam>_)yQ=4! zexfTM{G$bsVid<5q`M@ssCGk!&V@=4?(%efZR&=Eo;1yVhxeaV``19x1MB;Y>OZC` z{CpJ&tu=5S$MWBidB^87!uroNK92#W3!CO$;lqEX5SivPMxw|geMlvGKlCCr=aZ{RQGMXG&%gWq-H+lF zBR^!9mZ%Kwz)-skP4WNtT{N%z^IYg&N`<)p{G{;nWx%E>4hDI`3ID$M<<*mBlz~@1 zT_&PTK%5Zts1)f>nDDo|E4b&g&wwKFkMk4ud)Dlgh30i(xBt(7y@!jdoO|vbft-^0bdWxN?vD6yx8gxidWBnkQ$gh%nM!>Oinp=inc|2}Cr0 zkb8S1{t)^}wEtM*!JQ?#u9HeW{$q)Rca})lVp#sqjsJ%||8upu7ugop-TNNR8TH>M zc&?2Oj0cN6i8Jw^@z|#WhJERvP(O|G@387WbNAnu!W$TikPjP424Vl);~o;O{5_Gg zAeTYA_CFW>BJZ)71>o{?{_!3E&d`5<|kXy5X(O){O(9iwz zG>FEU@jRj9_Y*lW@ywXSNO=V^@9*9cFl#IBMp@q-a9w}$33e;Y0#mdl!s1slh zr8T818Z-7DU~YTT>C{zxsp?BG0WoPe#5}}0%u%LS8VV`t0vhwV61T z|Igp{`&^_u+}`sad;6cW`DYH%Qa>|8x$5lK7&vSu;!^qDES_XFnX+2^+B122gNcUr z99HkL_QqT>Hx56flJPsI;WGfM(B*W)+<$?{30pf`p_R{jSE^|5ITwc+qv03t_G*g7 zH1T|-y@-d5*Hd|Lp+djtF+6EA*n|Y{>}(Uy?er)!rT%m{5SwB=YbB7%9kC@9%MhnF z=fm~{XFv8CtDamvf{e3PJ*GP{-{0c_!-tA*J?FT!2aRYpO|B=J4JnjwvSm}5rk^4x z1T#zYnm+ir+R;md>UrFwohHFF_;Qlu0tigVJryuOfRDhpm{6a|9FC= z-MG>he-9l8->e25Q*Xa?;)|YPcZp>f!;PsaK`d1WMjqjxQT<=Wx|tW{(=KFWPrxJJ zc>9#WdA7ow&am_G@p6X@uh&KBtZt(Pe&Q?LCgOw2ZLU0nHH)$c&f#LF6$ zTDN4zNLGMSRQuO8CWkLB4!gYr;j(!1Qgi<^XzkWov1rYk5$T9m??X&LRTzgo& zB!_Z49qOHYjCvtehfzNgbvAHfANG1DA&_UCx15R3KGF47s`;Q z5CZRJ%RHk*bUNATN#aw^{xqg7(kzcW)BoaMPHXl{opWol0K*F(fwnzaAU(klU;Zfr zHqXf9vwtNloOYjjH8Pgr!*`1*$BgMxgDtXrM7NaV$%R~R6#avf?FlIx@|Sl*vL-`)FdR=&5il^(b4~X zi2Wiz12T#XdL#bU^9|ypRH^)Sr1;%w^08hg7aS)Cq{6@%j8RsWQ-wboY8~>A02mxh zhAGjH0v)I%VG5T*8*c58F<*RcL;u>ohHJvUh%y9gq0uEm^D8Rf8Q(x#W#2lTW`&vk zd=y1F0aNxhYD8bLRP@WtV(;oqBPl0rTiX=?I5ZzW+dPIPRhP+M^G@_+8xZk?Qvmz; zq|0LS(h9pJRI#=$e7oeY1R?*Fph7S&KbZ9*d93-!tcvf@&&oeo_Kw57O;D-ZDS z%R=JlXjupPTvGLKKZy>Qc(+d^rLke%DSqxgMW#CGeyyCTl zoblLb$5H;frCW2;B(f$C_#I~_cxSpH8_g4f$=4}N~EROS0+oa*ZZ|gN1M<0q-H{>kooU zYoDCmI*Pjr65Q8&Kvx&7Wb3@}N$YRWXo|<_SZlUrP*dPhZUoy5frx7~lbIi|CnntOi93G!E`O#6tqG~h zQH1R(6=kQk2xH%$0krVqq1Sy>kIRpGuH$+Sa{lj&TKGBp-8}yG8`*iO18r5zByukZ&1lvvih`?nhgyWl?_Zy zsU2Is=&H8XED6XRPyumGv8a9g)irxUVmAFVGKSa5LbKc=Vb*f944)meY}A_|KoUmA z7aV{^B7TWY`t1v7FGfE$(+A!x`DD`?6@#~naap%~TDdW%s&}t#jQB{v;Fo28D+& zKAj9EE;?5>%UeE8rHPiO%rx6%&w7<1yu)_!NXUMbkoQG+3Pr;r zrPVC^;r+1pWV4$nM+8jTvJ`@#7wb_pbRJcO^D1O}FmGWR>DcHowj8+}^tHZ%BP0@y zn*KoNJwdwqoLN@|F}M3)JXKss&RQ1L! z!Q0JYEe3xNlR}vAwKcI$yPdmy46SSdVc`oT_2LC!b_+^#OEkt!Rr5<}+eDsI(`+>yCtdRJU=zn+W8eTn*!P!dtERCc1(V{t* zwX2ty+WBuaOnC;^drcQi4jKhBKm?g$;j%qSFDR!9V)KOobY^A|_0go&yu~PG*jdWa z)?C=jp=9~oA>CIqxu7D8E$0cc4z#;ei}9H-4m4OE%Ng@ea6R|rXEsJ+m6M4rdll7Z z>HRNhQpy3?^QaDgjQ0QSQ?I1%Yt`8tZ;Tjf?514LC$dwq;y#XeO0W1WnagG&(;i>D z|Fqsgi6CxLHB`vnjFm<%N{q#jW#77ZVJDzHltw0A$-AVby=CXx!=*sSbRa>pO)@lY z_tBG=%56&)ffUpvQzf_}3!ef8UgpL!tOAstABJ7>dHPmugL9lEx%~+zqT}Xl8$P4i zGO?Kg9*BYvGv9r{#b&!>CJj%3rzXy4aY54`qo-l)=8Bd{n`5~{<3CGDQt+{vz7wC# zvuHOc5AwRZQqV{xtA|^L?TG8Al5pANhgkaVZ%^9%uwJIT=g?MwFB95&D+r#7lC+U` z!iBo5WjwO}G@mDkNwk_h*aQ@-CTymacMYe7Wa4BitWTshYPEH``-AMY`8^Qf%rPuF znrs3mNqmmmw7$LCYhJTw85eArIW4={ih*N%pU}J) zZfkJ=%5ZSnOI2Y%ATdk#b~yY!)>Af<$L^=c{=D*hZ_?!ndSfV4J`oE#Gw9TL%BdQq zal3!3N+K0Hfb%4>TNRTcP7f8@RJ zY2;JIFw+>1xY4{%Ta7iB%5}|0FIN{E)Y6LGtpf0@i|U4_Y+#R*){fIb3v^zy?DkiU zdBre~X4~@1!OIRtU6&gR*I@fB~mW`Yr#fSOq&pMb8OuSc} zREtvrBBEDs<%cqkwHr8E=NF+hoUzO*ZwFN;%?nt)Z8|YYgxfl0Gy2qIZ^QCb%7dW~ zuyZDTIq!1YNg!2i;7GyT!T19ba58_CAz-(<5L;So&2Hwh`w*h{=qsh|m+!ALm5c^Z zx|%%6^zE-`BG9)vE_%q9g2D%BbTA*|s-TRT04fF5 zm@~ar4tpiu_7I)VGy#k56KvwpsO`A>*O*l4tObWh&Gbsnb&K-@TwbsLs~q4`TDW2!?(JRkcMTK z9$nAs@Vj_p)n2>O>${o_Pcdd)1o;*{Is1sfV*1Y5S-yi9$IX$uh#OWCHZgCU^(8cZ z!Ikg)=ogjH!8M=j8M4)EolF%0b8^~jrK6InxHL{G3HK-6CTpkOM0(%63u{{P!gnt& z&o9r%I*aPf+e{W-52n76u96HR_TSmrS*@w1xSHI6!yrmvtU6Sd8XvC-X=5W7&=|5% z`Vo+xZ*dt_5Vx0TfeWS)$*X2?I-fHF^MhcHg3Iof)MW%~Rqa6jF32EF;NmN+NUg+J z3n;%{FnU`DKcPWS;dMxjE4q^0wN%H2?kC5xD!pY&3S17}?ZiQxoY`!I1DZ-BZx(97 z3RN#+omr$yJLc{Uenmx%3qeIa(?>2&7V@s!Yk73Du!Lj)hr3vvF5)deM)sbhZ|DZDwc_z1SO33CB~q{@Nk%fgM6x{;~5v0MH!X&nk?> zD{t$jpxiDuTYJX}wB%cPHofLWN+1t-s&bWv*k`9Jsa*9+6>@pD)*rc$xv(Dmup`BZ zK+2+aGF`CIU9&=4Y}@hbOzNUuiNI~TtpWAJxxMi%VT+_)_!(aqrR{x)@W$!}jw@~d z|3{|&1;+Bx3K90!y;3IP%e0vz-Bz2lq+lIyEj+$>D5+l z3WX~ElAB&v7@0T+SWU%Q3lvT)5S}xX%-j`=e(Fvp3w(6n>(W&L{A}RZos8QxXg;wr z%E}k@H1}l~Nx@MpzyspcYqTe_bJgY}6||2b=CWc*lE6ab)Hzw0oAxPqq_~F}CH>AP zb1-s+@LRTT@tNu8lIpp#Z>}#kn7fu7;w_3TPTUqbyG>-1`p2d?l>k1f|LQ zmBq6p=Bh=eRO@7G%GK_Llg%kr50b4NYA-ZeIQ>8Dy=7Qc-PS&?0)iqX0!oV@4I(AE zK?J0`OHjI`8$>_^6qN4n?hfhh?(S|joA}M``{?6&&U?=F|MdU#e&JfzX79P?T64@1 z_qfNH(^uwa`Xw_iJhhZf+BWETGTX z0XoN5QCJCc?ciQMoUXDPN>b9AC=kF~Vvv3Y=?>=<@%*t!t#RveNp#VaYjZ18kwN(Q zRol7JCoeGjwQ z-k%^gj`K^>hRky;X-Y&RZ@)jE62xaW3^g&5uX_qy0#U9g1|ORBMnN!3BgMQ|C7B8X zQRYSqYd&s<8(m(Oia+Dd#H&-|+4>|F{d!9dNIq$`T9T@KZt>2kbQO{2RqokjmpweNIz<@D7{Mf~W6NC~hzA$Q5lg15i8Z>t(2)Wy9LcGg_ z(GOGAD=armB?+v{R!A>WP}9EpiWv%Qrb!Z(Qa#H-oL_7W5T*uuLjozO*im5}o#!m2Ix z(cn`#uY(q^;7v~qpx`NF`L^qYw26J8P1ed#l3EyP);HrYx04Gbn@!O4gw9=6)K+`BWCS@+V}GbtOUAPxN!q}6O6a_O>iu6S*t%N%1@>6C?F=){9r z=RKX#_e16=Kl|=+ixWYDby|I?TU|y>LlxHF`m7r!PQRZkdr=14=i{4=C8Aaf{Zi;> zsZK3No-*%oL{i$9yJ&ZN9_LAJF~q5G2bMj}7C<}yUM%aJw*LCZ$av1~8{pg_A>n$) zLEf6=*P2)7T|9g3cfw;I%voaon6|(3&OQRaz2GeX&3(v0B}Fmc)PL9tPE-GxB>6Cm zqCANtVTyS*SE~_B=Cw|_Q}RKrY&-sh$U5(kfX(_DD`sBFJt>5bdv&KKUJ978jC*@o z=#7JOn@(Y8w-~hYyCdZ6_D`j@>pW_B#{HD=b^hJo|DBF;&j+Re2p4tz<7rep#{Ts| z{k5}pxnhYatpRjnXGQ7S>WbmiJxoujC^oe1@d|}$ech#jW!genPAL+u>bU!qlF41q zVHt??F)?C&7aYtICj?AJ_qwB5!+>KT3fz7L7#05+M^UOA;GY^*Jx@fe$db=}E%)@R zQsgmjnq-{F+ZM?_)K00`=U-bS!o@`b$y1ZOPu7o>p86B=2z`xY2?|OD9%p>xh{R^d z&cI;;vpO-Lk#GpbTxliLr5p4&G{r#=RubA08_Fx(XN{zjm8m(ZoXUEJ5^8qpZ&+PQ zlpLz1s@d#)E20Ue7m>hUQ(9s#`Jx@gwq%x>STRsH`av5W5YkXJ9K^VJE{A+8(;ib* zS!v1PVlfEcabnUQ;u9q$Zke0+FE<}{Yg6XnbxfzEz?nrLZt=sss3W@$9FwUaM_ypU}e}XUy#k0M88MoH^~U?aLL! zOj=K#64$`?zMr8X!aY40261OIrYC!vw=D|aj}au0vOH$~vgq+wj^u|CV)pjOryuvk zn!FM&a9tqMSTkf#?`Dx+cn8ESWj&&w6x&=|=CMZ1(JTr*SZRMk6SbyUcb0ZSXEc;f z_F?xdjm>-ex0Lz1{Y1_ehJQ7Bl}ob zmaY+{xYS=1HIRK;VKIoiF_cneQsuf_oz0hnxXY~f8Z^i;9oWnhkLIw8RW}~~DP1dH zY`3NaO($&R6|tAnO38TK?1X&1Zz<+odY`$**v?7U=*MX~Vkqogc=?45e2FMpZx2eF z*m-_73;H2vE~{$fX+*rRSRfxZ9+Q_O<}LRQq>#w$x;vu3h+A1^Oz^3dUZXv0)I41- z;HuE6)wVlb+P(TIp2EuwAnnOa0ydW+q|Vjl76te&Uj3=WtSP7&N zBdiT?UZ;aZqkNU4uEp*)ZHd9ChGLBQY$Y`gb+VHa+o|d7aL`PxOKTE8`b1yVeNRT2 zY|Ij~k+!YEI%br-EvJn8UKw6$2tlxX{n^KnW%eP_Wp7qqEI1jvqlIP<>QCd!2)ad$ zlLGL#W%6=w|H;>gXC7u#2WTej(pbFGL|p*-Gf}z0Z?DE%_2OqB2RD}4qCX}W49HLd zNSzI?KYA))UTs&MHYs2>3}S}zpS}K0CX*e`)PRk5g7Y6u+lTHu%9Achkeljqp6R{5 z&sZ_ob2XCFWfI%=KjV8cf*_|mmO}ym;nhQCGct)7HjjHA>-9^I_^H+T))T%T=IwpAb+8>z`B`+ru zVzxyk8q4C-Ux`rb8lo_l^x1bCdbpZI1*s(v6-I9&;j#T9`I-Hm$#iAR&a_*6*L{Vx z>jXNL=UIc?2Y`*m>9Rr>!wqHJo+whPjp18=_KT=056M%n5cR0>_RHM=Zctx6esJ@q zjovdMM|*Uv#}}(hYfF4ZlmT~J0xNSTeD}~cM&fC3M5EM&XZfG30X|qEnbRqH&bzp8 zi1VZ7GyJ^9>@p3A`-T^<896CvDy71oN1a7?#qbI--rHfdrkk#G3ELt{^f5Z37vG7y z4!hy2ho_(zBNZyIq&)+oTaQm)gWL~HRG4LEiHXE_V3u)c#SW7U4$f9-kykte4C(fW zr9JCbxC8S^*H^Nbnih+usQ{zl#+y^-^l>pPB6M)1x{!|t#pB*3zql^0bicf z6(^Ab0UVNNcU^0bGxwyhtPBqig<0BUpN4}tlukqPvc`HCLoo-%0rLtTx(mxo+CvX0 zD8xd8Wz$4;PZA$$70c2n>@^FdSpF!|?_J+0F@8X4Ytj+Yc{c3s%2%UFlQedYfvKIYb$z{^0Yd)>u}>h zUB><4=dzi3Ni6Cu>X=u+=eFJ-(DfyO@TTD9Ww2D+8!^mO8S)H z@)5uUEuU^W+92s98{1*PiQAbQvOZ5G`eHD^A(S)~ zgu9EZ(K`@(alxywxQQ}L<_hN`+vznww;RKG8ei3ELPA_#-uHC7T%otw8j;G?@5*;A zc0Rss!MMqd^D~won)UuGzG$k^8Vp&L*vU7LzcsC`g5^4%X;Z8V)AXg-36UUx3@zpQ zW3DbGzF=1=dPm+H%U{?1X8t+JL$1^jdIU@|Km98Y}LA^{+2BHOWpy`6l>_$WRwFB>tDuCaN&!$uC zyqsjK*X+BT5oUk6m>?khk^*ISeA45oM6Bo`F5Q5c5Ms?@JF2M!0YP1dloM9iDTUL? z7E6;;cx9g9U}^#nS6Sw!6TPO>3+uDjsztUj6RBU8IwE4u{giZZF>#pgd85Zv4y@2u z`WctyY9(4=KOOW9$$F)H9ilnvu6dbWkk5iATIa9{65U+Y|JY0 zPQ}OQKlGG(toexJ-P(JLgtb*#E(d5Jg>V07Z2>Ah@j-2q-17Mc(;~de@fvj%`yj8z zZb#>qNyL8!q4@e=Z2dEOc!)2)o1&X@xWb5eiXha{Gx4F=))1X>)OaXZ(&z3pkUfK| z=!lIifJn)Di7G4~b*%?wICE&EE3L=Zv|29pX^KXY`$Qj~wPVxBW(O0_mg;&?Wd}3$ z+3LEa@`mV$X2_<-|HPj4d|--=%{6I+<3$54$Hv39x}FFlA?ikL?QgwLrb& zQLN^IFm%GeL#V4rO=ZdPdF=Tq%Wy_o)#wuIkDw`5Bpa}?NPeOR>%>AJCLvR}f7({; zVe7^NJ4p=fo?bTLh2YRC$?d9Hq@06y)VUxo70a8Vy%0OxA_3WbYuIZgEmZBk!D`%z zSyNOl40?>;OZq6`9I!q+s4qC5t~|8R%1gv!8{Ol>jmI-R{EWcM6GR3q!3c&r_V;a&gQ*)gvO%|nq7vs!OG$7a5y) zDj0|nhRtRW9AzVisD}-8n|gcW95EhB_G(oJ4=?JA6F$z$JG=m?ioV!O;iaQK_Zlj} zgX9jK1FKGdo???_h-I}_UQaNEoFfOZeERtak=DV%2+zec;K{1wXjJK>sB$82mYK~8 zXf>be@Wn9%QwwmKvUA!Vut-D{@Ri!^4gf!}E0z7ah-#`t7I7Ww)SPg7WH{Ic4ZS_j zZZ=IFI!hH6!J_y1MOXo-9Dq!z+I)wZ(KNX&%v!M+PN-L8Dt6<`w(DFEs^yFT$M>%M zXo>HPvmCZ}Kx{HFQ=KaG(;GonK4gy5#gku%nK`?opQ*=?3~^K~Iw|Fx&E)E0@4tD; zZ=*silR0{54`_VuJP&_tw!ld;KO*7&?Mjh{%WwZS74F}z?4mSSQCQkp?C-SoFY5IY zX{oE0*@pBtQ|5-w(B;{eSui0MQezkMyr_*hPb zjJ<}rmrlb_^W4wMx`mK$-0zXT>|;}QTF5~rip}_IcpM*gJMPUr9!^(a@7pd~ol+Uh zkWPrhQ!55zyspqm37BCrVs9E|!|jd|@b^E@8zH14Mh~y!vrA|jYS!B34BnqZobQe0 z5q3>HcwJd+*n~99+%z&?s2hT31ucn$o$OL-S@8Ri$WzIsJ`nVOATs{t;vNc0JUILK zGip}Z`goE4!7WWrndR8)Fb+Ar{WtDcg38+sgnOrWbx$Vh+xunYy?WzK2MCl-*dy#E z3;=?>zLFDVeX`{s!?eJ(KY7y^hgQ_0nE>x27tp2ZOn=hl{}GU4R{muVPcNXT=UMM9 zD(*7iHzh1BH2Zb1k6L1e=2ZdoJQ$sDNC85^l22mjch|xq-Wkb8TOPgrbayfeVCOlv z<4ieFdJW#T`X@i{>3Kk}%EIZe@xW%YUu-66-ie8;%v{#vdWt=rlzVnFU9B!bHQ8Vz z_>O$`YwZfN+OU=JUa@Kws5{+GvG%kPwB@ZPBgVz2g^gUBqa6$ikw9fdnN=wCU^!)Q zVdgGIJ*eQDD)!>ES!W1ahP~UKD8)6mkz|~;u6AvFx-nf=7d<*vlRaESlZ!#F|E{338p99!wKYO`O9*A^wldt@pteYd~B&0RA?h!JA!SY7^1CcHi~7HOP7?@~F!B?Oh)xI7FFQLOt&+ z$c>`YH#ZWtuTR4;iWTISXY*C;w8BR(Y5C70tndcwpHMnjEMQ(=-(9A3+AnL9HLf6a zg-5C=Z{gIPGsHKCzYkk4wnSkmkHu?a2vFtiu6%(xzQZ&;y|jiFWdrqSo$~>&6v6hg^?CP z=^|aVCe%?o*R#A_wLk~Sf}e^xhR2QH#Hq+pTT*33$z!kMpgO>2W9U8}!d0OA+BBeK zOxRR*piZ^EzXZ5eA$qQ(%*zW#(~05(9W$SWN~I?>c~wiU&UCCs@@Xb2MSZJMaZ3eM zA#2cEYZZg7!>H_2S_Gb}>&1qHm|84dQS2Bl7iZs{N3s=qv3h5#jI+waqdy5R|GehM^`LiFh!Oq*3My&hE<9!0=u`n^jipEP0AXv>`>wk> zx1Ffugb`-!E_Tet=@-Y@R~)v>gZj&|No5GgIG(Gz(0b1a8-lhURfP2?glG1ec?5~irn=(T)wsBn7oBIVdm40$uj5_!pP}o)Tuw&Sl%UTk(q&wf z9z~Ayl>g`SOM zY3IvBLpuK~9S;4H_5sohF$;`5D=9`B4&wG;CLXl4X_dV0sitV8PJ`L{w~j^LN4t`S zZqcVoc{9?Cc?r+=az{m&nfi(dbYF)F4cwrh0B=;u3fA<9qu35`e`ZDUwQKTo61z0z)%A`MD5 zgdevI67=FWyh;ORAdidxiDYd!U@CG=bF(dmra~T^5uDd~_d&JEhJ415uERT+zs69h zi}1K!xrw)=hT&nTsU@+}-FeaFiP0zLes|}N@mUgz$l7x5ed4!CnI57K$DuMKBiPc253QfC~8n*joA zhpRp0$jFBZgJ_KM;CY-a%!hqbj7m^-33#~A2O>*0my%E4FhV&+bYoA*NR&F~{G4T) zX~|0js?cMGck93LBfLtvylq*kjp;hzSs*{0QfMo~`4@6k8p5F9yV5gK_aV&%B|d3y z{b}?^d{+vv3Ut)O+(1YIHA!pJU+Cpd^*nY!@7$fLRmeH0bvRT!*vhL~Q>?bff)d$S zTPG5*n|&5NKwdVpeCAS<9ea`Kc|FUx(e(qfSTZGEHv^4&-3XS0LVNXp} zOGxd?o=)i5T@nd)J12wUzbj)$ziFV^;D9JVlqY=h^W z;vxxuc5HK{8F2ocMR~PQNhZsf(+dkGV%<3y7sB~z%JUR4YKe`s`ydmbe;W|AwZ$3GopoNEZ~L6RG< z1X>NcA!_fIts89shT}aYm~!S^QAE%qo7SsdF3x4Ujdyc{;HVkGR|-JpyGvc)Ba9St z#C`6!D8UZeGE1U#z20k{CNp^)uk1Gw?c0935@_i-5lQm??4(`r^;|rr3+{JwQ(aFM6l`Xp{9f4+ z%reUdDesV6lW>sx->-MrL{ewgS*1AZ&K(Y?3cj_NIyD@Q{3-+riFzUmiQ{b44E)Nq za-A{CYx=Yt+G}Jqj7BR0wnkC}Xl15((+W|xC!9Xu(y7pAP4>p|N}940s+3u9nNF0- zxuhqArvK0?GD2!J3;zVbo;M|M4*zyWz5lN|r#HsH&*hgR@Vv$`ZUmXjqRm`jo-Q(n*lea+?cuIArTAVU+^|H7B$*Kq?lY2!fxO_i zH}yIPMEUQdKirEAi*{~)n_ftCzBoI_<6@0s#uvn#`HYD6%bEN4WW-hahmvXbyWJkE zdk3JOm7w4OY{AjZq$8L*=$@=vAsR#Wkx2DpcEE@bQdp!{;pIN+zdaw~MCa_blz@WM zigk0P$>+XI^6T#}x20owoEUc25v2>?iFD@%^^Uj8Rl?NZm`$=s;KSo%m6}j)N2d;H zj#-33+?3}Vgs%{GduyCdm({$X^Ebu(!*-c@-S*XSGi7HK3CHw!-LAv~hBkEpYweW&cCjJNvN_v~V z8`aE(^-4Gw^I9aFQuG_Caw~3=k?yipt<=`qo4nJ|t@zMI=vGVT!Q9dk6-4544p2v8 z#7*W;n}#@xP4k)v%Ar*c*d@25AzA=FNbYrY4OYIG*_&NM zzXDY;G}7teQ{o{VIEEd``=6rZ(+F5C(w+$0k^X}|(zh#sk zZWrjV?%QTziWy5gZ>0d!dUW;?O$o_xY)pe;CeNJFs9&AE2W^7%Vuwf#UdmD!cj7%F z4&(b+s_FzFJA;j-I}^bmbIp&47#K{zHPcl10(X4X))ia?%edm=RWiMMOXT`O4gyaRead%9Mbqr=g&QHfHx zYd5@JWE6yfkhaWbZ3Az zCaHO%Wx~qn6;CP279B=d8%^go1mA@?{z2E39-|H@iUo9l&(%PoJOB zE{Ga?`kOTO>^@-s{J9V2PX9_U$I_OkVd9m1#4k0SbzA@2p>z@EijJLW zPC4%r(~w~T7t1sIoeQtROAyGw!)Dg3&__9HhGUd89_+>$lgLt7vgXhy$+>6^PDg7( zAb^&|WrDV+szWc=3Kp#wTWrpE>j1ogQf^dbi$aIub~qsz6oPeTc-1v-omZVL)9uv~ zFo<*tRru~^H{>+~G$FAC#7W|&aSme$NO9qBSWQmVh@=yU{%OZGDdje%Fqyl>Pv+GKufPP-)C zlPs%UDdY8wz!xp{3=NU@K&rpBN}FHWfPEU>;o~>JEgtjyyDu(v_VlaxMU)2Q4v&cc ztaePs zk*-4I-kslB)CPG_>h^58^|x>I`St%879rkpbG($--?`fc3ltFNC+*hgV1{6}j$*m@ zv8J?FPb7jLyEs;a=jAa1_F%2!vE#QapwpC~*$J91*2BZC{zM|V!&wm;uJwr&ECzeh z9M@p|auapV#`+t(%>)SS%_ou{qaAE8!i*+Cm5Eh&t-$JzP=cKt0U8~Od9Gzm1?h#a z0`E}T?v09wA0M|;Pdp}%01_9Cx^2gp{c-QDP(X5_u4uaewMOZ!n)?`!#2-RFMT>0^ zuSrill_C@d1qeZxXC?8#2TB52pU$~z(49I7uuxeXm!4r;xw0X#qwMaCcvf2PoLM31 ze9_8KZm7sM86*zSq-4fMX^3*?XZ|hu|BUtAA>?o*`l?U+E%VDCs=#};1j!U(;u@az z{MR_Wi#jqvlwVTbkr4*m{GG8>_5-SI#PVrA$*%<+tNu`;(Kqzzy8yl?do~Kht><3R z>Y$dVPty2!JJjng4w92&Y@AoRX0FFN(Ju<9^}50j+!%({mo0&A z+Fzv1a7{`rtjR*TYqv2rvsH@hgpmYA#qEUF;d}N5apC2t6%A~$+3t=DgXUX7hsZ~p z65UhNOQ7w_nbm5K+*j4(p<}KvDj}p-f(P%~1F^ir#urYuxrVCp7~YUX@}F&U*~(0w zROZ_V&yRJJ+UM*9M||IV;7P)6TWnv%C*nYkLsQWQsMH35C^cbcSdQ#Co+O+)51`qq z^kE{=FR(&r)gteczox2doGLegidH(E5?}h0I&Y59r@KtDSx$BYu$#}ww#%Y4&=Q}4 z9Pz)av7avuGZtX2wKDdFsJk%7(-|Je>nntPg3usxNfR6IMlbiB=_)C^S6EqKFquqw zYIqHZqq#$AWlRhDFV6{68vwWg`Dw(+=zr~kA91+wgiOLCPmHtkkmq_mk42z~)1dOg zquFU`jmc;zIYK=ZRBIGlu9Nu^aum*RzBr1abbCCWiq=r)dim)9=OWG|Pd($7L<~2- zImb4f!e_m8c4z!3i{D=JY~)x>2$p7|-_kYCVtMp}qr#;97R?1@npX|qGmgap|7+X} z6^7Z2!P%h_Ib_}84N9d&gMLxGQjBiWi*@GJbj6r+OD@&7HjVsW#w;#r(`22&;UAAnS=Y{v6SQs!P@qv1S^#K`iZezse;ZoT_B7oEji? zj=BAHeW=}S`8lOOaWW{JdT6TOHrY?nu~d4h1eIV0qTrY_fH0m4DW0 z`6ALiJ^iIE&Ei)tfG@QI8~z4DGu19>5qF)3Qj8=V_x2utCE!wZ4iN3=`K4h7nk?iK z?U}|E0TaXVc!N~}*R)x1(kCb_-6rb9GS*f_*H|npC%*r!wfQX|z%uoe|7{(Aa@JN( zu(tt$P8S|!t$1qA3jm|IXAzqp20oMo2#_=OZ+-NGwN{=xxc_DN%(S(GAYal@$>W;JR97Xo@L%tx8r{>ZQ@>OgudQMwcgULwsF9HKaiB0 zkp$w(U2%;hD0qa&s5y{34KwJSiBlkEGgTG?)_pnG5-e$*;i2NjSVffjr?UQZ83Y>H zbj+ITy;%Ra7a~QgUDpLd)86wQ2#-PUDggvEf~x*;tJUS(An!un`xd~=>l4SD;50!^ zPTLF5=vEBGXinGHv6N;gUrC*hBTGaw(r44Pc@BuVrOjybN!-X#t}mcgG)f7+Q(vnc z4*H~b`Ge_HWI1X8O0coX!8a|4IInQ`%m1Q|;Q2Io0mg?%aP;M!9(zA%2%t>0cf6_O zn@ByRJFl~L`)1Yycttb^2dC{MU@$RAAhMITM7K2D(j?sm zlSSl~&jkR$NV|-Ts%|`*=9GB9uEJg5yhu+x$=SgsWVu4?HE*l4I_%p1IB`V@8ebN;dwfbaC!SlkULWZ2X5{s?T3K2m&m3$07P1Ulml3d-ytLb$ zUTt3b(Mg5Cm_>B5J1qjjfkKgtH?x&Cs7AA;UrPyU91cT8LuqIyTfPvB=yqN>hrh~8 z^6>!`#*qes*#BhNYIQDt6y(-wLlpW~g zd^Sfc3MfL;oZkfEi9pp$lr52N*!RYo46_(jq3PK=H>ttY-b6vM&n{sKn*_9h*|B zA7oV5j@Fqc^fI5E6nQJRg8FcufGBU!AMohqlq2dx@jdL_ajTV}9QBl`A}RAoW;^1q zQSqq5w9^%LUBeh+()ErrMT5@Uan?JPtiKPsI3CB3(A-|2Lj`{&xyuX zGq8oJ0FScJuTY(9pY(Oq2>mgx0PqtK{)?Z`5b_hvP_iFW`Mt;B=?zRiOkyph-yiLH zBa%%Zb>EPghK&#EAD!1-xAoV%2U1E2flLRkq1BJm`bO^u!icp&PbFT0+SO76j+DVp zrM$J~L%atn)vnKMGK1mCRb9n&*8?VA`#nCyD@ex&`$W$>9yNLV+kHvG0fNiB?7>tr zFAKh&@!Mb=L&0f-M`!!k3!=v&ul6pbwmjhA5)D)f7BrkCx`41y*G9viquhgx;6 z^o>Uc?jOY?nmq0lyfmG5(-FLQ|B6ngh8W?uK>yRN2SxWWe6K(X$97=s`QP9AKo4hD zck1J{!j1V)jimtZGNTZKzV!v|j`P`>4{%jX);_!LCcH0!u1WgW**#I_`g?$X^Jfuh zKr0*fgExQ2)?g}h@bH@KsrBE@VIJdwwl3cYbkH!LQOvD^ES%u7Bw~AFJI7g=4k9jM_gg-KDWW z!yReuCec5>K~@guL0+JKX#CXxaIP0V|G1)-P_o<)ad8lrlS9Aq8+!gz^)s|6rvA&f3_aMxoIE8 zGx*}!uUq_jWxr(M2QIv`5N;2d|KAZiA89uH5jUAXYzzD)HkjIopcPz0v=JfIe_ZPS z*^emlpcBxo<$Ccy?+9d)v>G(00xi9N>X93WU_+utK4x#i|L27MA*UbcQGoo@M+*}E zp}zQ(fnp*V(Z4bIOMXZjP(UXoPmbXi(!cKT9|p!u2k#i^AMjHCYvAqrh4=jy6%z?SmjZ#DOdOUvdvRsDWlsmIB#-i2eUE+D(+Pk3e3=8&sN@{*VOE z60jcHF1Pc4>v^&-U_FGw<-Q{Pms-uA((euwx=cKG_}^e8n5G}R;oavuF~4Wb*9Pys zxb#W+53&E_f&RV(<#ki+gLQA2tpzmXVlOmn?SvF9%G+9>O^F&ZJIGC&>@TH$(V&6-V6siaGt{bNp zy>{m@i&G9eHN7r7E+?zZc}l%8TvnkGryN##m8V=*P`;u*7FI6b8@EtMKm3P3O*nH- zN#)$xuWsB#1joU{|Bybud21e9hYai2&%RaOATdZadwu6W4@3Haj~^3^^s&!(3*p!A zKJb~LVxq)N((_&Z^-F03#IygK)c?IH6w?2bDfo{3zuiare<}g!#{W~Ez-Ih^RS8s7 zvjjORWf`O_$gtmLZ7G~D<9_RkvW8az*{yk00otN~ziiGoO2oSz$dL}Wg1-Kbef!(e z{ma!s0Utkf<}=@!^VFSj>%WX)8xF9AMWHc#XW7TeEWa&eD)=fN^9p+5t>0GjU&i{| zlGd_|e)f6whv96(A;sinz@WwYYclTH*mDTo*+t3ZzefLGM}na4)yTM-@stw%z9VTK z3b3w9svkj}%FhqoZ!!Krjzp^U8R!46V?Dn4|NpG|qwGO(M`16o;L&k#)J5!Bh zsg*Ut&omvoLpQ4KqLAwA?9@>;kZChX+WU?Fa%omc^LxfcJNZnJe!$lGCgb?ksx8Of z51m4Ds`$~_I{Rh)clK`zbU<+T-Ci)CLC_tP)`@LbvUd*X&c3d&#A`b%Na~Wnfh8*& zUG>Dqw92ohRd&$k`3F7;;C}g3!d^o(_{C$bH>@>PUUX1PpI|zfmow4N7=%k5R(muo z4Ts&3ZH9B?#E1=N8H6`d0?*(tqA8EVEk~nB@9UJcSmT={wVBzMJLRUk&XPOLP3`6) zWoA4dX}mFr$XBB^*eJ84GIIf-+cuZb=us^<6G_z^z{8ed*%0wis)5X5(Z$s~Z0ZGj z>VT!=;^OG6Gp+IH6%_%S35qZ0O8DEQZ0623>a=LWGJ~3v>rKe@odLr9>#M@uT3Cxj zujs!B7%3&v!N92yN5$-r*WZ37pWTduFYuFDO|`%M$OcCa)DD^n7+|S ztw`;KQN#mA*X*tsR%T%~8`#OV<%t8%$I6Z2ELxr_NX(OCSrKRVEl%Er(a7r&2CV|w zOpUrz=b4=Zwt9$2nI|6pLgY5aqo=e=E_fxMpX*A7=~vo*KkVUR z-*CL1xc1`h75Pe#B`neU@?Tek9TC1F6$RHrK7XzVTw~z)2D|_IjGPp{H^eD0Yro&| z+zk|T-(#$u(sDdqCs^aTGC(#QIP*s&mlIn+ZX}nL#lZgISajzdn$#nt2G}|))m2IV zle-$N+6ZwJ%TUPGnLtlXypFU#K4Ubym~RB-KHds7S3t&X zwB2vLE8Q6Ohl3Kx6h!2YS z?7^Y)!^39ZcveQ0XynsLYht%!3B_-AhbXl|Uro}p_k4}mzI)9iWnn&>5G3wsEc#N# ziFZ(8V_Ke4L-l&1&Mq%jwrb_;`au5Q$^(AYM(*PSuC{R!a>jdw<%LdtK(%zS2kA{NS2*BWyrAc zlX~Q(E8Np}P+s$yY870!D=!DtJul0%d2d|@C^YBC7I&&`ge#B!j=rVk`haYv{1w#8 zV+!S@GwWs`GzX%8e4@9ypD;PGSK6{eSZN8BrLNqqRjEPckR^8CYiGhM<~y5yig{Jc z+=Kp%Ay-{QD8QM5Bm7n*&}4@`rD|r;1k95-JI3RxGco6}or)zJx|UHDW{?VNYUpC` zb|SmFWW<3#fgP*%p`JuaV$@nzmecinQcS3<*@FQ#QmfG=^xSRw_wmt zTBP7RXqo&!MzyV)&-ILN&`ouxoGDMV*%z;6x?dh)ei1iTf)@MEr+(5zC800Dxzmp2 zIcFv(t(IE_t>vd|n6>u%Z*_OFBg`rwNm6uf=`O+!*OwR`%_UvfRIqSL8Qf56O5iv6 zQNSXq`OFYu%fmO?mvGu6K7-r)P*H^H4XuqX6;Q_ByiV4Skui|qY zs>159MQuD{L)~?~_rS$WXs+&Cr2I|nJ>+}%WWC3ZGbw1Z_K|HRb)8-J{*W@@@ig9o z&O0R$K26~K<4ruL=OcAaZEySS+mKU!BjN4q%H*?0yD^%^LhP{g-Lj)SMR)e0RP*qg z7!5AF+wsUm*3x_L&Xg+&*e=LtUFuZ*ffbz-IR>KUmYr-7rvLO|s!f#lyF!P|DoH8c z{q!jdH321CGWWemxM7xEE~NFKY$S15!YZ6AQUNv%OG#2s&-HSfU8T;|<3dkUigW!H zN5Q*x&&~-@G?UFLRW(~|7?f_a|27~5v;_RDQp3xk;!$l)qnI5I&!vQqTGpx>#agRd zL#Q<}<%iPpH-nC^ffe?-vb+!C)6B+Sv9ulb(mwG)Y z1!?!8483`wIeLLKdajn{Qj&tbMO4~+f>B5o%I0qiaKKS0M>;L?4 z(p*iq*%znKtpEWW$|M=;cDbn~Bdc3YHDa3soku=2M%?sRTrOYqO*2;77M{4a+r3ij z2;~t&XES|h>Z?$jI?N_BOV%>jYre^#vS3fGeJ-D!rhL5$xyHIoyHs+!Q!iO?-Vv(h z7Lr$YrRdKNvkG+h=v=ODT1C?3dR^$?7tFC}%`rhZ=_4Mo{X_)^B`#u-QSn2v8Ehc0 z?vCvt-QgyFfxkz7quVuXc5ZC+tpIF?Iu&EX;(_smcFXW}KrU+U$Oz*+nQoN$6Hh z=yqW-|Bh)^|8zeMCru}HJEygt$c`7XxF;O0;5~kd0SGYKl zDgo&X*c?G>wOJoX`FQxzOkXb+rOu@l9ok>K-CK7-c75u0jTf!t|Keir2Wb*yBJa35 zd6kt1SCZ>ko?|Otz{3Rou%fg5oBn7k>cx#)iv_KOZ(CaLW4+b4zrK&=Ia-?ttJZG3 z4V@$;pbUTecxH^in7g1g&P-TnRC+?vu!YMx_uj5Y?dXq%7`=PrMXPm|u250_K+9!@ zZ&;1Dkq%+u#+4rE+|R->uKO)`m`CyFFXm4R&Z~LM*ID&wR=eK2Y#dG7&34AQT~VjS zTKW=Nb@k7a@3>^!THNXSa$>O!;q>Z_segG$S+w|}8_%`dq)HH2&=c1>$C8VSLq^$? z?4qgthq<&9PDI`#Ickh*-5f#zVc1gK{V}wODFJ2TWdeCI>}FFB$eoJcIvG0ZFWRdl zcMeYLj?w1bHq5P9p?Ng6D;UPhJbJ}2c&(QhLvVy=Ig$;1{P5IT>o{^M*0>ScxXbK> zwzxXFsNEgOx`OWzv`_z-LM-?*%#AXNecYK1lJ`fLK>Drbr*7E7iT@Rgz`cA~JPJOjcvdY;GL}`UvREp#%wf3nKrm#< zrTVjWWnDdI?%%d2d8m}sWfVAh&plBO ziq{!3O7Vh z&J07cG1zmpFWi2=!=bBt5M^FI?25Sz$UqX(qrC^}yLj&aF^A!6&(uhME%#KeY1a>1 zdi}j`+E4S%HD1+XOYptZA0)|Ctl`#i5HfF=OfiMMoDI{+4_g^^S-({8B^^^iV5;I7 zl$&v|Q+?GRx#G3zBtnhdwJWvnp8-=Vu~>F4(}`HeW9%5S$m1@ht-KP{`V%&8Rqw& zj9W+`xWuuo8aZhe5-6s2tsj)H%Lx-&c*US{i}%PCu!5E{WNx>5G|{bc>q zG!GUj>z#$3Z2{!d01n#?2@xC`#XASl8YkioP6thIct_$e(%i`H*y|WaHr_xt6t2}* z*FV0?V_i_xQTzmRR*Yjdy_{Ils5%w5`x?KROORV9$r-`7wbw0NFwq>Ev8a{S8f(4U zec1i*+0`-APHMhZ}5)z81)Qo@hFHOalig?h7 z9GQXGuo;BU`dSgeBT)js{AwfrF;@LE*Ebz|2?!}0%V8DS_jAvmNOa!+Xc0kj#R_CEhtd$6jmMklZ=Ez)0MUh_Ra=#u+ zeAjIhiJ0di9pRMxl$`5>tj5DN92%GzaasuPKcR0hhR|rNG2+&IpDTT8*WQ5SM;7Wm zTYaGuXI;N{zO-gEY5D0;x0C!HMwfhN-;-L}7NuN4w(L9uNeyjd!u8cRKL<8pk zlxl1HbMZX8>WK$N@jXU==*%nw8pAgGn_X<%#z!n?W>ewc2iE+#o_t)ov^F|uDQB{5 zxYJtI?0St(B6nJ1vi&h(GTe0M?Jb_Spe@@}Jt8%1u<>cQyyc4eweqp{QqO@y~CQ?y6s^Lh^RDCIDilo0Tl(MN|AC@P&(4PN|6?Nha>_jC`~|m ziAMyaMtV;u(wl%J2SO3)gkF+B+PCqj$M@d%xxeo}9-ajDUTdyd#vEg=k63$a20To! zq;~%bN!q_MBuH+J&o=8gh`EWt-obsFv;(<3(RTuhj1)1Bhoz>cuQ16C9+?i>T@n#% z9G@CUa)rh-o>m$-c^X4E>O*dmSB#h14vQLI9j9>sjCZ4MOZP1E%%Vso=tYT=rDKcu z<$yj#$5HSbne33oTUo2TS6^}*N9ahpy0~rtyr`^X5l^EqKVnraTYq|3hNm&`;vER zxdNDVs%zt9k#RO&%LBo(+n&N{y_mVCn(zHQ;N;G+=7m1ldwKQiP8WFPR~jAB8mE1X z+}`|M3R!0-4v3?&Y+x_jK_rPfJPW7G08?PU{LaT#+p!8l>-}*Ox_B*_!7K;$@?xTM z3jIk=Ky2G*L#3&UU~npp$St=RO=&om%l-!z`MvJt?*NV~0Mxutx%=}314zKd%Rs>q z?{518Fgr{qE^<(C0I0xZ;5ghpX@oiPO}3UilnV^n`s=%b3P zdBPf#C$fBVb$+QQwx$ex$X7LFgZugiNAPuAGJHUXU>j`g;GhaWW=WXeUt`SSzUOm$ z7dPv|KEAydFE)3y90KM4W^8@cUBwKTCKP8>jIC>W0x5WsIVN=%jxN?=%aRJJI{A6X zsUs>-(7&!>f=;PsRCGA+n{gxgfo#9R9&EIFN*#d9#-RX>y@PJShbf*~;gL<=>apK@ z+;w`}cB4_JWaR5DNOz5Pc-mJ}o}EG+2Gqto4U4A(tJM?SWbvoRwjE?|hco->Ttm*M z71};Tb$$h7nKzG2S86w!D#NSH|N2w*vi*^;Cj(hCM{fB*5KrKY!QNJRI(@?tZf@g1 zg(SV^nl%QQ3|Qz9ONYe;2oXOY8+8tqEw}IVXp3ZUl-QqP*mX2x3Pm^NZx1Nxpi3-8 zZUy4i2|EE`0+w6qbwX6vR-%Z&xYXOmM8@{&geg>O#~aXb#F9p`sNfhvtmbR*XsJzj z6#V)iT^()v{nrfy=f%^<5p&D$u|8dbhPK$MV)4hH?qOa8Vb++&oP1y>WeMY6n#yac zUbmYb5}c26H{7CD>Pb6p(Cr!QM}St6Q7LiVU>ouZY`OIO+|7{$e2I@aHcN3>467hEGU&1BYBU^?B(_5 zmK$;5Ie<~8@)+dyVD)p4VO?m?j!q8`4ZVk`HD0z~H~CBF57bg_U2|zecLsbaFzA&m zs+Xw@SE?{J-c&)OlCa-GJt*yIJ`=UKNmYEDPf~e4A&FasZ{5VK$h8^>iB!gCIyviC zslhM56ju`za(0&kG~C9oSdnj<);Pp9$U2reTi?n(#^Nvgt>z`?agbu3v}16(#e=!W zvR|JbwM8;|^UCd<_2WA=mGLcB7|P#(Z@iLh$u<7vE#%Q4c~m8!Zk0MqyAll>PJ_3@ zX@DWkkw%<*nJx!qgj3CwV`-T)gb6?E%R%T=l_Ud%6ge=vFZ_FcgEFhrbox@)Wmqf| zvKTF^27d<3L0D!<>KKD$16^Jlk!lq0lbeo4);8ov9W8nN zS<`+@t9;BW`nWZxBWC1rOYaw z5sXmAZ;@h($$Ilbhk{~;^Kzo(Vw=~F>zBf5V^IsP# z%fB(QHR8B0CZ+?iUQRYzek0zj#J-X~o!g(XSE_;7({#75Q!tCUw%uR97Hy*RXvnr& zyJst}r@-oLkw}p1UImT3phPWrh#a=jscD!B-Uiq0mU0k%v@>>$ifkQ)nywamB-C5* z&N~v=9K{Me13qe7+i01+V2tvz8#h2bDtfP%tn|6AxCgt47jlHHPiBfv+#*zZI%+nh zm@7H$IJ?%m0%HQq$K~W_^7Ni;8yMGL=~IXvo5SFFByCg7pF9Cz$kibS>c(t!V{O-k z204G)+82Ta*F@G*s%Z4uHj|VSgJT6VdnZ?1d)4h|GarE@l$>XUh;j84-MVTKM#fDu zWp*fqe=}k?R+0MN{K=8I3+N>t_Tw43*lat}+S2J497Jr+o^grU*|-*koY9d&J7+lH zou1s`X7|PoIe>PBQLymXd=7%r*R7?^9G~ZTRjYg+jukMXJ}~EDT#o(Fg3)u5y8pTB zGGXsW@-a%e*b!558u~jq%)1l*zTZzK+q8cSL86zqShc!2B?r;sZJFrKs4@uS& zpvlHh%5cO;zu>=TObLSJ_P#T2VDE6OFp34cn_owq2?qd{4k=hKX2yMczR!1f+&Q^q zpa$|;&T38RbMY|O)W))NAwx$AT$z-Vg^)y0gXTu`sLPVP&Z~9pTIqLpc(jNZYgo=m zfmgXkw&Q}B$sc;E4KiJLUa*){DOTElal?DWWui{%TMc904H@xTGSJvr+`^Tvg4_mU zE^ikyiMkOj7Gj0vm`Mq+b6PZq7pN;KGN?meyP4o(LxRlLblGu#LFC6~372`<#e<63 ziZx-WQw!xY8G8oFuFZ;0qw_D9p(D-Y{av0675|J^Xi>TTbm8WDg}@1%wYlS8`7uYA zq{S=v<9vs&;>SOLi?*o&6C6vijv7dEwwYAqHkxwh5gV~tW^uDXRAKI!vZ5_liXd|JEPwn51wv>P#xwiIw#0X$XOTA3YV zjhDMH8a%xf5c{*IaeyAn3cLm(WxBZ-omBn>{P_b8-&Yio{fX)Wc@FTY?|F_@m=utB zU(E|S&#n1Q2m0_8sG{(2Vwx4r@nQI}cDs>T4423ssz){IeAlO|RH|FX8==rZE^HN@pQbP zb$d|{i~3I3BKfO5)7MBTNB-<$wWC3l%W=y7N|2&Gs{w+KPl`2N#65fLc`JUhKqg;o zr_B_x`SC~r=3``-G1JX5((dZkSvK5*03bowTb)Z@N_Vn;B|ZAI?@sLHYppEL;AHe0 zhqhct>rzs2$5W*}6*p*8#+30+?sJEP$$O7_Mw(h3^l=Q>TFEI00a9`CMsgcYK>BYT zlKMOSf(|&`nt)|_wO!qO?2%KNL$eQbY@-2mQ$Dcw?}(8xu_8j94^99*D#QKBc{unc zsP*6u6p)x74zu7I424!+WlF06XH!Puz@gagfYCCteexdL zR|9v4jBE-j&AIbdv)4PP5wtHGRLj5~kBUd|~@758Tz@Q_!?>TLi6 zNZ_sCkqYkrxFq=$J&(ck0^K33pZ!28Zu!gCId@&eg!C=Y-{6z6ex*mIfoq>;w?4_| z))D;CIoln|Fa<`27K9aLfXQ(TIekA?slIVK{Y{Hw>pI2Dwg|{-5K=8ZFQ!NkMfv{J z>HT+&)heUY+u$0@iUFd`r6pOJvFHrIt_Q0It&{H6I|kBj(pJhBWl%^0m7#^bp99m53obfCuYOfqsx3_FEjlXgo+NZK%CZP zwx7R?xboAS;lX^)L%{7-%FIe{+gLOi83gy6eYu~#Z3G*!EduB?2F|U_EhAgEgLXw0s`clauEy=+^k8iPA4Y#H5EG0po>Jq!&R`E){?SYfO*7`Ra^I2qn2UcRP z4gwk!rJ>_f0e%G{gw@OIW6Kw;Dr5hytE36a*?m!S`Dkr_WSW9ZADz*u#qd1VJL%FM`pDAPi)iYZxfEQq!Q4YI`H~g@_EAiQO#5HLO^T z(9H6T06)$rVq(^_(ZK>afz15j#3A?FdtIwYyjgMvR8{b>M|eG zw-9jf&FJ%HFJPL>DUQP5t<#xXNmo92IGWZYJ#EcdI@b}ni@6 zDPTq6$tL51Y4b4-G7ta$lWb!J;}RU~qS4dS{?L$w-OQY#Jz#tm>$4q$&XY-V01d(d z`%)IqtaA&IPhNAP%o%IgOn$Jz0Tt!{go%p&fr;*Z_}w&nPo5dP+Vbp|M}EC}M8mj< zarq``=fS9zaoO1v=)Au_Jg<^wHJ~Wqw zB?5|aI$V;d>7|2){&gjfE%PCu`Tr0~{2cONcdN$&%1#ujmaOS(td>Z&G`m}1P-{|P zZLFnzbK1ZD<}^@m-VtX4i7OvKyp}`V$p(qhJTf`vd{(+8fK$%6NxAFCBG=5Z zEsFuyX7bjY$a}Nh#??JtEZTFbF$gPrp*&!i&;}h^c+MHtHarD0)>f>o&=7}FkJQeg zv0((R)z)sqsjih#FWa?-y|PD?z|n>|kK zL9N$7vQAM7RNLgQlp2gGFq>1P#)qoecr4m>I`pmz{0q`iSjy$|BJ&NE24B66f?bPX z54+mQ9h6tUC{L+qoD9`B8YE)|5G9gDvFw@korMA&bUmEcQUuh_LH+9_t0z42ded&n zO!$&0^1+&^P5J?T%Rwrpg-K$G5W<9vWv!6EdjCya!^41LZkc=V$$>h_(ij^QLV^4_ z%ya#gU%i5F`}#l_1M;ddg`{5J-G~QWHwje4$on6ZV4m$~QH{Atv;WYd#6vj#$B^%323vUB0%m_bnx^keSzj*?`4#8FDJ|!JTza6VIXsQ~+ z$J49k%wYU7)zNcdLVP>HwE5k-(bWV9epEOGwY7{JFZG&@QzM#Qn6FNm8P}$oq+51a zE36yzt|&@26qoj<%8U8e51q5Qd%MG=K$J?18}$H^MoUk$EGHe717YnD`V%U?hyB6( zXl}rFsB_Z1va~C}$yc+4!}$U~llP_EwvD1x8FpLx>s_~|B?S&O z;`Ez?-@$+(3<7zXrG#Lf;sM*XN>x?tH^Sb^%_mNj{nAB16>OJPYj66|{jWtRWgqMI ztvY-HxWNEfTHgGJb4Nh+2229pM>UDLg-_Y|_7zevqx|L1#^+p>=~_+$(PJRa{&X;> zT0;qCQ>~W#$~n}IAdNETb0s{yuH-XDdhB}I%h6`nQR^Y^fM`%|!>xdKJ?w6|Ykk$- z-`d~|_l30n5MIdib8VRADgny8$;LqS)ifFVln$Yf3orb&)6nBh!k7tNkxt(yGuXLO zHBjkYIz*RM>nR@UEAOQ-fp=FaFoTgEvBVxR;6hc)pvrTeo#)wbyIN&?LuWr7`K{Z| z+UXlY2Do)q0=kEP!@n;dP(Uv5kN!*+L+;QY0Ns}X$Gv>s#|a(L%_i~`eGpn#om4cE zf-gg?x%okTDQuPF1wh{o7{3EFrxmX(?>n5N$YmE}zQ)H3>b#D;V)SNz*DXmxTj`pa zci!@V*?OrjGUvhY#i@UsIzk-{Yi;D?5iv2cW%i7J+ zAf-b}Dl94H>%|l%D?0QKYy-$q&f__au)EwDMJnBLf~UuemL4!YlhH{`QV9!JXJJ=P z>3pNfA6)cRwph%FzoBIKVGXBMu&EUfsSc?nlj*yBc`J(1UDU66vT4(s@!i$VwYQ1~ zh1g{!Rw^mcS|jel-68NJt5eT(fVle*1>{_Lp5)6;sv(6OidI2=8GEJUG-a%}%DOay z#x251jlYs;HyiN?`aou#L68)(0af1kayj9>W~zim)^!d;slTK5Wl|-&g z&LjvWdR#YK`!)J|+->frqXlnFo{`;sqd)88b=xAy%O*p+@tcLjV&JwjQ=RET7h`A1TZy3+R3-3zQIxp zofB2~zUtu-D6I2?PFo$Cm{+yv^>QZw*4w?(XEX0OGxXPt%~pT?@H;nU2S=c-8aUGw zVf-Suo=Y`S!{^*_@PoQ);2`zA?yS85>=|tP`8Ba@<3=mI+s&xtNtD&no~CiW-|*z- zC95od9Jy3!RGAog#qbSAb-htm*vyg|Eo^M%@%$9;xY)g%sAVE|rV!cR0C)p-O)`+@`2G- z@p??Ad>A!Prw0BAN`IdGhPWX;8a!xKF8^U*eTrU?RdGZy{@zFHn&Jnt)QY00BgSAz z>6nJwkpe>kN4n-tF8sv2F;k2Z)n9hy^XP{V`kN!nEA0~ynwK&b1vuL@gJmMEkU@sVJluGJkrRzKa4!}f@lfuGkg@yX zD(!Y*q5YqS*)?&55;Cj)_C{3cfOLye43G}%Ogu}cpK~-2A1EzR$dlGQPGhso0Tue; zy03a0cRQ%IN2etMFEqf8#!>%_%u4oA-{o$f(r=W8OZb%p+e4ebLbhpiumUJVDj?qt z1dB`8k%o&r<@yiLZ}gAnKcai{w*z2$9=N~}(9-=h{?5Z8h$MveFf1Tone+fv@gVOF zBegGGMyXXR^TxJKGPLD=*Nx$7+~=+w37skLCCS)gogQ^tah)RVpoNKiM^3}q@j0o( zvM+6!hW-KcTJ~BP!IXV&@V@)Lnk%W>xZx(46vI0h-)2$eu*oS(%k-Pzk+7^d3*dOG zTS>R7gDWn&rUv|p&>*=W7ACgbP?l>+1e{XmHI{*f&jh>Veg$6X$Cdc=SmeA=WP}gC zPivQ_P7>aqb`XJm*Efqa%4d5J7to`=AM^@cKW!PS#rOsQ^u#lz{V*HcKxfy0gc~)q zV!wKFty@MVhPqs{{n7<9qoX+2J-LNL==C^z2PN=p`PWmZyo$MIjqHb|u^jk~sfTA19+XToa+=DVlJ?f)(eQ$x|7Q)VV9{~L<4psoWA%j zk$0i>4$w^U_FtmIH-r^(}{8H(`K=z-; z5-XHKArw0Xu2uUi7uK3y&bT^3vfa)RiJ<@K(+|qE*LL4&M^~|?C;mMG%KYhc7t_0Q zOn#l7kk>$vP?kW@wa_SpDHY}TV}g##Iu%_k1zTZW?{apNb55U{Nwzuq(S{!5^!ia| zAd|(@SI0Q)?S~HW@dRfhpt_mX{Q&VSZ(_2>kyVli5Ov0{@LVbmx#2YrolgQfp3&U4 zZeb3JKp9^>#xmdiN|~5f=j~;OP0f_|tE)rQ6r@>ZtQlfP!M*v){;NUzI@LGED(1sM zN=HnMZ7d&(p}=-@U;ek#YG4ZJ$=_HZzWl(xLNz;I89{rEaVf7=O{VV>(N9&w2w z8*MSOSZ4t7Cig6pUA;1`TWz4BbsvrzA}0u!FDD($BM4K`rEUCu3 za(Vn*=^lV;QPNe-6h8}Hh^47^@CGRm5941(2?vgkDHerv0cnUQmJ4~m4nHGf*8*s4{ zfCADX2VcTZjubsFw065Re0jwA51)}6T49*V7>5MD^{Uu09MX3Ct_a-jKMaoY56{i` z?Xh3h1@PS10c=Z3$y4Ab490kJHk?IrTOA-r+hX4`Gi~9IUc5$pd%o2o%`aEGJW?W9 z>z@>3yY0j?OvV01&w9}Q{&0IVl#Q!j2~SKAMJdKflJ96TDS*dkMU(Bvac(u@yZPl+ zu8=@*MIFPbbLZkz;Z<&VMaQ3EzKYc|m4+@BWxHk{OMIS3JxUEwuSrT|2YS>Q&ei7C z&NJ;=(|zjOVdrh=Kj-w=a@ao6g`*b;Zal1Z)dDDh|NT9`fJ<#z>&sFf)8E^l!vz=N zoiKl8B?YAB2&<*#X@}M4I66~Xmqw$a)~3Ay%a2`#AIpF6%j^(;4}V9@o;2F#E@IN>|^yS1CYKD}Lk%zZzESh^oO zVMUWlL}rJq`xXt#Wm8VK&iADY>?5nfl@68^veUzNZl9jv+U;17(%TVK$$f8yCA{W9 zlMxoJsHb$UXx+_Cb(`DYnG}hnZfWf2Y2^hL#|=tf99l?VdPZ<_h{X5Ek8!3?1i_w# zpc*&F+Gv72;Gikd5s#*+7Y^GmE7IpRX2<^!VYS*;BI+_c$+od3nxG}ADG900f_*qz zcL*K@P;nN&eHfmt&aD9tR!oUnv&Hp;NWem1%35Wl9qtA)qkZ;uVwGrtV|=78i-lz| zsZmdx30-WBqeOCh%u?6C>e7^&~!c-kidKwgvjWLUi~XMesH7;-owT!@zq(eDDyA9u}X15TImdq z02k$Z!TA3R&I;WF?nLO!Nv4@=9`f8k0u9mk?Mx#%E~6550K~C1Ndk5EP>#{zb_RyT zIJ=q|cxNX&(l>D{6NPzSN&;*eW^Z5&k?);8LfRrhB3@7WqEbr&B9pd|qgI*AQx9|i za^)5Kv5i-N4}BRRUL0IH+w;&cRXneUX#|n%H6W{&gfKY|3{U(n&_L%ewFH#EHz|GG z9MstBHnn+Up#*DM-p=&kuVC3dAZC6Vci~^xh71_W?)7cvj+al6=6a0pRW3_~5m=OV z-i`+iOt@w${ldZS{=~t|i@`nszjuvdNJBm-kpPD*ee^c(zd`|Y;M(c250G8LU8}Q@9;OZ$9#YX+)brr3WK9u!d}jbYuU67$9{SG|XU z@6^yafSf<{@zK%Rng@b=UL?4<8}$g6g026ZfLLD_F|Bo%l_bZBI;I2Cl7QMNuO155 zwg%hKMwL+WyS0g|_>5WI1C zcxW<;BDb_@n1_z`90b_BA6uCd&f1d$YiazWWpkK^5a0=;^e_*tXttP5&K`4I?+FR^ zh?Ox_>?+zEr!guq@aBL-cAHuDPGOvZ#c3|3iWrlbG{&kVdqTyVSo_}3w#_&6dNPjo zB-blL_Z_o?9oj9t67GKNr#cdJc^o}-BuGY}T>jNDzxaciE-BNb?2fxt-A2pS<0I~E z7TFnt@~ab#TWK~hyGp(I|@+GY841%C)+ zQes@z<*GY?IPE%y-3?ytriI_3KT6euZ}rT{ZS!y zBM(grFx)EVJ*a@c)R%p7pph%l3UhvgLnJ;B&bkOtixXsoG`+!(WjuBZ+*VvZ>pYI$ z4&=q|l+3UB4>>=J6L?W4tz~!4orm>_P`2-I5s*D%sKjgwXq%LZ)<4)=HKF1C$=+yT z&jeYkl})q$C;?2G&3s0APq!P2COVwkxX{m%rWhRkp+`Kex9`08X5W3$RP-PbulfLB zaovy!?*@j{_0~uYM?0qJ#ug{<;|EP;ekPsG9JveS0I5`6a5%TV*Joc%!d}&)@CBmr zLG6n@DrsbE-MOUNP~(7I+}&nBe#ufDjj$HGZ5=Q|0o^90Sj8^CaZ&7(_nPk*nk@

TDiN z@;0Bta6kR~w)j=T&`A|{pQ7FAxqP}-qTT4Bm8ei>V+bMB0lO_3KEgeMiQ?Kloc%>6SBpZ~^3E+(VQJQ_{+2vcp%WBK#QACVbp$ozmRLTv}CZ zSV(&BHkBMUGRk%wK_AOaNnYlhVmd>wm}94p%$K2W7jhYAI#ifGc%eny!vH+syChGiogBj*E)aMN6RrG$n0(_GV_l>^AxdrH)vu%sW zum!ZZ?PL%DitPN0(uDDXJ00wOUm28fq_&fa2>|&^^GQNldxLcUE1H?O16fPvrF0!@!PQQ83dFqbnzzq185URt#oPO-@y+C0weBpD;^i?N1=Ze5 z4~C}5u+B;44@&^v2J$k~wr{uA8u+(A1{iWGXD7=Pj~x`z`AeJ$bbgsnzQ6Iyq>s-= z09Dt^9w>z&Cs8g(b7MeZMCvx@_N+-V6=2rAO#RLt@)l|u&KXvyQoJUjKq1c5RoD|I z;YRrJEz!(e451DfoGCSZl?w?_1kM;6HGum+yp1fd;Ctg`I#MXEo;JS@izIGq*TEvDl^yU-HC$)!`3Q2<_$WuJ9weP~}o#Hhq`H&ysH zsqn&lUmv?@CSy^QNuFDk>qMHgL+3|HpEB%3O_wFxk~e9+0cTY`nOnbrG;%bX-(VA% zr~@k6m`>-OV#9^`h`H>JD4zh1Yt##mGimJ#9!mG}lBVykW-Sk0>#M4q5QQ}6Lt85x zAkr49M92l~uI)nJL=`}@L%Xvz2T6Td{cO_=|Hi9-4dgJ>Ev4g%FRu;QzT=lQxzqvNXwEFUp=cGgCrJ8pfjV%oOa>|z~Z?(tpz;j7l;O>|iu`EBM)citN z(GvAV$6H%VAYgtGpRCEnRq9l`ZtK=b4|;{-3lj^eTr{9gDKZ_>5_+pxtTte!H*oD` zRP1~*)6nu4=hv=0V?K*#c4XK)ap>oiA+V-{N&y7WeQpnNPfUIqaU&JsmtzYM)Ay@yqyoT(>@?xv?DeOb!SM+0%4Ip3diWT6K7XNkH`j9B&HBRX z7LY04c|)4s;VOx+?Ko_#oRNBG>{^^xHr}RweM~7^3-Bn~&XkAp$^E$)V@ssdh~HqD zozrLqLGfNW7MJX{;FFofX`_{ILBe{4#V`gUhqj9??Qxjx9YBJrJ}ezAo#V-rJqL92 z$R9|Eg+bx0m3aIDes!=w>&g@$)ywj|9X+NJ1Qe!sdm*pP&Y%3*(>Xx+_%DACClBOR ze=h_)b)O2iENnzG7=-`mVLvy1Djf74jTd6omrl>i20^~)dG_8=%QG@j?4w|;VO=^( z;4|#Of`G(cHAZ2p^BLg|ykHL?TlS!*s=F2E@27U83a{}2Y752%9jU+elww@4`0z8A zFL=HVc;`)Y%rrwXzE1=}O+*0D@0y@`20sJHTeo1u-V_RXRmONc;qiR_h;Y)?%Lj_$ zq(m*1mnUcKLHFl~8o?~tdE+0ag>T3t(%D$s$B-6t+N=xOg>>caz1L|_q!MS|*?+d# z|E7UBp5+_-LnU=bEAOP#Du5yP>W5XQ+;|(2;oh>jg@92+?Dscbk^RYzu+d7K*tRnp z>C-sIz_LX<6@plZ2Z*o?quSS3Gg75XedrfByr9%o!qi;Td;94C|Y!

2|iLZmlvJ> z=}zjgg{_00nNwY_XfVE3amcOdEx&o0&UB1np5kE$Fn|=$h>|pJ45J5g$C2oWDy^5I z*n_DKAMt9V2nW5(re)7o043>5d#%fszOhGNR+y{HZ=iaYpBzDkIe%&G-JSF3IIBa% zc%CkX^`(J&#dgV;P~@_NJCU89QXQ(BL=Y51t(Rv+CGS z{Q5vDxK)2>if4|%VN3eg_P0U-X>-6|XaD|*yIeYRB>1iArL@${gJED>jZHwCc@Ahd z*HbB^1ww`wn61Je zM8^R`-dRQ$g7o|kW^n+Fi7iRu@w1csuQP!6?y8MD`TGK$<^+b9eJZ~9`>=$oz~~p7 zu8-$_ey;QJ!Ja+Vef3=@(P1h z|HpHA>OEbL_GE(>{JU=_qJZb8&%Hd#_9HFs!~y%3iZPD>W z-omKhD93TShpqF`!E}-y0CCkQeaz4qr^c7sTB*{HT{bCILmfF_=wwv>x)8|-^>wtbqU_I|WWs&qcp62ws(&bnC=EuiE6hwhH3up-W zTj5tV5fXeJ5WXNCgrraZc-<4Bz@$5lXGXT5|MPEsIe5F@Rk*CmPnG@me?9!`+Xp*x z`?CSC{r@_F|M}Cb2QY`Hdxq)L{IAoH)g;qn-1V$We-1k*GBGqJg-EEAD)CxR?cxQD z(W@iV?1vq^Wjz6G!+95PU#{PF>erk7{9VKa;HC9+O@8^OS(Ok*H?2e&E!Nq~1-GCk5V0^0N6^$2NZ zEobeQ@F+ShPB!_~vqz^@m6e}nj>nv>(j=V;NKQ$;dX@+~bm$O`CQx~1=I0#;D-#Uf zXADUln+!udY^VR^54~h)aORF4Tm54dJ-s4tR+T_~I+ZkjLzOg16@xP>Y^-ftCtjU^ z-D&=l{chm>i1X2s#DT^~9TQs8qZiZd{I)Z_ti%(WAC`Z~{78?IsR*dp!hG>P$jK+R0{WFN-B)nQS%Q^E z@bo1$R^%P1_m9mzOqap2vCELAefu-(|Fvay5sC;ePrRgGX{Yec?H|!JoEemE`~YReTU!AIr!Q|fSvzN+Cgpk+X zG%-#ZaBKK0EQnLi2s%@acz8H;gP!Dm=I2Vdrye|-H`nQ-zr53bJ5qMa2T#_h=o~_x zO0wP>Z%dQ^Uvd%dK9xVJYdQ$FfjrK7DK^Zxz& zWu#N@#>Vw@k0KdnUfC`o8R7O#_!mq(_XbK<1tldv&70_vT*w}gy=nC(v23i1Vke(ym2!#6%bRnE23Ske zDzt`%wbMJ6(fNY6;7k0JEYPF=;)QbF**a%#ob>R*)i+w{FfMPVQ&|HkkAkd*SlHLtL9}n^aaiKl`w{|!{-N7eE(UM zuL4KzGPSJ#y_Iw^k=O+! zg~9IA;(hy01)ZZd;oNRx&-6~8K3&AZnHRgeJIDD9E@zWPuJ8$h=5d~7)w$_0)S#-S zLhK&YQx)91Q=UnAIHTIgKQiM|SUHe|GC`7{kq~_^^XFq7vE1pe ztFzK(OF@G<;aX8KyOcez zCnhs!XQE*>7nh??61D|29J3S_<0Z(Ii;BA=YyoR{g>1P!nJMZALSa+Tf^V$G+@ksx zDdyHz^VT?W`5{)AFNwo%{2lgeo{owNal{$*CIPk z)rP_KA%XLFWk!|x29{FlRox)I$>de98tvrS6OG{+nJkZjPN|2owHrsKbl>LRS4!$?Xe4Hcqfj*Up(S0Y+0>#`Cm^=JQpKs9*gVQ*IP8=I)IQM+I zSfoQf4V8^VpZ6c}>mrYLZ^X1ef^E_rZ{Zw>w)$yA%5i`Zh1kcPU;N*`_~&PUQ~k!{ zuywH8oA@-L0X($1p}aL~U0@f7j%VLw99rLtn}-^Xl3QvD=Bvz>2Hien$R!pL@*mG*wL)x|Q2&__NrPa;`9! z#jlf|j*VkNaTE_yY2S;YHzrT{gI_Xo{mS&*wVOCW4stSH|Jw0yOZl(AG&xE979?QU z7(6K7vny%%I=H#F>p_WyY$KRg97QTIpIz$UtD1k!sTPw=A+g{fkJug!ZRV@xY$mEo z;hU7xebGE^OEZ^_!@cICD%Qs^=aBjb_qtt^Ny9qXVX_*<1If5~Mc_Q2ydcuaqu-vg z`lUOxO=AwXJB5Ql=X!eH)gqRBNMek=aM}Bzp)Lyw!8>>MoTm;?P*b&ay0rSzmUl}U zkEI(fGwdwo7Ixx}E>KHLBw>)@_N{RV-krFA@{=B>`a)iviU8fTDfeiDh=|V}dko)W zYe?C}hai*ck}+4y!CNT)dLL_2!3JVHKPU?rbcJ{3dClriVZ)Ki_7CUY8P?QptPC+Y?OX;Oe@ad6$5a{z%73&!V1 zr(7BIoSRGT?gOrMP2)7ImbJ=Ij0xQ+IKBTdj=guY14(YMFu2t}ZX|LLqc%qme_t;k z0Cj#<97qG!i;xbDPjwR=A{B*gph+VtyYxp$f+7;waUYFs+FWV^r)Ds6T_9T@gHC3R#W{Y)c(qY4xaIzmJaJB8 z+q=Ss_!{}38ZRgvu4Gbbrf6E;Zx7#9E!Gf?KYuhx-Cg%}ZT$zxIr?co+Fo)Lcd~ho zC6nS(;-k~%g6#^iX0kR0(QZ!+r6%=QscAyS_u*@qF8i)j$8^nfS?l$3<`yxMLv)b^ zrB(|YDlIA%x_p~qSG-`DW@nH`TMs8d*9W`p?HQ*^xy9LL^c$oHaO~PWXB!|S zIvEzu%C_=qmrNsD&Dt9f;9+D)y`LM=kla?OYp+Gy%qugLUu6FnZlE_M=OI6fP-=w&RwdE zes5eIBTafW)S<_!C?KFONBF}bhGbEV_oP%Br(>Pwnwne1Z9n;Y!9_whlw2pTxb+6~ zoGb5;Snc)cQ4idAWMpJSrzd8=d#}A3DJUptn6)bLY3NQ?@QO+;e_3d4u~AvM*Q};? z4BavMS3o{y5wx2q-C_63n zFCp~V>5d<{SO?#aG`~F4taQyB$auM=Ymx)T<&MLxt;^4g2y*Zrw|@NRv0s5PgMFit zA+5hJ&FKF&9{dp`e)7Y>Tef(2@xcuH))!2`+jZktVJ1d`{GYAo`%>PEB&N%6n~stQ zim5mJO1XxZPcXmPQ$>X)h3?-x(Nee>$Z}3pn4LCqTMh8k>>bRBO-;&;5d4ZMS7D2L zJ?#s28Dva1otRoyes$YkyCUD;Z4^E=n<>U>s7uYwDuL#da?d_FQ`DwfBP3NXr^<^Yyt7ZdEFpMJN=6xm?dzN6 zu!%{sn?h2=U3OsBk}2ZQnokBXRp;h^O#W09d}vuYfqYkQ_YQ41eT}q2Gh>)7jvXcN z%GFd`m?5fs*wL2!z%itcIQ0vuyeaTigGDkkRSf+9IEI`XIRsTVRJDa7&9hN<@Oing zhO+Y0T5#h~CFbFM9rTrH+7zq=MgbX^$@xB ztn|VPv$j z+JALmzt(j#k)V3z6IFA4q0gN(mpqcpxleHS59dTukiolss*rI+a6rIzG~?+rXD;!; zh)Sta>)>$8mk{9cTH8;6+ZW6&b2TnfRC0|o_2R@)4~J08+|g$mUH60qrL=N2;m+(A zj61GyO>aSO$@hFPww5WZuD@~|(Oyjx4;g!c^7h|ntDQloyZ9bB@PO6mT4~aXOo{8b z?mm9;>Rfl?a5yuc+y$}dREPdnWaF3-Avbw2@8zZ)J;x8HuGb{@c1<(K_w0+Q(rum5 zTIbk#-M{d_6ral#E%iPTlX`Rh);m568OK}d_0zq6B>T##92N5LINPpiMo~ zj1$KnKq`UsE$=|!xC>fQuWQYU3X5bKtKxQ+(Szb|sLs45#^<;1deI7JT^&D!dcvX9 zM6HnkpR1yxMq^5QA0$6mcMN@*U!R9pb;)b|f9$3P=nv3eqVm9ZE`f#|#}qoIO6wz!R_Ud#-a`=hOe&%?z{mzSmy0 z*ZQrsRV6?KC4L|HajQS?u}(!z_daxL)JJc4U@NDf0Da33L64FRU5Bc39k11ZQC**?WFnxy)6N zQ>Mup8X_SGRjbLhU^26vrb44ar*7Knm8h0OnA^^oFM0a-M851dt`4Ez&HEibT(lrl z{}Vchiuwi>-a(SXP0?#hEXh%qq_RM|YI|36t+rpJyC3&UFA1^iNEzt-TCi*;DI?om z|KtgCbJ|DtSPONO3kJLRA+Ja0dpC(vBM5C>va@_c%m(E88I1ai>iz!V6;jms z{ohtV1b4w0CU+iYmA)Oz;C?!YVZC;*DLVR~GC7nP`t;FLO|qlmyo?DVXd86Gs*-ZY zwN=H5hRc^pR9_!CLpm%qr)1cxoVAX40=4-lU+K>=M(={DN{KrC?okMM+#Qaod~|Ry z^SCNi4A^G&Gqcx}pFk#E6B3VpJnD|g^gS=?^v3*bb#xCT_pl#lD7q#2{g&o|bD6DG zb9NW+Zjq&ZaAuW&!#<|V%oZ>i@RXh}{5{GNyH&5zp%h&(MMtyUVK*BUNt~AM| zkwO=D7oHN>|NP`RO+X{f7!kbhVz%~;ONx%R7>xlwZ zqpL|pw-_9~;Z0F~5b^QidtxUZ5P4jU*u4WpSGENY#V6ic7?e*}C4Vosk1#8E$8FJv zbf*9>du1mC^9?L3_VEJ?ANVrwOyJSmJ`%$QWDv6w>_nR$)Kn*hViC*9$^fVK2732C zqe9ANI-pG3KFuEDI@#e;*y^5>Km}r^GxdoDsDu{jrBQDq(uz6h-zH@m!~rdKQa+qW z^m$)L3h?zSt*A7g_c;Fc_^H;G9CN(sRUOkU%G)N8}+C?Ce~Qxa+LMw zBVvu_0&^nDNyGUtOtF)Wb?MXOrTk?DA>BqU7K8trSyu}K)4znw8 zPEYy|a2DjZ6p@nlxd_p5cMoQ$_yppwgr3*CV1zZH{%@Oz`pi3@dUOdE)Kn{^|K^zK zlh}b5w+1_}`r=3cgs+e05^{CQ8B6IaFY_y$@!2V@)KqSXG@RwwrJ|O+f*h&;H$NRHoNrS z=5$AuF^AWO8z7HMPm}?1&;hM4J!E+UrHetPLX-08`*h=mmF@dTPMv<0eaQr?&!ImKVR!uFL{+?RZLZemJ8k{Ji%$-KtMqED)%m0 zk$xkB`-3i|CK>j)Ix_7P%A*vw>gaDs_LQ2gGo7KjOtzTGcmnxwOe*vf^dj7UOH3!A z3Hx|&Bly6k87@b0>K$?KZC%Wy15rg~9}$P~v}rs-g76pckhKTt?ePg=Y>Zi1E|pSK zQcC_JP{-%t$HSl(iF!{jW^BRz%24K+CNMC)Fx|pMAXrcV6|` zG}ou!PAsTWqX;HuxkBQ9%u}MmfB2qLVZi6U(QZOcJ z($O0ja&9Tk+iE@r1_n}2YBBeH18&pZ(wmp=t2Bnn3;-zg+>#2b=4|ywn+1%&*K%gL zGT&?~d1sS+5E{^`_J(KtVmK``=R=_S(={R)#9MzJ6ZcpLqo15Ayy!t*-HD6sWU^BL zu($)YIhu*m=|)WmxfDBPn}PTW4P`68BSvz5t}(cXh;CQ%kP_6^vD0I88D`tG923kZ)BVc3yYIdb~1izatt&< zzcPAbC>cjhU_pWxUcNrA;So(C0`KnkhxHYd^?w%WMKAQdrJ*J4i}=JwS3=fl5wIKa zQt&pVXY4JxC3@R5^)}FQ4f9V}*rzW!kiGehjQ@d2Nk1wH<~$cbaAM9HNie7EU2wbn z(VH@4A?N&M2O6`EtN^xbtMCV%Fe&&ZeSe`bRGRDVX4JRXj}u_Kn{ijg@B2#HeZ!B+ z+dDQdf4A{SkrS&42M*;~GLjS5KYAqs*HnOr2gQ}%9T(tfIB^I^+F|f{C@oa6R6g_b+DgQ}zp%lMc&iVR@N-iIKSW9_!FH%EbL+ zUL?ag`FUkj@UG&su%WO-$oX)9rlF|$K#tn(XN6uFg5L#0LueVZOcd0JhRev@hWHB& zo>f!k-S|k_UdFUXJ)yM`Xv?n=rq~h0mquob8`&fu`X=te&Q_D5 zx3w;z^b1&&EO06vFT65BEm%|%TX4#Idh7{aSZHV#Bn#&>*6oBSd!hrBc$q-5OOT>I z+?#Fq8uR%=w!Gd_?sK>wHVimmcFKs96wxddx~A!}pZ~&fDEeYx32N6U0^TdT*n2`K zMehJw5OjiX36OVIcLS^zvnBLjykcIdUp0b`S)P{2oG73a>4}9I*Mmg(b}4VePUb zZbl1|)JD>jG~V9cRlJQHJgMRI;b|jlH0wFygjBTL*#ifTi3wp$oHJ}uL!Qf#ft|4z zE?s3dTnw%4Z~{B%vlTWarWut_chlhOZN6tWx7?>* zkRT*M{3x3B52}Qvm5X>Chkl_hMya564oPoCLqa;!_Z~-hGcw>lNOJl74O<;@W6Od+ z50;{m$oIS-T!mVg=y@!_&F>*T@CPwvoicc}0x}Qo{f0T<`oCm^>Fz@!ouJ-oEB3kW zf`Bh?!SB)-FX9yK?LjIwHW3O1MrTRCLm-*DUPddH-abBXoO2XW*S^#*;0?QUZmc%k zW2MW#>+2{N1Nr4r!5IMiP}&QWTsfK=ZupC>VKwsdBNf3wd3a-BoNJ_)PGl?(%7&Se zc0tA85|2=vo1DZ@zUA@GTDGu@JG+^Xr)};8yER+;F!2EZ}0v`Th+Z7Oc=Q z_Ln0ls0uFso*5?3L)jrW=^91Kr^_OCS~fN|Zr5>P#(OzNK_yqH5NS6len(ULKWS== zpx*mN(i_p8oD=drBt$TSL5fJ2aV`QJm>xcU{5Z+_Re37ylkBQ_jV^A!g3NIx`imEV zv!LYjJaJ*7rR1AQcw!<=hHSb7U8-7C!9$QjYHVwZdhy~}M%iTBmTu{&j|4zc3}*?y zAyco*R$B4VR}K+Jy$kd)Ysc=@Nn0o{R-a4xT6x=%0N|V(>2~k zv)Z*^A_f15)!5L$qCixYg1^Mvn-i{VCLE1`nO#GoN_73+eUd+Qi7SfVG7Fc=K=5|9 z3Z2xF{%_q=V5EVfEX8lR;OqIJdw+GMN!jJ6gGF*gD2+~&?W)%jBlX4n24!~Hrw;+6jU3+5Adn-KrSF9)*z7Z1# z>TLcX+gV1mu0qd`tG>q1tk=H63~ru)a+wU?Z13abpM%q+N;a?!IR%Vry|v$a31C&E38y|xoqDkev|3S{&Id~oqL7o`wEW1 z>6rd)>ot`=_!py{{o=%FHC6K7zPUM`{Kt(r`TNBpX`skh3S^Yx-33H=zvG%Ehw8LD zgnj+$Y1mo@odO^Ucdgc~>FM4^#b6Va%Hz6uz$u6(#XM)Hb!C_fW@p-2Pl$Nf_nbj7U#&wi-%kwEwMt{dqKLwY^@KJZK$~HauQz1J#&gXCns{=$DncLySq~~0h`fr+-5Q5B8YKv^?WE1=>B3pq4z4fve6 zl{^lX$R}Q&UOVa7-8C0{+1vJ&_A}aI=tV4s@>JGX?tJ#D z14F=G6;`OO=H{RIqDL_xicsr&8-<6)Es;QK_9Q7uPD&j*(vcFvklDLTMrr|rI2#Ms zHs-sUQuVSnlV(JP1+HQbF4aG~6jgl21g}<{RAfl)nqXIcd^Ji59-IH;CofgpUygvD zI?foZY|WvxTG>onNT?~3@?fW5xsILqQ1!teKue5%zPI{#fhR2FnS#={4U!}(TA6qq znnN<;s>7{T7t*yR$gT~nWS=Ki$Tn`9?>sI0aty$L*mqdA+eOt62wu0gwq9LhA29en zdM*F3<_;bwzMRFuiYDIKMpofS3#G@fO>_30UC+bLUVgnjH4{rEpZ11^n@mlf*CupY zb7zOL?g==X{3IHB|LH(|tJ_6=Z`nr3?X`eq)PSYQjZiV}Rg3C_YcNvcK4m{A<+rY5 zIfp(25D|n_6j^;&M{cL07}1_aVK-9oS4Vp?mDD^}ZtiR7GX4DlA4 z%71tW8P0Q@UTZ&sEtw@_i-kgMrkeUb%>$g-xg<@TsbI1hvei{5i?_9Na^$_9%3qpk z8=nmu+|C@=GtIJZe88bEA)-B|Z4u1{Qk;*PwZ+mm&7f6Mhx&J&oMH~Tz26PXakq64 zii85RpAy7oK&F>mckn4&Xr0hd(Y~X4$+r7GY35Z5I_Mk9>%e&})AqRIv|)x78>_zB z%{wx8t2_F%ksLX60I?Ivi19>{DW!;$Bk zsI7h_y=ugAxj_#(G(N-TlCP(G{~2eCe)6)R{%)#=`@0)0!ls7Hdw#1;C4>0m5E1oj z(cC3n(@q}b-zYlyaSykj1UHp@Of~mEG}=Es_p>y9*w~#_#Im17POw`CmMmcGpZUFN zgX2lh)mW|_?#hKN-QqY5mj#|Yjl<~ys!1xE9IA^HtXqTjkk$`9k-hanE_br)l{@Rj z0~1-0rX!$VyppToeLGYveIs|kgFBB;n$mq%$K8A@*|Razx^&HX&aX;LKKEgZaQ5~6 z{9xh8mYg#t4ny1uh_JZbm{jV$bv^>a&go?P$PHDoDr)O){nn3NP2-Lt5%#(k-w^=R zq};vDI1wE7s=-;cYsmA>gR1Vy?pq5)75xrHWEI(~@-%E>_Ikf}$)u2RlTyq!VIUU? z#iOjGu6eHzX?PRq8((tQvj7!5;6KVIh9<$gj|@jAR}& zZ*!jPXvqA~DTFYys+LN1O*6lqk{N7kPhw~~o0&a&rmQGDBxDp50nkWWl()5H$+{1% z%06ZZcCKUzKAF7IJu~3A1lbb6uN|>o2+>v8TQ>AOB z_zQ@rKG?{hbZywP@d3ad0%N(httyfYk7bVGl{b{hi1f%g#k$$~gQ$h#y~4qasI9?k z!6K_(!!ks}@4o0Fj~(7?*F=d9+%j!Z0Ud^L%%pWuPT`dZTKWvKkqThWUWH;06CzS} zQ#^#O`u7uDWk`O$`Gn;1C)Inj8EFq+t$1T-okDfO=XxXrz=l}YtE%=-H#I4nt(RrfCXeK=?;E;o zWk&0U70HP(!m75a&bZG_?ZsVo)>TrX>&rJu?UN(DDr_UYc_nh-u)kOrB;jc3=(@Lc zUtPWJbiWut4I$YusBEmsrm7wcLdFx)Oy0$r?a3Pf0A_13hTksyQEU#mtIj0qxuG0Q zz9rC>Kd=`MdKG3_fCjr6zFO=n zXP2uMJe%vFmd6ReZkJmUenDKsUDwH(05Ylh)YcH41*K)ymU_V8z^5u*%91TvNa%s4 zyT>-O&Fpvi(mLGcEk2Scb`zolp9cI@|MME{R*dueTKj#f`F4F$LQ1K&-NQ@u`#Uqb z8*KK;0oC8dp%;CNRrP=oe(`-9 z7o-AG*M`+YK(3^GDduUQnR?fku+L%*RUf7rJ`OJAT_$I}BNmc;;tD%dDDF6F1n0q! zbFrGl7kJ3v)^8R*1*}EfJxYE%HH|@pSKj#WWE^CwlJ2!Z$7B}AOzTcZi8YP-!RPWa z3pck4En98d%Bp7vS4*d+<6Ca!sQRDvR3CQUa+iO?>T30Jd!3M_^2P>h#URA0V<-RxyByl?2h4r+KeNaP8m#wbZhIh&L*+E%gt&zC6odPo zfqvLJ<#w13th+Lm%c6s!oja34(YPf%xyUm5=g%e%6LU*M*y%B5WpL_~A&Z3VbhW)3 zY*$~q+*n8DVRJib17qP2;X6)dq&>gl(u2rcz|eRC;n7Yt<#=TXeHV0N0Y*yi&pgQ` z7hNq<|6(W>R93h>Gwsn&{jAee0QS|LJt?flltWmL3UL2(v%$8@T|{Oe=r%t3)-HkA z$)GYsBy!jtXI8P|rjr-#0m3SYdXJPwbm+s6LjIeFO+V#VeV?1oF9|q=o*l8gq-^1y z@FOoJo%YFPUfhww0b9_*NtgW24Abgv#LFEW9V3+}tkYN$=G|)A*F;!_v}JE3v=m#& zJMQ9aKQxbIDvPJ%%HuKZ%Df>%rc2kY|0M11{SuH?ZhV+(JJQ;RGsZfzSC8v%h1ek6 znd+tB!7j)v*;8#J?}yviaiQnz zmblfnO4l~0I}REvy4Q0ddF3>?(7eVADb*|fH&)GO9P%gkiC6o|tUk>!)95^sS=V>B zK%&%_&8*|6LRv?TVD+!Fi-c2*Ro!zwQ18`I9mpBAx@s0{b9lz_4I@B>YBo5ug6Krl zxwgUfspOo1!QI{fvM{l>VCClTlMc|o_L@HMIB(BiV%!!E&L(SE?IuUN zld_2C)DJTM6Jwnq+2l_1j_JWXjwSXGF#>-%J2(3bG)kpT4qU6Ac}pelU~pP5HXCas z+mIm3c#qZmqtf%>mw60&Vv)E<@Vb++vE$@p4LL|8Rmw}A1qV~eN5m#M1G8%Gj|}<9 z(PGQ}zW3msqwhuT^gEUlgdHF6X_5*rJ{ambNCth|Bz?}4fp*S@&Z{r$7YS$j*p_3y zovm$;;jYhgI*ilu=u6Kxoq8qMJCR2LLiwxau)!1ukI$?30)m^?*el<$1L*YduExaK zwcg^Huf0nDwd*yRovN{McVeQ zGHCBOEm%_wwd*Fd%xtX zkLqTBCTaL`=XY_%>mWI5rF5YRFqx!e#^h%C@{el{&CW{H9y>TN4Iaaeh+f0Gmc9!{ zdvPoCkSOEG#fO$F1<5$6c7Etmoy?~UC&}kbP#XV^RS++eg2^{Eb)Pj zcBh#I8#AaEtl)F3GMJfcaNqhAvb(X+ch#}GZUk{~i@C}X)}a!o?HSbi9ma~d-{ZEK zK5UFvYiw-Xj6*ebIxn85v?FKKa&CKwFy3&pg3iUZX?yH~BA*2*3gI`Ny4v_aEZ9yJXX?H>+?TD|Hv^TxQW>(uE%CAQ5vP*+tk% z>XZwGS(5V*8pG})#fKke5HT=iR)lgjU#D&%sjBc3^(TY_Eq5&3@{JN{6QsMkN3AM^x7K8cgb6PS{1U4{2B4CIekS0fRC19QJ-9bX zB#}`(n0}UgLR#6WnL^7%a8#}E2JfI{pK4cic3ee4M;ialS_vr-WHtT(UYz zJ56lkwxq0NWZ^>9t=g8u%9#SJv2HxqBV<4E_kgp@tkG9=#U71qg0ipG`L0--^;x_u zr4`Ir9+QZwDj2UtC@7WbStuT|s}ww#eQrdzWv`nX{SC&}Tc5Ph1Io@fO)HndvHzS# z?PS^86}gZV*_&USlCv%oUn1mWih#UYTn5T(tu3xtg69ApuF>7Btg6Fy%2nh^AJ^f5 z?P$muDsK6B)l8ymL;ON?AnF;)HPK8MUu+)v05++iluJf4p+l+f=CN-#l32dg6a%2m z_2BkrR|65dR@BBWkn(mcR#mmI@p7V|q_hyW$PUH^3 za7V^R!dgh;Veu5b$=Y`3u9L@Mrq^D881W81ae}9=Qo^m*4C}(7&p-y@>#!iqf~P-P+iV%Az`~$Pz@HbCf@j@ujKmR^C!tpS?_1HYN0MjmF;ZS>=~)alabHpqk1lEoRqxOvZ8WI@+lw4(p+K+g`a0;N!;epjHJ`)FH}~jL>#L1)n!T#p zrBa#6Vg`Dcs}HIabrLx`U7}md@?&rAKA3oYUT@lCipz9RsSnXBGA+E|nqJkAQuQJw z>fPzF*DfO6d^Ul)%uddW20Mu4=cbD6J-w+8ne3;|%yxJ0VgKO+FO}W2K!qn$ zIkp|)0f^^&3Q)ZOFby9eu||~qWb7b-Q}-q351OZx>XYTEm@B4zg$TKQ$E~EwT`Q2O zUf11_gG8~WC3P1$r-Vdqf2D2Y$YcFpQdIJD_O6C48cT8PT^~Qe-7w@ltdy<_JtXqj zT~JH|myF-OI=(*Fxmxo)xFJboOA5r0x>xFD;tl%(s2X?AlG9!rOQVE%?3aNfv=K1S z0-&aQ;hCE_&MW@u&82E7NWj4CSkYmH`t|GASI@ZbM9SE0?5cCVZyaJ>{=`>hH6oJ# zbmfBj;gaL()(E2IiA-Lp#_rZ!I^=7-z(Z@pKJnzHjD;$$;X=cBWBKkb0l{vuvMthg zTN_J#%5hv}wSrZ~un`X9&lj}!hXze@GU0B;vU14!7EM^x;-Wb+#GEs-o!aVr_wMtl zX>w>deIU~clZ1u33Upr~3bbnTCB^@;2K zD{xHyLhaUWi7d?~pCL8t+U|W&SzS?O^2B*L_MGz@lA=`2k!!one`Md?2URrdg}3L| z)}WpZW$eApZ9}fogv688t>*{*pu9yZ?YAQBHFa;0VLq`k`ELgxoMRBb(tNtUotQ-t(Jkb@?d_WO zZHaf_%2)Ek#7&C5eADYs{0Z}S0^N41&(=kg3wP;etKBwxW%)(qPYqt80FsK#+@xq& zD}yKsz@|rY2Aa*{8{j8--$??4(|Y9XWQ5(mZS*`n354|8c8yo9=Xi&^YJKd#J8^%y zbKc{4POGk|QarG@+|Nuk;=a}=a~+W9 zJXLHQ&I<21x>l4I7hp9`ab%#yCBiF1#4+Zxj!ol9V$e<#po9A}JtsVFTcfNBo`|h^yt$XPj z-AtCTso6Tv?4OXSLOk||b`fS(5o1xWXO8fh5f~uRZ8&;^^8qT+eFnqIDJ3?}LE}w5 zz3B(9nm=CVU2)cd16W#MHop4qAOM5~y*fXW!C&VbD@gx4v{Et&5`8J`c`+tM91kFk z8wx-PT-i#M0xPJ;eK_d%G3;>#ue6LTH*}*P)Lp`s6I+%#)Q~CS;wU$2*MolgfO8bx z8BZPdyE`f$B06qUJUK!`{9C}e0-y%K>7o`spRv_-;ZLyND~@jc;#QD<_h7pvxy5e0 z*YXbkvm##A&jdSUGpzr_mxRRds4X$drh3DjOi>$~S1lSJ?t?08ni0|>k6B^oOvBf1 z!KF6FVAJguHhlOD_R9(lACFjGxcZe{EuH~Yoqi2iPoYV%XQW52&?o{RZGVBIWNG)9 zV`sAS1O619tRY%rFOkSFBNcB#uIwMGZ^~+LM-kEee5Aisya2Zfkh`;Pq}Idv%poy$ zQs*cIS;T{>c!x<&H_>sMNoW;HRW;x2_ycKwWUx^uenbh^CAHt0cgvEr?c>B}P4R zdlXXb?0D)F=6V@^g=SP>pm}3MQ$$Y7;bztora!1ml)MQE&K7Y^9Tgk?`_e``F^3O^ z8#1ikM9aOQu81C~;jq~OVBLlwqd4%3e>;V#0$}%*JDZbTc9NVvD|~!>-^xloJUlL# zVvW$$-VIDPYB}3VVn)r6?j7-OGwU#nG6R@!$nXq19}Q=x(_tO^O~hW%<|>|%vGMEb zLr=e+3O9Z&VI;P!#`zL8t|#RRANLW`EP@p@K}d;>jbzQ> z|Hnc9Ct@fk=l^uj3BdrS=KohEUDH;Hpc9x0>?%(y!`eo8qx`PFED3Fj9-YR{&$l(` z)y4Tb0oziD{Fo@o8EX~+kFOO=!qVdc*LCgBJFes`b~wecrhGL(`8L)C0y z$)=|@9(<{%6QorGIOvT^JPw)@a}3e~=Sb#z0=nJo?W7Re`imFnIIc@)tLC1;{f%u0 zt!zmz-F`OdOG}p_C9U*e!ZXzHkryKx>4r-^N(K%cdDaG(P&1}H@&-x3z5-Z?r7A3J z{=z6s^(hsqNEl5Sv;qBP@V4KGsyMt^++FZWdXyMvec{rTQGxI$H%<_>L>(Mm<3IDE z`k_Wxc<%?uHDpC7W!FY81!EK(S5KXw&7YL>=3si9bz1kDptPyd1c`6+Y`vGs@f(y# zfRSHQlcLGeK^F}LbR(xssl0q^C0H9vdd`b#XerE5FPscvcKL`U{cDR#w&f?Wn{5?<; zFWcQqc2QYg|J`+=gR;cR8h4yvzJUmb<6ZCvx>shbZI+6j_7X zKX9LDxG*UYT^V*sk{Z;zzWH{@oS|R?@w$}78FYds|MGr}ju9s>`k%emm;gYcV%63L^-+yG=B9{Dqa15#dlKhwz5Zj0 zfhR9B&~~c4C;SI`@?Zcwg?%jcC;06O(f^K|{~u+{7*@zZavk3+{I=%$FN0vX0uI=` zZjKfrVepE>;#Mr3Ckm;4ueDOnAj_$r^CqHniD(}oG4zu#U9HzXhE9btM&A;8mhWMA^_4 zUe8p(Cj|4QE4YXw3WmXi`<&L8{Y1 zy1etKP!Rg7|H!xr$i8ub7miyqQ6Wb?#T&}_-xq*dnHOju){cX|37LTgyd5eYgC^78 zNbuJ`{}DJ96~!a+ojGn=@!uCybsbPpb(wYz{m%Hv3V2+ifZrRala#4313?ql*4%M) zO|o|wSO&_{}7YU!MJ*;Ky8=y z^2H_MbX7=PLc7%uc z@^b1$S4ItLv+Y;I8kgGI+7=I%^Bb2u5Pxe~AWDY3x847zogZouX&Inh{#=&a2x>=B zU;JmX6G?$)>-VxGHBmniJ$)HmM|rB6Z;=e8wEnus|27sDvZcDHmJ-_Ha6rLIVz(-v zwEw>#xH4!guSyt&CN~Q+7&Xe>jPK@O6cJj_qd5fi3k; zSemvVHZ(zQ0@Jy38dYXajP*pPV7F^vrI%RRMq? zP?1Ws6NF;L|IWQ$qNguDJ6$1gYK+%%_(p1-%q`c4)L=xD{|_Vbui~6ZPvSC@dqoF@qUif;6-=Wy(n~;PK^h5ur{Dy(ZSiCI2AlIhSZlO4a z`Ux&AtyYLH0m-D#esYZS%&?!I-^<(cPQbmRRquf(SP8tlkjN)GJGuMccL#XjqNbXm^H+Y3h7!7}M^9K}F?fc)w*nr#n z5mtV{yP%xuxSslZ)NUHDJE>4yNpu$@60SWyF(CslPa>R3EBLnXc>Bc*+;BeO2#e0k zsDJwpvwX&#JsYS(g9q51(lrAKBAz-!(GR61B5B0aN&_xGqcKPQoK!4D zyMB4IN{In%Le8T$I)Mr{v(8o7EVZfx?e>AdbN;>IoCOudo2NIIlFT)S`!iTffF=vy z)I&$uL-er6$H(cG$G;VI4L3Htc=j?dP-4mQDcbiqA+N_eg*qH>wq{+seCw4~OTXQN zm*ZJv2~~QXDx^Xl*}3b}W{QhDT?wRIix2urU6bd=lcoKV?w7Qgh*joJ2@T36?$Ut*?xu zx#!#}{If2tgoZNRt7y8Ate$~on87)_5J=(D;SZ{Vm?kQPizI$51XBpy0QX5nfqvT! z5sZqHplQE}O^FLtwld_#P?>%BIST$X$z;3V23FFPsQLND_=Se*E#Iqd(Wlo1MJ`P} zYB&G+=>c}FTtU1HM06YDOI_3Q5JIM!b=ma#R=fsbh=!D6VV?CJDRGWx*)w=a z(Pe4qCdcNbW5sK#HGa?4CH{_kxglGo^oL?_VWAK2OKr6NWC4dcf(q`*cgj&c(DO_J zZf~R`D0Fa&#l^)3KEn=N{2*SQl}i9lHU(};3R5pI+9@}AwZLuBuk>CnnXOmLR_4{f zyT7We<1PI4M9mGZ)sd6Y?mx0g-d#k`j7;U&M2FOS-F3s7SC1GOMekQTMWm^XLV57o!}O`WwM*$3qZbH+7t0$Jy`K zCYL&>f45>HCZHB-7Z8sRuktaKZ&8Nea&uVsb*M_d^}()K9`~2gG+5s@+2m)=Tm-;? zd}G~l_oNGSdGON8%C~WR4tu_f1D>L?iNdUY1jJzqs*xcT&a1Z}Ki5wA2SpB*UmUQq zR=rqSEkod%Xe&llpErb{Wn#*RU{hAWzuW-RY85{L4GO2Qx#z918iW5x7lD8Q82pngmE5=p zR!tTBOAL=$WRnCL%N>@&L=Lv^MX~F$nf2xs@(|7zg8Ps2jauWpir#Ykj-Vk?n2g{a zXPP~aPqPF&>jmtzI;{d3SNUkBK00hY{IYAG@NJwDS7dGK7Khog<|=|}L2G{KQC7ve z+xzgzSFdasdkN^}62lrhXPTmN2ZVaKGabY2To&~}dk;tTXf5i!cj+Sy2m+C{+ix

=o0w1`pLV{w??aN0_3Wr!UsP8_DzrP>8RQo$%iGF&Kmj)%H?mc9C^moHwxpN?pL zenTN(0P4?Io0y>uD--pV$}1Z0EhZ)=D)UlLcH}`RK(y#K;35e_#c%q~{BXe0hsNE_ zLW4V6#gVs=uv>`eV4g#@@r}0Y(nedX6fra!qg`l%C2Th43hB+o2Yha`FCLcKzBLNU z?pT|XZM$9=1Pg>K8ns0#Lp&X`U{&xMch;$8qvCRsg1{rsVvc3>jE3XsS_G5havqew zswr`RgNh945Kl-jqxq--TfeajC$+4i0441s&WfT-= zbTb-qU3y@WIG+irgtgww#7{eVmACoU{x+Rypyd4+xGomM?>Ta1KN3RvgCaqDBYc}# zxV-wo8WW%Ge4u&;=a=^QoiOcoAtD{<_r3wS74sUPBDoz8YE-|ML2YJace5uTgQcu% zZ_k560@s^QXyf9mH0q{RfgjPPh4KCT`LQsXj_NaWZ@>22JH)f_!ShCB{F#V!9Fc;h zP#CiCEa70W;T19SVpE)A$F~0*4>y4Qw z6D_NT*aQuqqQ&BNDiP-z*6!8CIPc3rm*u3+5_?(6ypD2Xg}K4A)VsVFpfbsib5~=} zd+Y==Pj#1C#k%l+@c6vgxXIH$y-_UkXlwt%*3?1IVPvfzLH|PRQ7p!Lo%ok?J{wNA zDSMl7|Audn~ro)fUmM$s1w!#l)KOr$ibPuwgN)%y|qqE{dHkT9_qu7nuo_ zZSOZ!yb%~u>v!Q)ue+hIZngF6z}TCu!_z&dd7Beutir?S@;)@jpa>R7KYGTX7ck$4 zD)}HC&Zb)~LALHT12>9=Crk{6g{~6`uT?oN@}Z1?y!VyJ#Jjo%&4JqAu+RR(eLdBdZWu#=6mw4@j?!P!(5tU zjEc;3sMf?54hf}}7aF(az*S-q5KhtI(l+-2Z};+t-&Us*lw%0Jk>?T7G(41wrB=eU zwqChFl1!96-I1OMyx~ZBn(N(&H?LoZTb=V@5zrTW-e6C(7vyXidiVST!A%+_!RSjb zSSIE32xRN_nQ!6E2=+YCz0J%l=SefeUZPXeMf_SUBH}U~Wdwr=n@*CjyT-fJ)byu6 z-Z@YqFDd7@pKo}1k$#b5=8QL7_U!a;;sD^ViA9d({klGIOq$6#QGC;tMhR0|KwGd%^~vMAP%G)l_aZ3^@+#W$)no_A(Pv9EghrC*Kec zrr7E;*vg|wcW5Pd&ojn8Vw|)@FU}|BP--VLs|5R8js@;gi^R|H#_q!YN$#6K<)2!6 z$AhxjA*Pu8cy7`;TR9V1Jtj=tS0gzCGAH3#{Bt!cy0Y$5+(g3SYO*E~)qaj;zkq6tQs}qfoM(d z-`iQ6TB)aUWGuLxZ$@1qyc}z*6WwXtCJDJLqg)qZ#u9dBrR=cgn)5^=AsvT!n&8@f z*A43XZD&5}E@Gh;`cHjn!iv}QBpsuv{gQ5*;ov}FHVxWC zWgiL}t@J~hC03eW$L@gr$+d7?9%L2C)1ryzv5KT9-_-cFx$yPY^QFzjyGBQ6QtDRh z1dRCu1JmBmaB^M?&gzMcw)vDz9#)t%&zEw}7sIO4+gu{sEIrq};p&TLr=FpHw}x-v zXAY0@Fg5n;?Hg@R;0u2XE~1-hW974*j(B@7plPtwHr>20U!l^)zOUM|+HHF*;Bt_Q zD77@wH*;Q_E^KDY%%HZc+)&vW3!qx$FuM`q4bPBGj1uXw2myUBG6UG&$&e7|Rm++x zL-p-vHk1h^%M20c7-QI>gANOv0HR^!++kG3Yvt@>7z zC$}3FZ7lja6bfw~%rFHz@2~CNFBUdZ>zQs{4#9zy*QcJt1~o*&!a6+%A9XgVZZpt# zGn{i_(wgNcT@Wdza~8!Rq=-uP)Ha(R>`t@x_q=`m=h~exT*;g+5zg1u+hP~GcJds8 z$kQhyRkw5YKhq~S|DX}q?u7(E_=R^<$gIEG#_0FHijKmZTkBA1xfWvM-QpnpT|6fM z)+3T^A|zP63~oTV+>E_Z?Y}WEVPel2cd|(x&(|dQN@r&avpVugEEs30BT)Gt_>f~-Uxsq4eYVR=Q8L!fs-aRC+g$J3hg zE;EGa8cCB)DNI+v*Rr`e+86h2AlZ84D=@zKpZe@r<3smx@f>s!1^BEKL* zF;C_~0po)4kILGdtS5A^Z!FTw3l*$59T~+-oG!8FsD%$;cx$ectNIT(*O!gz*6v7& zayIWgUh7%PR~mXTROOzrv!4;HK60jNvmcVK9Cc1~rwf zLku@}flk^mRyB4j&aYjN*G53h;XOj3*Zl|@nXEnFn&m?77CB$Sg_hi% z_kJLzio`aSHd>B?oua$A=7SY7VnOlxj(N|!5=lok@U`D(@NCgm#bL7_&K6zv$p6}Q z847VSgSGauZOo*TzVf*Q+-Hf~Jm?J8UB(pluZbd5R82QJD|4^%U6;P8Y`WO+@-<13 zC*MZo;)}k;9}nJqNa$ejBw&mx@U1Us=ZbV^>oI3B#ir zEZUwtZA}%Z&XT+;$!yCCxH2Bvn=(qqZQ1~<)+NJK>Lq+YUE%R`G>dYd?gqaKFQd!T zA7;%_x7H$=blC09db!rB4kQ;uw+TfwnW#MWO8XK?#Lq}@*#1A1eP>vc+t#g$G?k{J zqEg%l2nZ+$2nYdDkuF`jg`!jiLJtV2*g;wdElTg9Baj3Ir4u@#CiIff1VSJoWd90gO^=)Re`r6Oz-r6eDm!SS)srEba;&L!r?>DNvF%n}ZAeHVGRp!=pv65PD2p~zD869t)%+_}%yqL( zJfaF|sYt#DH|~DXNxoVaSV1{*T9ldntQ@IKJB(5eSZ=+|F;K!b_q!@tRp^Ux;&Sg2 za=%lrFD+9aMwD^Ws=A}ub@7_rqhxaPW?>tN>F_88N&F$>zyJ!@1nW76Od1gmw# ziOhY3|BYM}Arahs7znUPF4^Bevhm4IwG~s-H@yTMM_O->u+tgqL?|ceM`sX?(!BX|62<5L`TC2g4iG&iN=Vy3AAae6 z==$O?HscylE-QI2MST+#PVgR9?i{l71J9p_G4jS#-=B04y<7CD%&F@NY)I6yKuQ(V z=lLv20eq{lrXv3cZ>P9D8&eS}RmisWV_IW6lCpJsX?$PaAk+~PYimY#lN{(sQPBfT z;~y8aY#?BlSIU8MrmgCW*7Tchgd_ABu3kBS+$!H{8t=;Yr7~lxWtt|eulhdo`gBYR zN9d>!?=yOT4L__MEVnu3W@kaWXt@);JCnow$Q4eN0hI!@3`IO-g@?@BGB+EhedWDjusv#2dcr&MvbUIzyQfG zLyH?a0f3tH5`sKT&Kdn!UP30n=Ikr@VQ2b~z%h~92TtUmic({Qe;C{{Q@=d9W~1rT z_-hX3FC6d$m4KM1{_QS6I@ysFn^g^R-!~dQ)_yft6d&E@1*Bj^B-e>Y0)Z{%(%y#w z;Cq2AlD1^+rSr}!C=>2CCn9F1=l3~SN)#MR#NVVeEc&#X-HskUok6)D$)(cwdj2;W z_IChV5&MMTq5&yFOQbda;d*!QhqJM>l!!}sUht=FBuvWBRsCeI6(Z?PTUt-|O z%F~jESQ}A;36I^~pz7;BY+%s-dzVB3=^I*+VLW_&`5aJb z3J+vT_jywi)Uyg9+c?K$E?edV;(Vj6X)AI0q8DJLVWV7z-})JD{ABBtl2X7jpS~tZ zEVfPVjJ$iS!DMEFAFU&4escP~k4ofn@ZHDIdDtyBLes-VzB{l%mV>G}4-~?@L zOlAJvQpB}Rtv7iK>zK!#MJE@>wgP&UD=G?rL0aHC@#Mg1l=S9Z1!C2UuB8p5F+Z;Cvh63iDb_KQK00xzJPUoNM>0 z114x_f7@*Srux-+0bkxDh~4n22{w~+^vkjQ^cpKBt{)^A+~~GmjWPQy0~CWF&PIvo z^9>f=4>K;b8pH~VO*+N9;L+ACaS}vB4QA9yQNtU0NmuVfna{8wq|U+>qT@yF(@rz0 zI!a)Th;*FL*6LKnu#`1HOy>GC>nBfVvSOWG9G$Igz%J!Z#^oohZOjlfZ2(mjzE7Z` z3+_Xrn}1}4S_~9De!DOVy?VqIrC5I67L8movBPZBn@jkEE>B|)kMnV(jxQ)Zxzq1H zaBsqxu zcoQyI;zkqhnhzm8sZ*4R@a(QX;Z6_8x{=!v2w>X_CjQ@FkG9zAi|E6P!-WftYN}&= zYQ{Z17f(+tLb~8vwm~%A@>=Z;-mB+<8u_EMaYOQHC)-hX#Q|Psg~{a|LPzdSTeW^- zFXKD)mKI2Eu&0E~R6#%$rU4D6>KR$3S#XP5q!$Hn4Vam$XM3P_E(iQn{uiHF@$|}x-|y8NrEuik(t#g4F|&ma}>ebD;w$5EjOrc&TzEB}ZV6s@OxC0MCU*v-yhni*0pdTq0t{TXRlot=lbZ7zF&L} zB;;DM*Kkgc_aq=D{PWvoM$SrwOFY=s1eHce0d}dI77#w7~aFUmk0au z3dUV{O>T0iD>|>9a{Q=_2orQfR)4*S@7g>u--(eG+)05@Gw6}U4v?HYt)}eHv|jb2 zDe(7>)FZKJjUPk?p`-IZc#`eyKOZ7cjo$~2ITPq3G9Kzbh}Nb`rBzC1bDtL7lKDMP z2r1y*n;s2TE`K^0Rnw&6c=yW}Cr6k{^zg0|i%?MN_#R%ubz7bdtOm)fo< zd#*lehJ)bRXO?r%y#*PmpCmSK^-|Z(W?MVNr657pavqkadY8>3!Y=_K@Z6+r%6+#B zRVMO-A&TC`2VW#n&X4+#ZS^V`l9tuNPzKdTOEsR z(_P_9qzl-=d=2DgNwQy?xtgOf)nPCq^HIxlu9@+(vy^)?G>8;<7z5YM1HNYRHP?#f zc!yX8TO}3$+AD6xcyqfMlH>iRS3fuI0I}To8`4a;+Agxx3e^XeTJu+7N!2HN)2#dV z_owxBl%@niR7lH&#+nT|4wd=O6@cmKQg`5|Sq4JY*G;eD0p6Fmjkd!CSj0PF`sf(v`V^Db2eVB%?#+tIwa z3J634mw!}9Cw}!!^J(|RFj8)9CD};x!h0j2$oW*m{>cClyl6>^IyG0(W4Da6AHs$dR;5ZZqD+kSg^6t=OFc+=TV#aYfFMK~3?!~z>X`LeT1Dm_> zy`IUR^2I$3nuSRa0sNh0^Sr3?eP0ahWzpB!`Bhw`&EePlv{IHvY=G9md`T7h=ju{8 z|FC^+zLm4vr-P?iK=w0M=HSbdJK7J!o2JEDbv&X2E;&0f)>%?B-4#s$stslX`M5_e zmgJu}sLQ*p1fF1_-ale}3V@Y6^7`ZRo2uE8l^YMesa+Z>F&vU8-2yQ9@hEO9X4u9Y zHPnyuy1Q+Kmu5~npG~KHWzf#F{2Fd$?U{hs zWZx`NTrEA4*J?qW)5gq;>5&{N?c=r8duK{fJB|$tjH^9TuR~ISwp83{dE7mOcC${= zEQ9h9?~b5)9W2G?#$ytvHh}Z=&T-INO1MG8>PJC+Imsi2l@*PW1edwiV4XltLB$zl zi$5(ZXHvX#AZ@=1Cy=0Df>h&?68X601hHe}syWY+iLAa2H{G)`1A@l}pWT0GPhO&B zeh6yHvgm8^*lgmWCfT{3%~kac#2jRg8G6SmT!SosUkkFWMpL&` zp^QCQILUZTc(9^!Pvfaa(9aD1epl9@r-deeK#%fEFLcjy<=B#ip^D#YoEiBHuIS6QQ_5Rw^|aPfM@>3LGR2mq9FcHvUAb712w$!Bv^doSYmTbf?y3 zJ}>E}Lr`6e*)$o{|C&UD6r4u6PE47#RjUG*+MIKP@-dAId4+`xq@Wqr2y}(p@VWX1 zosw1(crvdLX=u%%@CdX}P}`#qyJR(b^2_?CXEK~qYqkC>&!I~-K&dR+5&!eVB4DH=00q+wsv|E*L}fR<4#6|bXKYJ6gV z-?%6-CF;}cP>&<*iMOK4|0+K)9=!Q9V+x3GePU)_ ztX^1JP{owEA33a}P4WvfS`j1@V)wlAgGQ}l>1!Xq34wHZ|`>m(8r1KF(J89r$Tl~Hx73UC9J)og)_Jrw?MFcwCw6J8&>Mr(L5oAHX7_a6Mrrw*_wFnGu+QDlJzY5I^!~d*(xeQFYo?) zxPL`$PE-PbUqMzX^?@Lc=25u&-jtIEwprs`UVSc4CYzWuJ=#8nHuV{_b}>AOx@Ci$ zGbz1gGXJ?G(mXBTDyYLh><(kyaB?0@6{n3F4XwmM5op@CD_uD>1`p?Ye+L`7C#B7N zUiry$9?ct|t1>2|U(2Okh~Y54fZWO)qJ*ynJ*69%PNDnImLr6bvc~;JWif2c)8QLE zrbGU<3J%lUwgu%360n zI?D9tP{K%SMzndw^%>O96oo+LiP472-J zFsFXg*{^AP=({mgcf_PrfUa*ciGD^`bNv`%lb zWW;bHK)7AJ-uMo49#6|@RF7Mz@48aDS6S~nkaK>l?a(rgQEg@EugA3j6gHID^KS*r zU!2Fzh)`0UvV{Rg@Y@;P4Er_u`N6W%pzP8*;?y4CnT*dD9(*Ss)>a)cJtjC-cKEX{;yI&DFgY;m7#;>*48s(k8um^ zUEjj|wIm}uc-G1HhUr>snU%1`D2n6MTcd@xnf~^Obb?5^;<#93n{(hj7Ui# z`9&|s9VRrPdTD?WapucFV0c57^fl>y@t}2N0d%<4a}<7M4HKNi(FovmQ04qe8Q*F(RHDwM z5>KVz83x+cp_}lX1!yToQr5ra>-|ZcE}&s&!13%#iCJS#SAe4Tylc=ck~)VndJ|=Mk@8%WCqhXM439-y(T)l*#Ebo?*e8*oR>SUlTs9 zn`I#%iO3gibPT#ty~$e94QvG_8+l+?3(yr1y-(BojHv)pZ^UpZp(F|vykGXv3|I7R z0b#{{7epoOid+0~$(XBk`J)&12(3VY#$$SKJWG2j4d8rJ%|35~yECVg4T}R6YTm)t zcUe!6k@-EDBWDp~sucU%$0yIrOdhe)&&=J79v@W}4>WY89J{-Wh7?}F9?4oFGFW4p zLRJ(Lo%y)atM1n`(}}FciB-(iO;uQj3x`U6Ut7SEonzWYOIXZyCY_ zfQMHKAYFo@@Awrin1DDi!=&EPr80UYt8aFBD9 zMr1D7iqePwBDNAo zyB!y}m29rJtm>`AJ-@izBZqr zyl+6xG+Vos5=a|8bnJ_YuJT1*?%1n_iE9DV$0C%^zjvz|bi&ON0bZ{C`eu#;&AQzU z8ouN_Z$R$(VcQbGFDX5(IJui!EGYK-7(!6s#zIu)_K%MY>2|Nh+mL#xJYJihkZP? z9yh|2)0Nrhn9VT$u=nQF+}xv{?ePJKhxCqZo;X!@sm{;bXUR+<_*{czsl)0mE8!o4 zT6f<{BEly*njT9n&TI9p_Cn9ze|-|0d}f-xLtH4WCE+jLgNU$k(B>G!(z5ozBKqC^kzgk? zC;)S;)xd6weO>0|p3BPG4@vo`#a&1a81601mbu!Vgi@_=!^7;%miB+dbm&+BmDT=; zC3RLyCA2o<8b{#x&C*AV^NL{s1^&$_Twhu7wcdWuD=%Y4%6cG7r#EvepK^3AWR!Se z?+jKOd3vBLcMjG`?Mzs1jZNMZ+f#0Iaqh{7iVC^RR;8E;BAQ~S*5CQ zq))mmctSw~;hSn{+hyJ)X8J{sX=8q-c(fZ4*POT<_^72v2}t+R0SnpC17vu#K8UV` zv)qJf>n3v-miS-&)CVM6xly=agvFVE9h?F*gCE#o^$FVXSSC^onA4=UUM~jlBO@1H zvF)UAth2}is9Xh&+Kok1{0HjHAhy%M@sH+%*j~z`HM=akh{SuA;dZ8Bp3~UA+C=8@ z6KyFAF5PWP9g7%M-&x0ohk{a!imkZ#`!rNO7we4sQ3RX8pNvJj`ysxhI~&=6mg23J zX=>I6N99N1O6$`S?by#swY7d{Slh7P)gwRznw9;UrBn&6E|(Jei&dC2XrgSe#Cy@! z0c&H3Es&Bj?R_X;09m^`o;J3H)7PHXnug?~jPt{$;7hWd?^Lo8GHr=l-5IfM|CHyb z-T^rK%s(1_yCtd4?Ft_{mUza&_jZeds6d9)$&TizW6N4zb@S(mj3ukJFZiG%Tz*j_ z05BhTl#A;e(07Uk$j|`=d8NU@G!k}~RUSbCsN=B-vQpOHvE0%C4~+X7B0|l!6MohI zQAs^AMa9!Jg|LIf-Eg}j&qm2hWY*N(Q<&N^EIzykskpdE*}`#6DD zX2DMs+!u?kzTM-&dv9yx*+-{=(aB8PWP{Cyme}+sX%4pW|a*rj(s|uzcL9wlV@yP z-(x*%AhUZFJOpwdMWvl-EZ%NNd~Gomtz2`vuZ4SYx;OqQy0V$EE|@mz0i(|_JIJpy zNXX|bm~F>v-Wz9}x&H7=F~>0lRDia2&TrrrS7Kg3Z4jkY3FA$k#X~&bd=8 ziE3jwV3KDVh3-(9E*^#A1KsKG6vA?q#hA=Fn*z4Bp!gO#jyACFWm#WsIMMR`h&{R* zZNo^j4FH=%`z$MH67Fc!ogG|NV}_NKF~a7izt`fNXK%#4V6L(V*HSoVmER2s86Had}k zPw@1;Y$7-R<8$*yq7`Fu@e)9QO>n~67~Oiwwvap`>;Ng6eT4cs$J%38-ykvLzTaJ* z%t80Okvtrp2xNpSn0;GBlZ|hQPNuTpK$;Q27Rs?+yM@RL+e-ZItbY7Y=8d_rYOrlm z{;lh!(y_$ldP+bs$fyXv1s*mnszNwRo6@E95*`<-%@d7yhejQ_6S2HXKCP9l!E$Tz zd0Y!_cUTXo>qFl>IErbWrDs9}0z(ck-e9iA{>@s^YkL{= z81X$nahEgzZ0EEhPg=G?n%&CSH%zJy4=`UqjC(}M8QuCcMiP?o?(B&C;KgcQco4>} zUexeQB=VZdk3p2W=UGjd)ff9A>yx`YThsPD6= za3|SaAJ!!x{O;6+Enan1A{t!bnM|_MAH2+I!>mL6`q)-p7qmI$xwYAo(IA3i7&pti z=ooEMA)`$4ong(%IDidH21GQ1M0vM3_gZ@M!K;>wWdWbcuQ+?;mCvmvbi)z%C|-jh z`%X|ecm;P`AHRs%aoiw8l-UgOv4!vbds3o4x`TF>YqNn*76w@7aUwr`f`nqtZ9&@^ z^p{WcXfNNzL8$Cn*$2kK!~M6JD1kB>_LW$*gF8CAxxU&QF35IMv66ViFS&Z-a?Om2 z(`yjbxlgkhf8O-WuMX{ACSz{>b!)WUEb5Rgljs&^*&QfuPG&U$qJ+T3Pz}Eq#Mb6d zL8Hno;1C&eRUn=>6|fQg3TBZz3d^yt9nRlHaEDz;lA#BeEQH&(#YufQDF>>QG!sQgKc*xo zf>AMSS9*bT{FAsAqs0q<#k`V=ITtKCbDiZmhgpc-*4ivoH&ISALo4>0cfqaqr=`6Y z?p95HP`cJXe*l;RWmQ$;G7=weJ(wFaC|#!%769fPPu4x%3P8Iu;b{!#!Jq65d!CCT zT9$+JWlWHaZB_2hjH!ftH6xJ^D=dF`A>h3pK*UZCScl4FVK$GvB=yf~z#->E^pc9q zlmmnFB5^uhZ1iQC{?u9qK3)YEzg80d-V-1Js&1M%??j4<9gaz9U|*Ipi@kV%rXaWh zeMP;$0B?5!g3@o)Pl|CvF~Jmf4ijnMD-`L&21**0*pu%AKT!E~~ z8NudOX_#RWBXhV}=#x@;_A?7uUA!O@IWDg??&ko9E#r>EXZQ{waLaGneL|cXk9fC& z(ZQBxeIWa?IE=I+n&4v2=oCXiAOZQ>0jtocR$l- zPF>!xFTMG9WlJbsa6PmhI99-x*M?UqpoE|mWw!Ls%G`D62pAr5k-8&L+ZG@5DBdb@ zCyB$9zLBMqqSAOpc5};7dB^l^LnFSG!M2ONAwy)Rxqxpm)HYCNV&l zgTFLgA$JdacKWjQm!z%~b=(mqLw7Zr*{P}Xzgn2uG9zE4FArVZHorZOKfklC_UD*T zfzrdstwJrYm1$|uQ?ky+vV!O;P4Jvixt{wiT>a9(K&;f#`xx?GVbI9tA%&mdjHnlEh^VCqmx}3Vk>`WM)l8 zwOGVXm^nU(Wkl#nr?q8%F*yAX39G=FT^84Sd?5RtgQ)b&U1T$sq~sN{Ldi}tA19Sm z?(^Zjz0sU1wXpCLn|E&uE%rzy#^_%lCk4FL29f2$rSrmKC8&&}*%(Yy*OjTd<8188rlNP`ZZ$8gG=g5{f zOdZ4+q6ln!jgPqlYg}Hmp-x=|Ic%;O=1#@-iy;u6w!w{dLjB@3|6&TpX(%T@Rnns+ zZB^&ocjeR(*k+BDpgxY3wbUnPbDbKh*f?^cV#o(=@eLTGDJ~W9F>q_u;_W#<$mHUS z)gM@VVnst9ITh%cTMvPm=Bt_O-0{UmFF)wCp>Y6jB}0v-xKZK1Iab)gUL*qpgiu)c zs0MuQenpMOd!4Qd(x~?toJ0N@TmR(BXO7~h%n|S$2$qB3PDGQby@2~u;QIYt)<=hT-xB4yWdML(XrjI+2L|zz`qJ@7bCrbioY%=_Z zD#I5eCC;FnmY0aq31W^>IP-Kr6WBE(n&4xM<$FELWF3DxWaW&U>&LFiESL&2X;G59 z1zJk=fp>u-lHm{=O!E7<5kHUSHqD3E4v-66*A;s37sn|fuQ46#GP7anp|K3g{CBAn zi+y+%lq*pmw&vM0Rl&5{R|$%5Pmr%s=R+>@1t+Wk6ro1jGj`iQnC}f$)|e$+ahvPj zIB6>tEsxXN4tyaE+(zN#B~aZu*zaemwErVeGdK5iccslhse5hf^4PtOE=)&GWw_JJ zUs4Q;565#Kc5aYEJSA~VY&rOtFSYIEg!@sBF4){9GYpCZr+97FrN=vrj>eZew1c* zNn-M0VE9%{P?$zyD#6JPg>!nYY#U7_8VFe`PR{;|DQ$ZC`)>WIZ=3Mb2NTe#$#%W5dQ;y*0s^Y6&#vJhdb}k)?&T zYk#ZRr0ej6{e4P6Vfp>X55G}iL zk@;GzMoUZwXuakU>$H;Y7T>hiBo0$QxXLG)>)BaR^&4Me0v5t7C|0#h?t6I|6ta5Il z1Pa`zhp2b^+<8!8pX}%Nh|V{oc8Dgk0#GLh)o43`k_*X)?+NeL1@b#gb@Vci_ieu5 zkik+4*jWpzx44M%Qb~#&Q5W2}wI8zykf9P6bNN=U=J0KF8iBxq>r7yEW4giSaJZ5450f>3c*iXWdN8TL9KBnw)g2pbN~Kvl_&r*QM{ zN${?c4DFVlWTgR)rn$8YG!JsBb$JTGmT0%p`?SBXbv*b()5_)On-PT1VYGs9Z@rwg zZ*>tThi9Yqd_CM}RRF+Lz!*}!99G+k>0-}!MJnI1EAR{kTTm&ZW7lTqK3?+AaiQ32 z#a_`>2-X9uWxKyq;=j}LZN<8G8IaY)#a$xpo3tTCwgUVcS&kxwj%A=}M?QUNt`J!D-efU9+}YDqx9HpuI8ZdJQ0r z^=6yibaOYh)v__aekhkD(mMys>(7ss$L}=hU0MYYIS=j8p$+dn?)=~SPSq#7#r151 z>0WW6+H`Qjm=lSy;2|_yUP2;NkoND$Ci0Q1w}H-mPUyJN!|rdz=#siDumCcaEeNO! zq@9c(2q08u0ALmNX){Ueu1wBj%KChFfu%gcq+5^^?O1z%6)5s-$c?C#jAHd^KoT*P z6`N=BWSQsyUbN^-4}dmW=MugS?h=!wd5aVb?LU!XpZd-uF|JN_IJfyHZz<`gj7Yjx zNnK@lTL4z{rP_i$za;>TcAO9=4%R%HJ4x&_yY(0E9o#%jMVO@`)W6B@PVEvpVdX;Q zf$Ig_IUl)fKo?TO?2_3Pzm6Po@jeZ#ACXoP8dd=5RcpH3RqSBEeBMH>6Qin`R|w4E zmB+Ud>6VWqs*(ORgI9dM4X-Y)!WaD&J5A8k^mUxM@2;}+B-MU zW+ee?bVoG;4tBm0N^G$%;Om@w40gBKCGs<(YG(``Bw_%_^z79q0~a%I2_W5kw`0s# z8~KR^lGZjTnu{Jx!OXN^XS6U2-!oF>F+9tNx(;pM6+F*={ifdSIF^BTrOaOPiJ{PD zt~L_jeXrA)MYjA7O$Bo10zEpGCo_%k;8>Waq;gYc*biGkUgRha^vAs(8acK{$L949 zEjl&c1*eotH%=S(Q_h|u0GIT$W^E(K)|c-s7tpBoFTpQEl{M&9ZfU0r9%YWc*g5he zhk0QAsiFv3)fa_PIF7RE4_2D$fa}xPe244<92++8?C7Ra?f!J=D@pxYX1+EJHL2YK zKT=LvIuGzRNfz>*&Q6$H`Cq+kv35ydNINx=3kP#ivlUFfJl|c;ZtF_?c$t~)%(6#; zRj>~>dM&|z8X(TOP)wB;6@`P7zhtg1D$rG)Ec803VC(CyIg4UJ8VBoO(r&%%kV#kLHc z0SsvuYArFCPQJMjXnaGG)r{^52UNxBsv!qTf@uCZh8b_Dz=RZMXpERiOWKIo-5T$L z5sy|0kEIcXyryYcfY~n|WFv%V^Xn8pb%RX8J)*o{{Imf#pZ_R#o z34(bbg9s-rW=sD#xrYFw(So{k*ACA=}rO1!wyuEH?_c9#?@*Gfqe*O0uU1jE##; z8BMt@-6>b>@Pw`iI%W-bks=Y!D~CQKeSHZ+Dev62tXk~1oT>^DoErNU+$Nl6Gw0|S z>|OC$g?c{$?pD}MZFC$NWHoj4NiIXiaD~%31&Ld-E@Wo6c#}6XptZ3Xg97@GyTJPcau}#tI;Y%Z9c$V7RP?+T zhMaGi@~m!`dRNJ&1>ZV>B&94227J9bzkp*qGjj1oUp@OSy)sA6k>X2w5;N=9*@ z206CdH^>PO4>IAfNq*u79|?;uP+4bSSi=Qkr(YJ7FTD|e13UNK5rsokk zp5O42i*fJ>aKhM>7q=&RMHZ-&x(lX2#dV7ObAQox+5;we*H)luZ|1_^`{yK%zwK2QaK`QkH8V~qRg!#cYWWX?tW$3d z-<|ns-Z{I`@SfneZPV$>9Pj~G=9lRIWRjbzc6u~!hc=|mCIJFQ3ZX8brNd=ZYUcOy+=Sh z`Xs-*VEcTve>X>XW$>KJ$YBe2(q6iJnbGshE0mZ! z@-wl*1$qc;>uefREBUjp-1HID>{989r95Zzja z@3u^%zY!H}Uu>;TM+0U}Io5!d#s)Q%HUuy!({*6deACsc%+2yH19vtS1_An;}Zf#c>~~Ik%0EA?_e`Zar$N#$JtGpd1k?lrD3UXcK+DwFAlZD3Kia#TfxJd^@wRSkr5fx6hHxC8x;&<9W)# zEu@YLObXf&uO{R51lBFYFyt6a`SgK)K=H0m2ae{kKJfGy(sts_#EyH?dndvg*k!>_ zX^e*AIGMlK%sKxkmO9`mgRNQe$u{#$HQ~YLlV3!}h#9FU# zv($o#rjPscVRIx+UCL(3WO~?YiUH3sI4&8doDxMa5wpMW(ppSu5g-Dzsq9ipM!q-Q zQEI{fKGoHS?{c*6UfS>Y=kYAZ=g~b2|McP5%hLY*S>)@3EYCY`xSy!m6PlTusWMXu z_crqZmm)wz0Y4FqIS^!OpG9&!WFim<}MB?I+FnO|2CSo zBT`ypo#Jauz*~u@EhHFbC0e*hH`Oe}z$dvSt+x#f&I@?)>?M)f5|9)5T^ zUt|%-2cBJ#^L1 zxinp9k2fK3^Ou)HWU8EWOT$O5aba;ci{%JR3V`0^oUhS+3M;An51)m{?ike{~kUb8FH|Hr(6w39XmEy zN`BUAGu;k1v$#^Do*<9XU02ur-`L50J_!&d)-%g~_wWu1eJcG%%wI9T-#s|iBeNtW zEZpGub!#Wyx@2d;-3BH;og)eKcKxlh7*63l^G&sjYlP!Mu%Y`rW# zLC#f7JMJRumppIY8fGNhncp#c(EJmv-Zj11^?ixV{>DF?LI-c$!-s@*yLZmHc$C3r zFX01SwPps*H;(T;k?OvXBfI^6F&pna)3qrpNNISt0U+yxwFy9` zeEvFPq+_op^LOIKpR*ls`I<*4!oHoRrrT@#>VO5a8?JC&1&A%Jzy0a1{uh5b0S`$< z3y(amXYv1Ysr~_T0?#WA1)dj1q&;xoGxRXO0Zm+c@#O^fmnY+)z~`Q%C49^(y2N{s{~RUuf8HSf?OAUh20pXrI!tBn6@>^e z9t|6xG`FbR%zv?%`BPJ=Dqy*b zl^5WH*!NzNbed}5_0aD>gkE8e!sCqY-i^O;>!tP$UtbGvZQFl5%71#ifBx)QIIxfS z&L98%Gnl%7o_%O{7#`8Hp;eQre*g8^L#7xj-08#ToyCH-rVWekg`Aha>M`}|tkMo*syW%RusYW+ z3T*!R-#Lk@ArRa}GV1>{1Na}?;LpF_HXI{(aJN&O}Xa;MMT}!*VBXM8I_cq*>gv!>rFa;@`3T+ARWycioa1@y6`GW73N76F?8e zdibzRUeX$lVEB84#P1!x_f&ts{AaZX0Lq)0t+nY12<0=dHp6MJcPai!S4FElXQPr7 zJpId}twM07?pc*&$8K`R2VRh*-`dnitE%nc2Dmpv&aoyH;Ldt=5xf_O^LC%f1aJP~ zw?pTXV*^#>NNvr*yf-x$OaE&O2J+cU2TTRFgDoo}BS2eiSANb3ax@ox4pUChmgEeg zy?nnq@A%eki$mP5CsV?K`jU;HjjEQ{9;Plu12s-{S=G}?VZU^B4Wo@kGi|#K=e6m# zQs|h1;flnP?u^1H#{jSBeu?F$gBXjeUUNk{@XJOxUm|bbZLeUUr2dQJT4(>khGFrj z`1!%Si0ze0eZ#B`U1l z{97U9ii*~M=ZGcMKyWdq@y=}So$ipgz=m5h_%oXC{qpy@X4@~$8PqcMdKKtNNC#l9 z^|T`u5KRS@avt;2fEN_d?ru%?p<^@9o}X{MC#=qdR!N%Uusc7q%G^9_bNVyE;^v#< z8W`P$t@@L!pOgFzWLIt0LUmu?qs|a4Y)ri@R;6I@1W8Nnd@3p*qSQ+A?a#)7~@?aZL=2_ zL-KdsrcHx<6y)~e!q;uH0#f6(Y(_k6=W?yXJWs&9DEy&rr85if}`Z!ZOlon98n@ZZasQdfk>~Q z+{8fYsE>o+$bdn&s(mOxUN+#Fu8q1ZrFUsM!pGN`iG$!i6HS@@{yN&Qz=7AueqoNw zqjrdScKrv`w$;eG29|2u7j#_0!Z>s=d|=?N0{=-7W&e$jh}tdFvc@URN!=Qt2nk|b zR3g9q9We!DcCQE+UU_4$^UKMi3iP}76=m-gV}G2{-%)a`a#u$4W7H?-2oWk>Bk#Py z4^vpQuaN-c?o#?N$>=N{8B*Lu%3U$z3-0b|jt;eQ`nMu@99wqJ_Q;)_R2X2M33@_@ zNIUdCF><4oE>Nde?>1uihG``pvDY@A1Q@9V^m`q<#dM$48bCvGZ?7B$1++vJ*OHOG{2X(4`$yQriqXi*`D7ow z{EvFuQx_x30-7wjp(R(rZSq*>3mOr8o~-@!iB`mPPT}Zz0E{Sc=!-bY&23+y-3)~C z9{naB;%t(_ETO`pwo~Y$$lv(@oTlU^m8UQ!AAHWsM}pod)xw#^nK)xcwc@G=YncE{T#H? zBdHlHmQnsT$R)qy^mBw&dG%S)Q|arLj*JBsa?v>D8l}89cQyt)$5%<|IffNFqOi-Z zB{uE{acAV61Q;j~myMqeYzN16Z?W%P&HI57F8W{P;o84ai;(BQ7#NlNqwa#3Ge!*CWqm6@a{WX&KAbdl(-J)ei zVtAc*1k^K>W9AjMQZQ5oEhqMC106NHh3s+4-4DSd)LDMb)q#gmD(e0a`G7`$$63Gx zBi5-8-<`LyG$f{f?W5J|lvFxg%IcEsp_7NCYy3)?LRW+ZnSbrx$Y;Pb8IiHXj*Pvn zPzOw;ffp^woBhrk|5Jvssr}WO)286t0GBmyUh)`uaoByB%O+OdsoQE-G}xXX7lnrS z=cTeB{`#Ta4A?E&cV^KtzSRJX*Ey!~+D)g_3MJa9W~hASMaz;4bym3J#(6eIAI>TW zt1|_QS%22!4oLF$c}PF#M70J6DhPcV4?9xx|{b#umZe($Rk0fRzB z@~Q28|3q_t2id*!WasFsi!8x33Y%ZKhph5tWZS^9jnxk7+ zg$@g8$F(^mE5_3o56t(~AZ#k^x+)DnsRq%mR;z8kKO?VM8n- zZ5o+(%fI`06@4rOX;AQHswK83`^`Fp^3Gg(tmoVA?$V_78kN`$a&(t#i#*0xye~k= zJ6{T287xJe&oh&#eBa8tcm5*p?0%aVrFQVIZ=)It%-eyFfe!5ZPwVmfXDVF4*Fy-8 zRzSiJ%DcH?mQOvWldwjs4h7Z!hrRa>r~3c@$6qNWDoT{CR3al}WjE}I>|IuN=CMzT zR6=AVt6|TwImZY&wyY!LIApKm7{}pQ-^VF>)ob*Af4;x#`d#1eb$$QolJj()k8$7Q zc6VA|u~Qqys2*4bCe?${D7Om7C0fwA6*#C(inMHbc^cq^BHP|E_-~`pP^EIh;jzzc zfjI1D+R>|^BjM{JXtR8(*TlZ}KogNJMcr)s&rfwG3-mDee*6>^owRzv+^$8)PA7Yg zJh5GCyoG-M5oXR?DJngP^*bAP({EP5&nQK86^+2{_*HgIBfmeukSvyqaxI;^SOE8h zBb9zQx^ZAy3?$_WG3dB^eSiPyelbbHzTs|0B(}(u%BXw3cl`zYmXr?DWuNr>8<0Mp zMUGo3muA4WwD%g6+*2FOSPB6H4?Z%sEtq_HE~Ad0>kIi5mQZ}RTLs{=ye_^q>{B=c z&)0V8F+trJJPDEBmh2EIdE?k9wz~={EEveMcZ=59Z+9`$FTuDh$%DK9{Tt1F`=Ejf z-1G<{Bz)RI!&&rGyclG(rFbxEgk9xnJ}Z2>#JR^_7mKX@`GU>YOOyyiI+x7~e(>mKIew2dq?fj+f?h_+P){#B7I1q3OqZ6oN!i&4RU42Eh$V*l#a(l7k1F_$7w2`c-|6^2{Jzk-uVUY9tn|+>URz{8 z)6%KBZZ2vKSpnra`W>mN?7)6ovTp5l9&bQwO12~J(R_vo$+#fz;W37#(CrHGg<9(H z^J=FpYyR|d_9cO--1p(x2FHKgg4%sn?`zaQrULC+*?+OYA6xk`2;0|RFz1|xeJ;K` zs6kl99PPYyk$Lxf?%zKrO_dP3D^^&aOA{IIIYW#2Z^Z~%?bE8d8?ojn!r?rpZtGH~ zv5bmGO#MQN1ed8h<`T+3EWUkZVCB-4H3mN>cwiHW)I$DyY9U)yF`2}M@!$Sx=JS9k zk^FfP3RCL2tM*r9e~kAzRsz0z-`1X?-5X*Z1_K&ZMewgK{02N3oD{aQ>P;dv%#>Y4 ze!ty;RRd9H1rV~{s3f!a0Sk8QIq|z)@=G;#E?fQ!AbaYS&57wBk@W%iRSA^Ya$XTfJ-c z=m*Dv1R56X;=q=~IQh4WB>_Wbz5hTd^Zb75AnQk>wTxxi^T(f?hZ85_kun>8ClG8g z5D3s9fgjTGUle@>HTWwr{Xb`u75{6yzJKKn1u0b2=ZcRN<#2dKioZ1P_=cmD)9yYi}c7k-!`|HhI3_`UbH ziE;%UNB3V#+F!TyDK`{w zJf$HhAlofrOWA+s^8dJ-|I1kZ`At%lZMRczj4t>5Q1I83Seq!O-gbwov;sm9gS`1E zfBm0S%m83J$iu5YQp`iw03%Qz=Y9IK#`rIY;MYg~pCJ5yg7AOF#^3q-|0qHDIEciO zH!+f1uDH=p;FQz!qSe|A(%8{4hOZ(eHNjgkPx8fea*HG3{Gv)qDVF@7Rv9Z8vCYW> zc~j9XdD~rk)^OFiPIGS6B@Pvr%x!-so>+zUsk1{Bjg;|kk2mVwY#igb5YuT95^R2P zH>gfL%soVBXUKu?#5X|;!a}V^lI;0KZBd5&? zC|Q~l>ge(?Lxe9zC4KEiNB5FH`pK_il?Ac|ai0C* zcZZEijO4;fZiLzeL~ZM+qc^g(`*OofK*pyAO`d+3!6RMU_%9-0(dVcJfS5#qd~ zQI@wP8Q6Q8-~K8o{(39}6b5idHQ%+l_qI}bSGAZ{`LY~OhuI63$p@eAMnO`339X8Q z`N;EpG>5qm@1{4 zE;@`CaShb@wF~1fDrTtsKKUnABeCv7wZ-0r0hf4A%x+da`{ki4R0)(j;erL7`|(2( zl^%{(mhL|6tI2&T*uKn>8K%*>V}%lxRC>R+>1xIN=G}P09WT_)%wSf|iPa!r4D}EG zROerb-j#@lzRph2Ba9-h+*{=p6ve;zsNf=LN>Kt=OhUum1Vr6OaOcVG6F06gAA1mb z7&`vb4gTGrxuaBdsD}Mm3~uXCL!8hpTkZglKi^43Mg}zN9!G{!GU6rq`Jcadcgiwv z_;a1V`&4_yIQd#*lbplKQ@8DxlOiS`QvsEErt7_bx}1X&IQPix&|h!%mwWefKYh`? zol4hrNX|nLdVNY8KjNNe9rbg)zyL8IC_GP*`Q00?%C72dvMfXm>6hAFO?LK6&t|uR zRU=w4^p-sFVX`B!(-iujaSRI4i@Fm*Nx#>CtgNIwgC*x3Leo->`?HR878(u<92j+V zQv1n1TfGVtp;Sc7_hJmcle91M?Q)d+FR?w6yK%B+m$ojP5OeP2ih;!+8-?Ww)e*RH<48TEi4 z{_*qO<>iW|z)o1bY9;1{>NGxj(WB8rik0p6f)(8PAOEc9G`C{BXxQU0 z+8faZFnrJb7@6_Q1@G?cF$2j72ZM-Wxvm#o4T<4rC1?@B*FTW|{Q5WG_3rMSIy(~@ zBlm0X`AFecSPor&3t5Im&FRa`!`EIMZfEgN`00NBVb=>r_o2d1@x{+6{CdZ`VHviZ$ot@E@m2e3$0d9lInGWI1@4ZmYI8xD_j8l7}p`fA;s zbaj?gS0Ren?(;Bx+ zInrIEv^<6!L4{Nx7m^d#o2-gI(wcF8=>@n@DH(yC;pTiCLqPyFM0w3z154%{bIS6H zd0kY@G0&CPX=SgQWs2gzinU=fdSaNqw(kM#?vk&=`8#@|x(~Lo7n|awu(J%Rdn6dJ zi%-5?&aF!34d3IYGM28~M-ws6pQ^8H z5Q0VkQvu?1?zj4o>OJs=icRkQxQtr0JvLMoS}oS<6D`&R#REqecma-5PiQZVq?-dX zi1^|x-M-WJ+p}Q*V16>jx)?H(v#@XYo$t&lMN3+2`;ThAKZVktKapMAvxPcOHP6|z zJD1e&oXKIh{zGhFXu@-un!2fIB`Qg``OANQmus{OWW=kFNAJXVstVEu+ZjQ)v=NXZ zRT%T5EbGtr{Q1*k7|8}qZJ=B@>(GvjJ{EPKyl{IFjlp<0Sws(r=0D-oFEL9loq-*T zkvPk>YCpNJc=9geQ&V{(&6B3|u=?0w(Z_0lO86QZOuirfs$7-fO^LD#CUUEjjUI!c zm7kXfo<{9b&_D3_eNcywotO3VhLrBs6Fuf7p@`0j_@p-^krypEXrNWKS+$J z4L(2mUAtgm_j4a&0d*xe#n~Ut$yAFm0jwKlB>A1A-28d%Ur9S<}lJaSW4Gk7w%&~c;G7`yS9h@Q9o4H&$hf<{ zK{BCHhVI^jhm8!u{~?g*|4@EuOtDX$YHJwaY>&@6IL#3bm%7dIdhlHJREQeodvUQv z!0ODW7Y!}WP?n7?gvZpvf{AmVsubq5ZoZjN+l~SPrXsrus-ZlJ)-89*LNJl_5TlS$ z2%|(!=vjXhQ;b zCav_$`A|thTHA!H;75p(Ry|%xd#c+iL515ikn&!4gwxJDuwnOIQ&;LqrjbPV%-E%1pe~( za!bQcngC%#Fb!*dDVP7av3PU64aZPCR-2nTI=_wLb1?5PQy2|Wdw52-BgrC0>WBub zB&glkdf7O~xNjJ>c=NlZf8TgtmRMotLBoLIY%v5(?`38Mk_tX8=B4Cg09_d+oN<1FTxrTaMr9lKaxnDE2u~RX8J%>LTY8 ztyYH6Q1cadK%T`WZGM(BLP#$b%&v-uLB=rgFoGjkIPN~2JbT-d@NA$2C)Kc~uI2Id z38JN9dI*E^a#|UfcUqg)RLCHm?o9E}!_Rh7s6gP`6N%7K==QpP#rCkAZaz=*mF%(n zEs5nf%uw-|GYa9gm$Q7mrxt1KX3>Q+%BN%aC3({eB;h;!Q027;@f2|!nZxOlHI952sHw-5U-Gno)jiklFv6*o&C#@~G zuT0Kh#ku0`inDkOx)a8(5O)pCQcn_(7cVrEk5u7LX*k21?4{-;5q+y8RgtClc-L=j z)CE1(Q-G&~)nyin;-$j-g&k(_*^8Ox$&L?;aU%=VsSRMLJZ|Leb#g@vBX2|8TF`aU z&hE^s{TAw=9dT2Kbkwq0mla^I2+!ci=O}479R^D(j&nqy&v&46jVn)qYD>PRM~64$ z6Plj4ifbPL5L}U>6cs%rYqz(f%rzwA2^?1a8DX@xP4>ro2Ps)3c$O;lIYCK8e)dGW z9CMgca*)ysS==?I%$~32QjOK~(D%6F1dVJ1qX#qHLP1fvdh8UmJo&+yoZTS=yH~>D zKt#vb{g7{y2SxTHvtdR05ElroS@DYN0F7kHbvb95R>G@DMt5#tpW9s7|6S*TyvT>6 z{@U82J2u}ougD|IGh{3^ulw!+9nx4d(r=E>^%@;Wb~d{4oX<{{W95z-N8~0A$Mc4* z6Ol%Tl`VfOJwnLN6tKeYu!y+CF^alfsobk8P|2FOGMB>wj0jG5UtFKDXX!$qAeq$d zM&;fzpIo)E<#M?Rw{R&3o_wA>){Q7TnAz7ZQ424`UK%qNtePb#%6O4IJ(}LR*f7L* z^uaVzme-=;e0d?$oflRxq}*9DDNJU1p4a@InVUyLW_q$Yx-%>2-^7vZEc5ZJ`+^1b zhpSAUT^w_{qQXsV~jrY*u(#J&-)pZ!8^nbx7l4BdoG0e=)$H zpQJ;h2xQf;&ILGT1=XKti>6#0dDCqah&F7=xU!(Rs2N2cfPOxz@B(HnEBZSLK$W}i z*c~7M=2b_(1~d_TIF}UFtUBL@`Pv??W)N1J^W0KLauEwnJYeH9;S%~TRddmxq|QN<}SFl-m^6I?j#_mP5^2 z2;5F%(}Lxg4`-g3kG}R@T<%O?UJ8olQ3|biXy+Vt-blV9S&7{X&-6Usa{D6UP>!*F zMunD<#=Whok2S}RIm!EWyFJ(D6|~1$=T0F!oX4IU^x3ZpI}C;w#e2?}Plr00KUMnV zlw43~5IuH5E7m8nJ!zq4G;G@~PFU>TZhT7Z*gYCOh^9T$fF^_X7_x#9@eKG7FCEQ} zf4`;;PO4ZLcN!gF9>;3;^rtDmr+q-V9>i_Acg=Ic;{mp{OM>^-p4Bz6ipT)iiH7p^ zmE6E(?})TQ<#3TFT^w5a)6w36(%xN@cYo*iI^JY`LpqGdP(<<>D+|rndT~EYc4@dn zC^e<3xYHZnk>%(kvCqNn^O0ADbGG~2G);VNqM0pXok7`WIpGLjtTq?^>7|~!o4gR2 z)!U}*@^>eBJ;ema*;!m))IMS9ZgUtc<2GDB6hRa)6E_~BqZDZxrx8usIgx8wr4yJM?~m=!TW& z>^KS*Vy-50u@@}e0HE+gU5CTfVly^Eqg6J*WHzGa-?4_bPCRY_JfP?BU^*;d=s~kAuLffd~=<08bSJSQK2U!+itTfF0+tnta~O* zjwP&5Fyqym&H4veDHh3diNR#(4u6L7~ zAIWQJ_i)oL=Sc}TSS@F`jL)oWN{ckThRQF&k~bPiU)G*E(6Y*U&F0!4p&b!#VUEa*`K)m*iH=JsQtp?)tEq`VFydQn^JK_ z5#|dw^|>~opl3ok53t|)xw?8$HWd+mM(79*55)w`!E;oyV8zAjST@k*k}{aG0zd;~ zAe*%hrA|{+qhS!$KHo3;n86QFfqXsv5L;e_hC2bt{wruRg%_R|8E)(v29ft&d)yFA zs9;H?;czfN0%9XQs5kKi{|>`+7T5^LbBu`lKpLEsHv1OZ-Lr3ln~FikT`)BtHK2Ks zP}NHn&pN97xPPduuriZsv;I*{p|rQ(b{$0|k8{~45nh?qw6L6Yk8exqQ=g2EIB|_L?l>0WRPQlg zFr51w4w>57M9|HL;R+uOv)R}@Yw+p-a zcbYtQviaaVyick_0#AN%>yj@kO9+(#!O%*tD4o*MWVZ(DPp|u@pI9wONkqxE??gRQ z&yqhyJsWdLRS9Fzdp45RAF{G|hOE-LO>)h<+9e-Go|I`)hAXt1QHi-^mt&MPodT86 zJ9bual%EVbgS7LQe`qvLb$Gh`tZK5NRYK;&bv#RbB_*?bc#;Cq>}(!dI{?O7oWcpC z)41lla_na&FtsmO;qBYYQ~7yzJ*q^ZCF(jZ{O*f4%UUlY9MfAcF1lL}J`{6b@HM7E zWsdU@bhZBACiF%3A z3xn;S0j6mmoz0VJnCnK`3NFRRd!3?1?Sfp>)6GnHsaYk?r~Ibol)>E9$`-`6e`K>= zQ6NmRFozBvFq{FkvdQM_SO*Wxz=y))M;a$`g0fq|p8Dtv1KNUnQduBjS-SiTsf6_i zx+SxkaVCNzP|giQz7jZ7ZL#XiMxl7mhPl{coLl^$!`bWPV?Cj4>tJ}6foPsWA~Tbn zVU7_D^xiu}1yZsob|fhzYXx&!lnavOKENl2(%(gvij_FybFz$J3KfP01GdOBt9BmQ zS9F)+cEX%gSmi4EoA@!6Iiv^3pu0J|?R!d$H{bo;9-XidyuX=-$m&$-}3EM0Q_$(w4M8^E&ba2PtDkF%j^l7I2Y z73T~D`|!v)Mfv(RPuL~Z)TlR!D^X?*;jc!x)@~Qt_7&ZdYRjD4=zH4$BP>k2#+r=} z1X#fE{g)z-eIk#YxGL+z5$G%VHc|iTT87C&9*g-fUh1V?C8L16dI}ku+n5wZx#_Kw z3+1!1u%h1x>D3~gi7nc?$-J=+q8M;TiF6$R?%TFock6!c-j2NQ)d}dz7OTi4*Gw^T z9X#aKqZ+sIrcRF-o6?6{f3=18-m0E(=09}3otP@RAyW2TDN@CfUtJ4XXJ& z|K_YgT6kd$9W4G~2@!&|X4)|}1fwIwuRC$XNrf6zy6XE%y3`&Dlp@r#@w-c_dTf<0 zd+LKjeuscYss=veLuiN>tgd|=9YT8qdGm4iFh$ZD8i&p!{rmXsx*cV-dcCf)5y~7cz77*&l8BX~w51mo0 z%w7AUeYIaj+0j#KH6Am)HPje+VG)ryF~&E?8=QCppCulDxv(N`oszj8y}h1TS|C?Q zU}uI70B3G#9D2cdq9r;t!BbJM(3Z!%$$vAsIQuAjJ8g;M$Rp)gNhMnDEu*Kj+??}; zR;PN6JURMCJe*YDNtC<@0eZC{GUfg3SJTX)+VuOC^U$WQS%*S3D769UU-%WzC>02elan3U5Ye_ zb7`wwz_S3Mv`|7#p@OQXo6t1CK=qNM7<>0DMFK+#a`P~)ZrS_@834KA_e^MJ5cp6QJ2k~bbBp`I&ax@5q_xM~%h)ZzAc zm_~pk5h^Mp8$xe$#G+iv;`Z?)r(vmTCpkM(h@A!Gkr{n`c75;l_|2Q|UKpt@ye~B$ zJ)^nL;&9O~1-!c!Gu7itl07v3m0sF$=0dGyhd5i zUG8=^+o3--`9_RoBI{9Nc-y&z9ye15B$lZmQXn>4JdYJ_4PQ^CiV=W<24&9{r1sop z<6JFQ4&#(0@wWl9OvaY2Tj!K~b-1NiB<%Z{jBsArSLldtU5MfG(u-7;rdf}m(H?+B z7eAY+op}a__Pv!0#m-menY26U=HGVqpHW_Nv9`*XQ(>VoO$~az+vD%Dhvu$!uDEFa z%zWwg`u4F98GP@px+z>8%SZ0Z^nrcb+x0xI)>_ovcx*fMV$jO=;Vz|Mt7n!%<9{if z7-I2TW%Sp+0>Yg>bPu$bsQtCcS9<2Pq0#54U-eUrto1&_r!+39O3mkI>flD`r#g!( zMSZv69%da$2g6s}g7y;OfZ2a}AM0^v!-2;(Y zA(;USo@0ND*M1F?W+fdqeYi;sCW9DOsqaY2ii1FCy?|AR{L|yAmkpni$Kx498nYNA zC)wua<2qv|kPkC^a+(S&a~jzCTIbs&W3|Gi7W@nQ<`%j$_0Tgtqicug&gkYP+4t*R z=srU1m6HxXqi}~?D0F_H^a3d5;|z={|Jdeq!Eci=q$;l1Fsb z#g9fc$tpk=YfG0Z{~D-4X7Tgi30{1yAf^|H((ar&z0%Ebl5>y9Qv>4;h5-+vh2P;w z7o$wf!z7awR-c6D0&fWSfF^nNLDV2S6f~g30NdCKFFeb9Eh{+j?DUctNJ1rY`wqW7 z+fnGH+P`Xe8Ol0e?A4|CNZW(Ap|dQHft`iMvOrRSOq! zyz{5+BV84FUy#ha6!sge$3t(@ap(a*u+j%6$S=^4~ z0oHBox)*Izw9UCM6njUrP&V@lb8id1odjD!toLmI#kkV&6&d@9Gi38o>~$BwrH70* z7}pEg_wpwvxn|O;!*$f!k0HlZw!%oSvxvL1*2slKftI2fu_l+<&qjmU+C4&U$_Wvf z*jI59C$^%OB+6W)@O>iaUdeM4!7U$JjZ0XsJ~|YAw#k>OP#+qa!=lLNn#h%F4yM2( zR1CS!o=#N}fqp&(&Wxe9i5mFuB-W@+k4m_)-3I-h{#DmXo!x=MfBHWADuDP#qd_8e zsQa0R)5P;ZLo|MKw0YMN+H-2x$mzE^1vZA5hy44wQ1zI>@DENXJtfsB+ zSVe518nf}>!RkPCdOyTLsfL#x1P6+1DE3ZHHkdV-tE}sWi(4(8tmM>|SyiSpe$Eob z9mSsLvW1JaJ_QoO8*U{`_n`7=T{*YfkQVLL4oNp}t`-}ej7->neR~scT;9YcxkYIA zKj|-w`Q$VA=3#B9q;*^$cwxT?$#iXJ;>Y`Ccclyu75P4}p#Pg1m9Dx9j0WZ_9_`J1 zWQm)}T#L5@wZi$Ylk>_Z@76S*xwK~{2OH9(LN=&YyFKx9t++e)&$(l}zvhlRl9bu< zE0$(+JmYn#LXQAD7oEkyklFyBTdJWT!A_#IrRVR67M zP@+%NupKB-cva}#CbZIEDixGZ9>4MoJ7J1

Kt2`Bk}`1w*Z`DiVEz>Dh&F}oN&B8@_b!q#DiM**>dHcoe zV7c6u2-Jwj2!5tb{DrKMJ4V!_RQ`6l#)k*Qbi$N@8n=dNu1=|<=QZ9OGpcq1=Au7w*=PJ&mO-hWO1vmH6yQ#zF?A(EhdmnNI9?DNT7h%$ zsc*my?f`wvDDG0v)s%+I)OiCRDEHF7Qh-81=6W;r#lyr)LBIahr^k8Dm^VcT^=Cfx z!ebgX98T2N1fG`KcG98c(z%;S>^JmC5Rh}@P=mkpW!5FbR|f>;S>x#Q=K&@n_V8f# z#3hYf`c2SMtDEl@KZ}M+S3X9SN1l<*@xo(@) z$L&t*ZmYsV5!|;FIj!;f<;Tt*OW#&Ok>R)c3TOIG9vG`{H(hLMtE?xTOQ={@*J^lU zXc;owoBXy8BWQkfrt4_3i8Sh#y6QaP!$w)q*0JTYZr+f` z(!4cY<*C|jSy9W7Ot7;G9f+jKrhufI!`0!cGutnVQqn3^ z5+%85ID4TO;Vk{Ly8M<=P0_qf87w{Z3)JkRp;E0j+e3?4@YN0_^>>7s@bRfsw)klR zj-hYjZY87CS|+1}r$K9+*o7DYi;l`VVXqZNLfc)?=69M&Zc{vMeYt6vt#4gzE+1Pm z+7PSX)h3f*P+E}K(i|4DaC7}*UxjHZ(Xi+$7th@^5MQ^|1A)5z9EJoMj$36Ba+U0H zH%6y@(~IOnnR+HPdZ0_l8x5<~6ab8ac@y$96YU44B1eG$c+4%D3yXQ9t?^3wOQgLY zoQqe`@o#OFtSk4v5;_PSQgtl>EuXWSR*7*Uj_Q>T(-P(x8m~Nf8saacT0~vtx0a%4 zHG0@%V{;D|v1=^eMuNk=yuaVu7Jxf-p0pRg5oo%#_ z+!$}RbkC0rerq$iq^mXaOGqct z(WTlkQOqE-*=i&i(oVq0pVIpP{Ct^vw?;IeQr>qfJ5AP}={ld@*K5aA_+q&aBOs7+ z1G#rW7FDuf+&KARB=}bGJPcwRR$!IL*Po?dsV2GlKC{0~X~mYe;Y4C64GuG9Gh4!I z)?CuoGOn9%5r(s=PIMFIB&;y7=8u1=H2?Z#_%qq?;FN%65JB{!QB1+bg_Og$DNl+R zL2u8=1O}j6X>ii=Ap^H5P3blb%7yJ1xi$&DN!nsCFeuCleeCnwLR(Z(%@o!3rxjcR zsVY$jme9VvdqMP1zt}$dRE=yIoEN||MIy8HL=1yCHTDp4P+q08xqWHqy)P=aI=s`a zf(bgp_5;t2Sa99UE^F<2jKd|)sYg1B>=+bgzxtiPy?#J=R!Y#-*qgB~dS{i2BO{>o z9l^=1vdGSe_hI0n7cu$cmzj8Oyh=6idi>@n^JBGOo31Qw5azI5ndvPL5>J)5*7f1m ziF{?IG_=69WxG3*sFOw+Zqj_lrb`#yJ#U2gqddbnS7NgLt){-sEj9GD_EIH08KvXN zVBn9PGL;Sf%%XU)F>8KsE5XfTCNq(ArcCrl%i>QJ5BGSedbUvYYn&WE^(L{}gKCe^ zetcl}cO!R5gIMk{!!w1c!xzMN1|Y(&%m}VJD>%;aUW&b8%`w3fzr-$+)e;Hvi!n$} z@u8E{6i)y;!1&GEeIEIi*T^C~>HS%Ryd%5?9m5e#c7p{Uya!UF5p%(htc6F1>Ta+d z_NJf9w`e(ii~Yvq_s#|6dnz*nYTFVkO)j~v=Tv7djW<7|s&@mZ7N|ZJk27IKUW+`% zBi&gGkDk#yydH@(l==8+(fE43R2Z#s6 zIqfpeZXb^RQjx>s;P<;&0D@xJ&KxqzYu}AHd~7!Afy4MqSUv+yd;x2rs~q0q27>~a zvVpTk%;2w6RV>;>nhkC^1doaMOkc zzV=lzI2!--+2c@PnU9PZnznp18V?$MtHBbPC1WuZZCn48T>1HsgxAfk*6o8+f}HOD zX%2#q)1v_}LO>bAr8M4=LKHkZmnUB8IljN#lDh*^ttEDcAs0;1Zqn{Mf# z=s$uyy`Q}<=yoI*2gu6cU$)kRmajWGje^$APEPf_LP8sBtCiS&f(i) z6i+dz!WKrv3#Q={-o^4|_(T1{1GB4WKbTjsX}_#OB2T=xghz(AkXebTcIw3X%4`?p zBf41X+H_a8gU!1E;~tV)jsiT&Yvo#rFenn~&W!3Olw2-_%qNUCQ=4Tbi26K@;Mjbs zy8;mE71Q}?TAW8->-SJ5qr9?%Z*7*#FQ^x_VMlW-JFcva;VuJ18bqh$KURN=_YGMN ziIVCovO0@`xgwPq=hKsQlLANTo<-1i8;2Iew^!*>4W=!Pq88rlp)twT7z-vd?4FSK zp{@#JL!9nPCn;=#l)1FMa=r1YFrr1jS_+YIMZyH@H2h$3eI8n$(xT5+qO$4)*GM;L zU>s>%niwnO>}D$vsJraB{Iav`27{K3SgqwH#FN>o*!5(|rT*fq=m?IQ$`93iWb@X( zY-G8<(UIphTUXXD5k{M0#j4xF)Pxr@7v5eOt7{@IEoc_iE4;6^cHYXt_8lRgs#j2b z7slvkhwrtI(rq;P<(hF-p(?+hQ;UBKuvQ$Y25vDi9dx~zqOtf3#(pwEa zPR?l(b-joUPR?|KhoI0SlLn+R_}Kx@P^pCG z^G@p(_*5VDVxz1241#rE+G~={8jdZx=b7nYmrle8S{nH?wMKIhw8)Q>d6k&dWr=n` ztEBryIPXeeU-+u8j#YcEZEYR^U?vBylZE?s!2%0x2fJoxILXlEHkKYE!5QHT0VjY7 zn@>>b1*2>(uBY!U99axxLKjEceRC=bq@LZIuDz|#FjC8RK7$0IcCdPTe!lXo!`Ay5 z!8(n3p8^~2d|6isxcAC{=$a!=k8Y#)q7Aj9%gx9Tu`K-OYSg^dDF;OKyBIxu$0u}y zbDR0a*c_m6^@{9m5;FH9+ETJRhLf&my5;#^U6raAksSSCyGB0sh2v|;cy0t2kF&)M zVYyp=)7pxGn|njmPL@H7ljOs$NV(GZ3-@I}ES)_(Rf8Sr0fN!T!wm98(MAzLCwLpM zVWy{Nj}K;;3rwN%VeMx-AP^aM+>(BdEu9+PV!hRn->M^mOvVkzeT-r}VA~;c#2!0s zxg=&6xC!t`rfo_+=U#Cc!LI>W7i<0g)V&1jo|POUx25qL95>TmW>;JdbR)DWyuJ2h zc@es`CO4SQ!gb)vl@_-#FjK|qR!wEtThr9mSL=%TMoi1|J#xT9*M>eH^ILgc-RH_M zTJE-Tz`e^JlNiciN6c6)Sbn~n>83cp>MJ%k9Hj09utB)LR9=@2VP+Y|j|Q^1JP+x^ z(eVOyZPcr;bV^nX7gpsdpLz)0?hbKoT4d2u45X}PetWU*rdYpr&c>^Xk#&$oGD~&a zW`w4S7^P;i!U^L~g|hV(LBMN_5X`VR7ENv5SU^{&L4YlAlgZvz-b`>ap9xANHJyY1 zQ`u7qu??y3(a(M{5rtqr{rKE>asBuzNd;MTW_(_ENN=4IDe7vg|KmPUgKj~NPMrt* zfd@Rg?Z0O5NP=f<#7tDbsq3K7V%|QR)|W!>lx+k>`!<=xE`vC9)$&xQWzv=8jH2-v z{t8wpE`2Ul$+aXSX29gS_w(#gm-%M}ygnNoMi!(0;7*c63nh-U))pxSQqve|pK_aN za%8`&nff4)h^eI#f7wo`5R-)AvMpOdu#BGNHna5+U2+_V8Q|;6CWt-C(TpKqjp`!l zM#hh8*1I%MV$tRq@lp@5M0?(C9{M^(&{&H-|EAdFU}ostFp1oOZe%Cx#yXQ!1dCP{ ze;{OqdvAt%_WajmB@5WGPeBSQ+|OqkY{q9rJ943`mftK}yTxnyHo2jM#tjGBmmvOU zd;}xR(y9X-y#im}joYM%&$&_$unyf>r{2jPqG2P;K@xIl{c|#W>OvDmtauF9nqy5V zIbtJKjsV|3@G#vOS9Ffu%mRW1syJpXXq#xPRF8hV1D1RxF@mYL{ zXNgJp%G?2f!X)&l9nBzU#e)p=?Al8icxC04-JR(jQ{iRM{}<23$S~d&0fA^UBtjl) z_e6evh@Hqyj2_~BRRa6;!N}5^)|8Mh1lv5MkjUxt%pehP3B7k%(N*VYdkMRxu5?2s z*+UFj(bqxJn)ZCr{wLXcK^TdNL+@;9JwD8GcsD z!sn1Bux&$zkx$Cp=I#1F!KAm~dDr?6q6kj3Q8*VoHVs2IU&2gt7gcNtt1vBcy)msb zcSX`-V8UX7E%Z!ys3-2y-jR%IK$~eFM$9+d*L@g7y&rd_ZIy|yd@&to9Au0W@lZF) zR7WvhF^nw(LFC29Mp}6;9-`2#2fhZy_IBK(EX?EyVqRqtTt{h>+zwHRNsbPMvTo!7 za4_RR@$tfw2dFC(xUN15zNCVKXeZL>zuQ)cBOL~*iR%?7rRsE@bw5nYS#A5W>!Vgb zyh|c5blaMT#f$i2=!Mc!eHgAn%lqfHH&qvD+ZU#xJ(Oj=?ab1rq=I%$z?p3}OKk2wYD2()WfvR{MjioMgE< z9b~Mqvac_XoP)hWM2{VmAgf%!o}()f;d(x&G%*=6w@UY{q0K8%uRPfs^Y~QlWL7s{ z$>hnL*)Qq&T>2FqLIp?f1#^+1->co~tuT%5?a?ji&+0ld4eyj*0a$aiqL}w~S|W^T zhJ((dN48TAXl@Q4U`jF#bUVV0O3b+LUA{lqF4#gjmKcL7ilk%jF9N09buGZ)15+PB z9aLmE;e#oaqz#fQ&o>yQ?vS=QeDYOm;I?N=o%QZk=7~P*a}?L^g%%$n$IUBHWyKxXpXK9c+l*O z@S2`of`3!@=|R16OeN%5VieoVn`djMY7VN=^$4v&*+GOnad}&Zo>cfA(tyXHb(`~3%cz3*Dv)iU;v_9;e69Bo}WH@S%DoZa%7fIZFb6M{pc6z`O;SMtE(-0+4 z@tJh<oP`vVnq27qmf1Me*pPl^e&W z`_Ib;Gh5nV6sUYBif)ASpz@1+r?1UjLB9{%Okg2q9~C3jw}nE81UT;ajQ|B%cFMnh z@_m3`ga_UaYW?O<;Dq7VSCc>F(_5zOw+>!3%4-k zb&GRGWRb)JFuUj}fts3jFa|p+SCYksC;6*QH!&bH?tvkgS%ItDSeyheyL|bx1W;OEMF(**dOHY1;)1YU@hPTltMsSx@@;5e^qliaLRmz*w+pX^#be`Hour;!JYCU*BN2OCZt{V zb)LN|MY7xRA{5n70Z7X63tC;e&gERjH;?uUUoNmhUGa3Aw=U9_na!<8I0FV+bYHK`CkMiAQW7uWfwJB}< z@8T!2FXNuQS%$EYvD~Um?S36nRI#D^-tZ9n&pyv?Mn;7QoniiFH{T0r&F?(ujto6^ zB6z2^iGc*103tYZ8O3?bld597dQTp1(ha>w&e_a3aHu;^al-pLq34P3`Cd;eciRmbqPC#bJf9X-w756ZJeGX)E7nHq0rg=`uSpb21(Q>^o4MC_eN1bw62qUI-7KQ^aZr4Xf(}Uh zA(_N_7;btRu^>t|?>^sG-lJFMS}1kK1csGY5kFj=d0XS8?-u|hXSG-$FMBw%NYZ;q z?LcIT>JP9yI#AYax!5%&p#IK+$@wy( zts;Z0fORAELbL*imp9wEiQn_aUZ0E0al5IVOUhLDRPi?Z zF;A>zv5#i5=e^t0Ne2lbqtgSP&~6V6;~n%>udWupET-Oh0ZZ64G+jbRqkSmyM_Y9V;Ag@Mr&07_!$mv^d!^{mrizx$6j$nTo>#?h$^ng0DVX*#5 zY+yCOuTs?ucA9>U5Zhld7H5`Oy<^koMn~7u0yxX5L*d0zSFqMJyvb07b3)tvi=-Z( zt3fh}-lFTv0>3YOU6iS&D&%8}Vy;$%0(IrR(u@_R%T{~FSXv0Py++|6><~MVbHsiS z&Ou5}ldqZiM-KT{*|9?38dFVgqDp^xgJmW=MQ zi;8bvlZ=rf*}3SZzDgkH|L9MXA3v&|&($h;|g*O9%C3 zxZle3M8E;cO)iHDTv4{$z{@aVTBeA9#OzZBOY-J={4-#D3_n;?FS)71`i0L$|DOc= zGm|3S*G#*~OT9CxDvU+(4D2As4VfJ{d-y>teG3j>6S$&8%vPBqyrKndXmu+sl8sKp zOTDG`hr~^UN^M7UXZAD}ObLu-(OxUA0QlOcBw&H_efHVTuCpgW{raRn5k@qBt*B*Y ztaO5Zaw0NQOBf>H(4P`%P~xrt%B%)$VYlaqIO{@4oqQvb>5@g-t;{o zPu5P?s?o|K8UJ_~O=BtMhoA1f0Zfl`PmZ0z5=2P?}9xmz3@4+QdFnW z-s|z+m1Z%8ic&%qy|Ia>5-?XuMvuLm<3o@5|e6uD_Hrl#z?b)$)n+ytkiwPM<{8c zwBLA)jIDb0_3pir<2#>*BA_!&p{%1-s<39JHB%68)k}TH4O-wZ3;!TI{*tmKYJ(~w zteg)1Mdr5zu{g=6etG$^)567QTv6ZQ{x zf7vOSDvPV*BisIFTGjIvtsq~Q6rR4c{F>W8YdT9zQNRDPnEjWu^df;}o_s}8ww`Pw ztNj;S$mgo~Ufa%U$7vtd*kjit5z)%m*piXfI=Aj=qBp^!(`dEQ++` zxF!o?q4Iu#%tQI}Zy&pGo)t3plZQNXfk4ee860xo1n6n{N5G z6b8ky@by#vMCpOyn#I=-G<$30LPxn>nJ?F1Ny*2fKP^q@T)(m`syv4lDNID(Oj-6? zVK|uJ)hf6uW%8LSM%=uN!sFtSD=bC6-2anz7^-}9k{0ad5q{&;N{f)D*F5LY|)5a%? z2}Qq+HIM=b0JS%~UNMqVE0{sr^UpVjKJN8cFz*>eUZwXhzM##9>A6&Hi9Vy$}+~ONcp+&duKEPIPJ6>C^VQBr>-_JeW<&eaj*K2zsxEYiR z8Kk;iMcIMcionpLe=nwB)gu<5Q;0GAnecgh8&I3+mp$f?Z_4aWASfunGEID8_Qyvq zv1eSxB6X?N5NL@@^ct^kI(SXxjdc|HuxHZoBw}P%rM5Se20eQUg!q`XG-^Qp=@%a@pyy1Z#UOh!ZS2UJo)TV zMOR-lCwsZfoBdx!ayB7D?w{g~y$qw@g$fQlWaMq1i zg)>Q0mfUl$%k|hd50f?!EM?7|)M|8Fxa0mkcAp)rUS=^LdU8TIC*jgVFW9p3(#6^e z!<2!IfS_ch9_Bp(+lECp>*v<}j+RRYHsh`ejIjIOj_zb@W!;$M8!_v8`C-M0>5CCz zjD(1`ETI}Y+Y3|W)XEs=pP$LCh=>z4;(N1FeO<9+Il_KgH}9tU8C6VPh|qcbF#Fp! zC7YK5Hm{CzFTVO;jJ}22f zE&DnYW#6)mZ7f*^gAp?ZWBFb4c~|eA=lgkmf6qU?=A|+2x$o<`&htFZ<2(*A)|t%9 zEU25K;{3BE8T+W=DSZ8YBtv|4_JMujsO_=B%;aMY9Fl99wY#PUQQI=IEo-21)2dIl z6WkAB^$W_vmDbp6?mHmsy;4V-c?8ak!%N+GzaIH7vu_HoDTMHkZ(sf!0%_bSpk#zU z)jvPlEO=3PBpRuIWj^w@4wKY;{t?qfJ_%8-?VX}S%OVvWk;6te2TMxJKBS+Hf#t&* zwpkxE%ayw4K-V zG2t6q&bEP@1g(he!I%#Xq7=cDm-=8n`F^)6YaZlNe)QgXKQvi0_dasC)+)A4dQO`| zdO2EJU*Nl*9U95!y$kD4@5p1-PIoW5D6oD`^y!%_(hfGE{TjaBOfOV+w^S#8C;Mv< zhuJwXdq_*UbLNIt%EJd97jVa&fDdH!b-NF0@#2l?dmGD!s(nrLCUaC0$x1td*yqT@ zFUq>Y1$kPGppm6K>Ssq})1druNs#g@#Bu3g3$eR?=PBe+k^Z*ZEqR?}gm!nrt`)D2 z(B#yMEV_O8@dn)7LNHkxT5k*YpGc`YzMXuGlpIY9uh8~^%DThDZWY88-u;e#;p6fBk#g9tV$Q*f;k6-B9b(d_P$A@^0 zevPjoS)yoD&8l67tA^fp$H^I(P`?{qS32c~W;}D+v+N`A-~spbFk~xY7z5CA&?ML} zVps~y3TdHhQ0|9~U?PDnJdN;w38lACFA8@YP7@JbXa2&9Uf%#I+6fyK^?w%Qt!RZ? z)X}MVQ*y=0uW4@vxVm}H`ddd|z_;%Pkd#I}3uogKE*vXWwxN;ps@*KEZlomvo1%xEW7LJs~ek@Tt3F8Xp&y$PjmMWvlha z&xz$-j+ZKR<`sWTz^-j!lW`_v+Om5PKY1X~Uz-a@Fm@L89mJ-a37lrV*)(D^C zx+85<^&UT7bQ^d~X2U%u3gC!BZ1J_qW5~K^6UF0RA1_p{bb$=CmhpY#4@ZV8=PiV>K&frzb*!k4!-^b^MY}vu09(UT|8wr)l*SCeJi!vBjx%tgJ=`d$)mhfpcM^*Si&b_xIi zJ^SG$V*f<3i$FZmbMtKcm*v%g)7522@Cd(k=e9(!`5?a2W5##%r9{7M^n~w1kDri; zTzWH8T)DMFlPz>j(H%8~kzWeAr~9NKSteRS+Eq3Cj0QbNkE zEKMOvm(*Zl#x6|?fXGPBn|1^+K(2rkm9BKNE51f6r-N+TeUOi)P~W)vSyie11g|E0 zf_1H7W$^p6ayR-J)gQt-KS~EWp_*ki)p@bI@BrYfg1iC~Q`tG+!As2+Z zV`x;EOruOH=T#T_(`(9N+A{^IrdT@SxjPL@X@%(u54QqqoluryOUGeGBz;y+16Gl0 zfoQ}+h-HlwAh(&KWof)Yt8ihgDz-U*Whvr3*I{ZxNL!+wOUJRZ1B~mXPchA_c{F7eJ6BH+LL@!2`t#TczL4#>1R2}7OX@NG>DxU=r@*8u+{)@dVqg;@gu+);htPZS&SR-D#(F>KMlpxKFW;G@Bc&n@=%V@0skiXQ^&7{qX zv2Vj5tyH+A+xSh;(=BM$EE{(y%ZW(8Q$nM!=R*dV~npyR+#15 zN3*8`Sv=fkf>U)vR&ZKs^*4ol8^Gs?dm0Jhq z^wwnJln(8KiTSn<%WluNxT^I-@j-f$ z$oMB?l|E=>ahu0B!L^nkDFG^j;k@&4z7B3-cj!A{Lgo#h8N)n{lS6mhCzDY3&;sn~ zUYL=M(##Mt3E%#R*=c`;O!R@=%+{{)A2C&0XSe&FlaJa4jr2(3L_txe1a$an*`nyD zOycVb=Y<}?_at-ZJ?BL{-**6-9+!>h7T&BHXQGCWj=YXJF^+$kDHZ)k)Bn%v2kLEC zDeJ$g{4MWa>H3Z;0F%7g1r7g!pkJJLB43!gB)`pd&YYZXq{NlO}0sM;3~d~p6U~^>3X{|Y(P3$fMC=J zlPZ&0SCe`JTr2z>^w8dSUO|cFKfb=f- zG3-OYviR1gnISnAZrcP~N(&3na?yl_1YYM1A*%0yKMw_zm^$E5!v$lwB;4H07PzsG zD@=8oo>MRBd><|BZyP(lNGjVRtzg2G+b$<;%eljazOanscE1W+G^g5q@owdtiMO}X zh#tEoT9J%HLtis^<0fnOy!@?t&|zT6rgd0kQ5qugq4uMg-A70oFnFqsllC@s(fGL? z6MkLkC~#u33TAZmOGkCV2*ibVc^eztgAK%{1No~xWQsG8!6V^7T zgQQuMN8N9kWTx6v5B)lA{*Pno#bz7+&iTJp1$PX z4mTyK3SQT%JYUepB9`N(6fUkX4EczgD7J`_FAtg*BS zC@%0kcRG$bKGj|xq+WBwb~fqTm4+dUle}~sm<#TxM(3E$@KTNRz^z!5x2iy zao#v+wuQM;>011I)|yhYno&3s?aBe=zXsa;KL31oS*LK7=zHbMMK>Q?`pm5HNs-F# z55=}-)yW1nPDn||sat6)H6nEQ;}4i~Rhr`UYQBgclG1`K&?SJL5e|_vUg^`)3QUy4lf_5ySjpuS%%!Vr zLMhuqtr6F4{(l!8gvxWljVp-W@|URAf0m;cCs=fh3@6q9QN06C2JTN1m^m){#VCKh zj4cMngN{1Ia{poHyYGptT2{de^ci30k(Sth?&bgZYLq+r+53p+vK!ZaNNW1OE1qTt zNK)>}A8-2S7gz0krdc#vmX^WnZE~O&isQt?__^CTs!idn7jxR4qo?g8L|K*| zUeXpzZ_m}ew&E^RHPSJ}vlYgcBUi(nskx50XsQ+BR z;|j32nyuRR%iDsQ8CuL(g4zaR^)0Ac`cx7lv0B!VGSN<|Kcn7eAMm`BI0`kZN%p{v z^Jm|vE*|rCpTOo_ELQwE{VnPnKg-?FBPth*4Hy4cLzaeXlXmrNuRN3={^h*j^Mo>! z9Ju-H+^==`IO>Jqcu!olV|B+w9p@d6*H>>}RYiru9uwq>o&UO(e>U;2pG}$_s@tpy z@jljT_}6N`MbqGX=FeLnT$+3*Aqp#89cUT|j&+0H4!POyz+>BRNw{M88QeF>(6#)3 zbBX;zv-tym=ZmOh%Qxi@JG?=Az?>=VKEH-6(K6Ud(6shSRH1Ri*Ec6yOS&@{i?1F3 z>+#Tl@qbTEscV0PR4;0N)`vb3*fo!Szwi3^$VxU|XMhmjq*XWBT-V2G`10>P@qd5o zD+kSXF<0`au~jUl;< zu@t~pC6-5h=&CvY?;G@);#Wo9I+(Wg`zFdCs`FGTSsvSPnjBKx|3asz>TJ`--Iz;H zuc-3&&wc#AFNpFV5BMhA!PC9nK-++Wjg~$Y=#;lxEIS9rz-(`pO_rR=w;EV1qd2;; zN?TiI4W|DIPfXeX#%m7eYW4c{`7gSUZ?4gIYDa0PsMl(i=Ku!uY`*?D-K^p1|5EGyyZuw{d`5dq(B{8HFztc< zKzmzZ)V&|K-W<|pj^W%XvqJ+vKEIRlUONiJBN9r5x#&JzfaNAd)NCz+FZ7b*W0j9W zmN>t(@t*Kw>TCdyx)WLa>mD;{Q!cZgtKIAOWp*FiY^3jGjkRfs%U?%!35e3?nC89h z7kD5onydML$bSEOegC|At##0Q6LUqBQ;%scm2o1fFtSi6cQEgPE}XOk$%>f-QZxH& zxt^%U>69@R29CYXJhK${2fS+|xofuH-iBTq))s*6ZR5V^p^o<)0gL$J_wv%zHL@MeA;pN1Z28CBjFD0F}hd&VnF=Q z-F2(#MLj@Yw9fL;$OBKTHXF+{G;)QtyR+e$*FUF#*@+`M|;DxKE!c?Ymm7km^rN zTgyzm3kV)o`$5!c$bc@Vov$h?p_kq)jaS`1dHJ5@qQ~ZZi0IQXV_+)yC?kMwQSh6t z#wtjzwPoA9`M?N}$U1JYfEO^!#itc_f!<~{7;haq%_^EysH0{7oM9r5eE*z;y@T~` z(nyG~)xBe?Y%3G>f{YNkaJ~G|cDIQ_mgVbDXEe2;FOM9%d3waH{GH>UgD{bD5Uzhu zUH*L#>L1#3Rq9KG`)6Kuuiw7f^YHgu|4i!t3Hy&xFX}-qHFnVX=oCQj?+`!jnRTzE z-$U`8YWh%~yL!ftL)JR({W+1raHEnnzShWPZUnn5*g7t(LeF1ESFGXE*Y0gZbykYd zm{n=rW*w>ziiYkESK5~JXQQw5z@^=mBLSZydMoU5poou?9s91NL&GIjFohCLT-P8k z9YB@5Csz&_A&xO;B?9h#`(wLkh63F3!UY-o&Ai)o24(Pk%6wnb2iApRexBc_g3r)J zkZP5mx7|4PEA*zAeCB{Y=^BzzntK+{vFxfSmIN=ls#SAqIqB`E-zvDjfBmnt5xC>= z9`xS9T)Cn^;;a>rwi=;Z*oO+PH2Twp($30@>bf6qKxdCCh&zm570c0$ltUqoj*NkJ z3zy{j>2NTr^roa4X2aw)_&ys9DW8afK_{xX3THZO=8)GN^J>>OjI2F#o+hqX$>EM0 zSYk#Z1aR}mY#8w%>;-#zI8$Gpma|A*tg5@e+ho0wU}D4#tctC70Adv?Yhx-RXFhh8 z7XZ)JGBWm@G*FtY--dQY#vmqNlRr3w{Pk>KOC9t^C2#F^@84?t zur|mC_0JPZzoB)`)F^w`^j`vJ-NnhmdNz;s_0xXdg!z#EJQIy_{E(I2XsLO;Wk;~04AoTp z1yUUfahLd1SH>a7FmNQJZt?213DU%;S`nIhJF>M(yp(PdEPo@AP5NG2IK2B>1^fy6 zRP|*KjYa?~bS}xEs#OCQb!C|yqe+^}>$$h@wWO;?nKW$w+9?8H_l2NpDzeG#$2#Cs zwucqnS>V5y2WKP@{tZf}PQd&cw(`khS;ex&WC|pPp0K;q>SIBj3h8P*LrD+p?^T2n zTKf==GFmnSI7yT#sZT8nIB66iVzxv`XGBLVkD5Axwk0F zYvsPb0$hKCZcDfAd^2mWCbFqZ`xBP0W@{ZKG^N`R=B0J+$H{SN$Zv(eKYzZm;(EzR z@Bg{deP~XvVu7U1^@ob<*Hiu5$+K-kw*fz#A0M&xnef9j>5*5oyiEojo17L8^13c? zT;@|f(<$pwQ-=Qf}BmNRV+sx5Z=NAT3qdMX%G+j*IVY{!ti&Na>AgzP~EtlS^ zx2zQ-Sh|LKqntRP9vq@zCENK#v;Kq2LjRY(6oqw)ORSk%0uT3FmWb%S%#X8rlc@gy zOEzwZ_|;P2`PEg|)+gJo^nN6A_bU?Wz~69u}yeffYN-Vb|!PI_2C;^|#MKzON= zT^8h8uHJ|!u@U?@CzmMcOIR2mH;aK-jCzna*lmYuQ-|wis+{lEoT0Z^s=G7$0V!u& zy)=TFunLP?7dW3Bvx%F(zA)}qX^=^%f!-~<_ubosaezVd#$BIXYVSxO-9Btp?e>&g z&@`BRZ^+K}&hmwFntMh|2~1uHoPT@9iRB;cBcm$CZ$Yim8AqzK^rI zMUulUMT>u>e>3;81S=QnC9(;C?qxy(TMEw9R(9C^La$%@i<;sTC+-Y_hrs3JttO<* zTajuC#+s|_B3poXIZ*+vR}ET)d;RKgYh}99FMRKGroKf*%Mzgq)Is;_P{6qN9=uEl za4yt8rol5gxFtzm`3FU~#$>b{EO z?LP}@6ASaYk4)yX!x$S_1hq8J4dj~hGpw!eyi)DW(1u0{@167ek6dw&;J(O(7JXB% z?^cN)UqdKeJbWDxgJl)*%E_*5$=yn~^))B?9=DmV z3|YL0lY$K{b#=6+-U`5A0>DAm+FgG>a;&PjJxoJps_uqi22f}GKj;s?Qxd^L*O%Q= zK|YTz014YTzx~anGjABa3L1!Z$Ap`6-D$74sPCkIZb@F1cg2A`f3xvcay zb{N&Ugiw6A-v^W_|HrgDemhW|eI45su78|GczO^t4p-V(!ZO>G2Cm;3?!~K=KNB=~ z_i{Jxx^p>UMOpz^dtYr$@G=D1D);h8d8oDexvD(21o4N8n|XqkAuv1r>i+gWop>fA zO7R!T)#>$XeJGS3|60lxbHB&8**ny}9vX%5-eX$PtYxTYf!8IS?lipRWJOo$voID} z_`ZgE3P$t1cj!x~kmAcnvGYr}F;|xg72R`pJ+~=R60haYwCj0ssYd1tGkXtKme(2z znSF&`iX6DhG+xi>e)r{SU1b1}6Zx6d!}EWqcaIAd)W`sxz~h-S~16BZ;C#G%91H3i;aW695@7SqI!VP@!_1_WufX`WqHO^6pj_a{s z$(g4%Wi1mjJU2!RBt5sioo2=ooqqH>N;)sc8P~yXJ7B7KNgU%6x!;0V%t9~SyV@Fd zDa=J9J%o-|_+F%Fw`#eKX(+%liF#!kootK>b(eTfYrkY;ZE{a3 z%IOQ$RAE+vR#&{NAQ~v}>hLB+x-9g<=`lFd5n`aW?4{R<3K*IXc1?mNbr+TQqzd43 z>Q^bUVRMrEV>+-zImCUSD0snED>K|CQH9TO`e+^y!I1SJkhlOFCE?OtrU_GZ(g&cx zWg-%P9k&T{npEvaCIgpTtoJd_uFhdsZ9nnn^RDbImrS=Z0R~(hgO@yf0s>Eg;|z zVd_e-UF{&`=ynTeWWf^4pq~+j=&;Q4o%L=Dr}Yhcl|o~!bhW7;3)v8j^;)VaqPRw! z4F9(Az4)uIB^MHdVpz5z_L|taUt)7<9sI zn7|y;G>_VL#vZ+?g>4K7>fdtI>oTj&>)LAW+)G!9Z^Pvzbg3gDyIwO{#`TWcrfu7` z!}7@c^IPq@OkP9@HZ^3voE@iN{pr9=U(BHT*g<$@{QNjYcjGSGCSE;3gW8rjK4tx|YY|KeuhQ1;b0IQF6hY zk3vshe0{^#!&SRJaS%sN-Df|xc;R|Gw&5FJ&e-%??c%KJ{zI0~3!?UD;wy2`z0V*i zF?M-tbkbGGgUV80)|YB$GywoG73pgKBS0NE7k=IDBMtf)|Jsh1}QXkK$a4U4sDuIjHg=+jhe>j^ciUmIdKC=IYZOu-+wv`Y? zeiN%6nWAgJjZeZ3T)(|$DGYyt=Zr0dH9JJ#u=>Djk)L_FmMT?>6q(-)Crk+b4l2?U!g{dU19HW zlgx1a{$)f0UE?Kj za|I9^9O19b3WVCF%ItY0i|c2!1S*g19{Akh@y@Eb0O|7pWsKNdn?CbZ?*FfE$8RoECxg!cP+ZEM2=alt}#o98y*m|DmSik z8W6~>82EHw%p+gZ3pj^=aEHCk?en{rPVP*`OL1O)0J=tGJTOgVB14R15G*7n4+Ce(eHB z{eD98vS0bK9(OW*19Ys`B%{Olk)Aio-P~4ll+5@#v1}m&a;e6lZ86|*f+l19_5V_Y zD<1n=IuV%1xr0qrzFSqCj#{Z+F~<4^$%Rbp-?*6Yauxn9zQ(ovhhS7|MrZ77=;AXa z#^GYrm2dBuBc#gg?p2o-#9w8X*$nf1xTmeXGP#FY99L6Nma=ZD2eDS*5j-YQ$l-lB z9M6uuNEu@2h!ay5w;#Srq28J5fojpVuefS~dpv>lZODM#;JuQ${LXVGI2SAjb zq7t9w2uho;FSfLzx`qI`)Ety3{tlMq&=+Z9J*LgH7YDbu&P+IL@-lFRxG3ZV=6{Xh zyRf5H&b6WHUb{qzQ`O~zu%lBjXd94@57Ex-GJJO!S+}m|IAGB#9;CPl)!lTOt5@8d zQ(pvvsvgTkGTk}*fEpP4gjfAH(>b{x{EsSsCA5 z!@?@y87~aFJqack;;!yLVCCKS;*EM5lodO*>;L}P+U7j-xol0CV}(!!GyHz6c*IwI z*^fq8zxJbh@lOV`B9Dn84CibUM1YpsEQ>BwoJB`tMPG(lY@N`^Fx!9l)QZyGE_!x7 zj=_3F7CgW$A#vOM14E;QU6ogHE>Uedmxc#|x*QQk+d7Bl_h4GT0>;@x+3GH2T+eTTUwGz@yz6fq) zes}!zf^AL5pJv(40}4q2A~y1m=OYmA=lx8tUN7^^lx>-=fsw68eFU~()3*5NsolV- zqRgqUD$Lbn;w5?sR?h!oOl)TF$`FzqQ_f1ab*8R|_VVfme z44YT?9@o+`4l)VD4@3)F3mrsYYHAVC9>8hFvCSV22nsXn_dNvfS?1d_R2(2XGf3t- zxZrop>v!OrZZ)z}I&u2MI(vFqofmuEsh2y_R)%*wPu{-N?i@Wcj-oC$Zx^qjcE#|m z6I8g2;k6YvyY)<}f%8lMc6auH*{yd`R$qP;&Nh5gwJu_a)i^7u30Nd{%#3S&tYf9L zi&Cb7+c`73%^MVTIaMIO(wA(O|Lw<0hP^*>?-kNB^fJP5+#Qu=iTLRu>GdQ0*;ug> z5XrL`L@Q!^P*T}KE7Pt>#Xxr{dD!l@xlde<#Ja{ zZK?U4L1L5+ul8#P+%7ev?vPjRmcV?muH5cjQ%!6Ngu+@*yC2c;Uhsj>x>Y;+qoavg zQOdS)8?WP!rVtKYMETOr7lu3ptyfLrscPjVVEb@ak9a2lsF+7P%bV7^XQ=CGELvME z1~Ev1`C?^Oi z>wY9%nNLc67;-c8GPd1m?%5>BrwcAK+92Yl-3Edr1Lbf}%vcCzqGl^`Yk5-g0>;%T zFZXkochTt11e1}WT%76G#&dpX^h{W0S4*mLkmcu|qKx~P8kE<8yNr(gFfrWm?r6td zK$Z_M1~@?M;(4^ss=c&Z``|KS$=-l#OhSuNBpb` z0X>=Iolp1|OT*5e<(lwyx((_2?S)pe zG5C(W?ZrT9u%qUTJ8t|%?)lwZdGG>&sKsADMxzVOqAq@$5JmN%bB&CuVGCz1HUioI zC`)>(fPF8>PCxRy9e&;L1zXH@+^Q7lDX~l@>;{aY1T@adY9`!`xzxiTji6q^%hjr|VqJ;+PAns-T`MpjCm&JuT z%xDg$tHNh}4V?MDMt#fbgHwlj`X@etJi>Q4v{o!I%#e>Ac3>mODXI)$({xoDKtvj;5XU|yk=0| z7T%h1Xa}Q%C&PO#0DXOOjCh)Al;JiEKiO+hBp~9orY0@3EF%Yx79r#v-jnT*J>pW> zJ($$^Ro&379B~hiM#Wop#Q~|dnM?%J_-I|{$C4JWFdv3{8tf7s1?a68z6`>Q3uxWY z9Ee+MgK_t(#m9GmzTt9fO`UU#RkW_va~MBk%_ z;jYDJ-jtB6WVO8$i)Eq=f<_uaGQM2FR7sr@ujr0+*w8XxAS_-u1#RqQwzFE$&+UAy zq4-N$wx-O?``qdh(Ps{b{)d~q=E z5;P~^beCmx;Z+rEdXg&7&T&YP2l zMo)`8P|s#?@;c6WI-`>^)>3S}ELG*QP&{14J}REmW!U>QkKGFTr#ZoK0&upts9#$D z%5N1qSSJ0gJEWSrlByXR@OrY(f>`ATmMmWd>0YUfv%4}h{LYcm5c&E7A`JcM1oAil zw5XyomeouSqI?*I4l1~P0aFgGLn3jwd*oUcDG)jKmR}RtWJWp!3*bEtIu-t`=nGvV z^hU8sRpSAUjbxRb&7X(!3lX+^dG-=ypBI2?KMGssv*|V!g1J7y+;W?ZZ zphZf7XiyE1NlWv+C5R;E^CBBd`zm*`Ata*aRv2TN@j}INdRNc-st<-+eNksa*zW7X zc&WKMvlIU#e%iG@S(=G*E3N^ND!UdP{nZLug6sVFg!oyJ(!xs`d*RumTTBg~<#ICj z2>Zl-sQnJ;@Bx@Yf#yat-*B3CY{p;;r+hp%Q(!)Suf`3pvc7>?4mfD7!D=s0_hV

0*aNfFn@jOjK;y|s1xjZ(7x;-NpZN8sG_@xT;-RoVC8zqGPiREopnSOZG$pl zOUS(c03h(D3Dh7ry}j?&U$E$q*%fNt4B0h_#Jv%la0sqC9vQu*%ZI#V)z2n*Q|bt8 zw-@FIY+DH5_gCiepEAvV77r{=3?i)`lx&MB?FeaV+B2#$nVRvzvi8612BdJDixGCs zLh|7dUlU{W>E2lFG^G z8(#0>zE65cNftdTC49f%uMkc_LY$n$f4j^U>Hg>NZ=t}xngtKq@AH(#&K!eTIL{o2 zN{TB9Scd0NV+oLSkmVUt<+Tv+o(#y_u`A&YPl;iZ*5Y`;!rDh1CW4QErAdYDxs9Iw8`#}_q zcNXM*Ru&}q(@{Rc5~Qa*64%W?>e`xS>)d$_>&wt3oX~^;#lV3o#MUd|S^0KgB4(I^ z=~1oCi869Llr^OukV9$p4VqBKKu`7`W?30^Nk|-jIW{Q8)vO+ZA~VFaenHmU-L^Rd zx5tk;Ct=ujljP!OhY1ULy2f&s)uMSvU6;hx<0h_yMmwQrJoKW6a1?Ru^2_55(n?>t zCd&}AHeCr}8zzSjwZmiKql1s)Pi=2sx7AU|8Z?~SaoP4bUiGKSE-8@ck|J!zQK?!j zSFm+H!;W$`*x7+!K%^sRN#Y#_S>^QtkH{xjK`QgZV7;z%!nFsbt?I#Rr_fD)!0Z>) zvi`-WWjAF}<(~X}GI&5uxGgXTepTtt{hrjtsrOE6iofBfi$Nb(0DGg1YM(`1 zGh^*>hL7q6a)YVgjr$cVLq-C|wU?yMOIht`dhGmcMjf4!u*Xo~mJ;-Uxg{ttll2(8 z6pBQy7|o$1(iSs05XLCQbq1bF4&z6=^@QZuIApDwrF z;2Smz?Xr)JcC-nWs61fv@aF3-Cfk2Hu`d!jF{3PFr7QNH_&wjXRN6Q&0eZ(_0>wqb zk4u<4VXMbqt}5CiCVUdD{1yXE8Iu<}hgtn9#^olTy>lMh(mR-Y*%_t5KT0Zn_iARR z@73iGkSi)-JVoer@gFksKn>ycNX3(x{H$x*6PetLi1dIvuw0d&DX$JU8NnirBpjlD z2A6BdGo^6QefJfi;Q*yd#;5I5#sFYTl@;L$T-`iMZ4$+>Fqk#btApDM_F?vO+y4;V z@In5*JahtzC1tAYp3@?fz5HbwlAy01o@FSi$b4zTU-=JRoR*F~OMn+dsm%YiO z(GMJ!4dKHX&p;Ga4vG&(Q8Tn&AYy7|CGq^Qb>(2~e zP5>+}jMu+CH1@uTygqjS7K`)^{F`0?>A+X!>%x9sWf{ zZmntUC($gEP6zXMuCjoZmpkW^B-ljvp(h|`DNyKl{`vcCI&kQ^7RX^A&zM_f{=4c@ z9s)wiY2OmbU+eER>d;uvdxPe20Jg*|CZmSyWlQVh_;X$CC&rctcx6hSbSLH~(1+kD z%k_TRgd)<^FUAsuy}#=xA9D5|Ryr9z5Mg9~3+13Gn<9|OQ2u3@& z-EC+x^<6b`Z2=xm>L$!O8jBpckO36^{pR3}CdATL-h~mnPuaqiLgf|9D?8?iYMYPs zk|y&r*T4!{0pu;Ni!_NM&F#HKuv0c=0@Gd2fQ(n$Tl;|B@u-{$Ew#w^LL2JVxSzw1 z?!CEDv9)5vodzJg+JX{$EF@aUBhp|+wDjh*sXpOTlG8WFbmQ!bj?mz^*siPKOfgzk z2^OkTV`o|2ID*~oa>OArv7OOUO{SIuC;BCmk4x`fy^$47fnX6=h*6NvoLDo&}FtC8lG35-+p4FTBgbYvsQ=sd?bjyggr zlA0R(HYj;If^hhz)QiF(TzHX^`s8l^LP>8C`9R+oY23@?sZKVNPM^VlVfR+jtk#4a zr#|)Xqowf!Wryd-p@+ZQXiPLHa26;*=i2Su0lRgBfmQYa17zo$D6&7;6v+2GvC5jq z?0mv1i$I57IE0F;KNVS_dmeOtOdIrv%;%&Lei~5%WjdRm-bdpXx_!xvGe0+jQ~?X^ z60|^H4a_-O!4z$b0tmRP-D#@Rwx^tC3R8PaCTL1JWYgFT6gNRJcsCPrxKrHsc)yfV z=LPx(vpOer!&02XB4YvCMdM1qJ}H2}wr2KP(z%HcsxG?|EDpP;5_jW`Ofviqnk?it zC0tT1@-^S)p`xan2Dbd-MVi8HUZjHuTRDD&Yg+fC86yW$^Pb!TYz!i3HBqd+b+G2oeb;yB1hVR9aAL*Jw$5o=g&h@3pUAUD{LJe z-kKQtRpn$Z-c-wAtCO?=y8(g%w(+jL&PVw~TB4EfilZq9pR7BMIBt8^0ITHwa+Q+C zBN;WA>TGcZ6sa$hKs8ngos&Ftd)4N&x$;x?2=FF8wCCmH{Yq!y^CQVswL4mR2Bj|P zQXJ@L4_~rt2%$SoYzj^U9=oCFjq5G@cRRjqP<4C}t0RRQRW6zaD8m=Hy zET+f;^ZFU<{E+p(0e=&m(So&`)PEiCLGH5~cR0 z4XF#P$5b@+&?V1iC5kPFuWHYNj552t4p+1-c|5Xh1S%0@wen@T1*}oiL_3LR_HZM# z&s0(4HqaP7cDPJ~)G(1^!dRof0g$SAA4`*kyw+5mhf?>Gwozbi6I$d+21d_A!|nAY zJ0CCztFg3bYdsgx*KG2og-dsTDd%U44&a5z!Y`GDqaY0<79|G5?Ht&xErPa2+t7*K z+VIn?a<0Wo6Z>HXE1nM3;ioQle{|hg9@k6IPDW_>t{-G11)TQe zrr|Ih?FRA?1vz-D=g^psVt7>p(yn#m=@?GWngZnP((rL>lwP7(@F5T9qbUh876s2n zirYKfj!r?xDtXlnVb|B=q07$;A2r3@Uc>n3Tf2f-@xW0IWUw6LxF~z~Ju)^V?oru~VVBK~-luDJYk zZX%IFA$thPqJ1o3;P*RS_|r@%xVoqlZB6Ts*Y0W}I6wAFULUgn zCog=$MaYPH+T}&fv+Roh6wm^{yJ+}25x=FS6MHKtgcg8iweO>^JKXXvl(kQk_EvOx zA_JVk*ZRXWKEmP@EugY@a4)-h>JrM~l>^g*pO9$N0SkA9k4D|^ayV@$8-fJ0MaXA|J>iPnQob{RTtX7(pV{I-SWYN+_5wdI+2h*z|)&Rd91BYaoLVm z1uYyj6+EALXh9?UQ8gs)0@_P*uH~Qdwo9! z#@69F?vY#BpJVvl0uhH&uHS;6$Tlh|HtEs_o4gEW%7#&z8SGlbJH*~csy#C8K5omj zH;y7}71}BUt*a#5y%=_fs#dHa2fgjG(Xv8kD9!OnC1J^yRU|KnQVXo_xJ^*o1g&c1 z*sV}1rNMzy|1PO<<;;-vk>n@?-wH6vkI?*8tJyI}b$@HuYhOW;JL(ZNn;z*NsZ!QstfCA+eDZiuqKtMqe#~yLkBDszQ*yHQez3Pk z<`{OHw2=f9D>sG*tbX#2i={3&v9Bx3CZtNED`pU&G#t|=(xno_P_Vr(bNLh!T*OY@lKWg7eN3;oA{N@=>+he5U z0+?ow?EMubX{xbJPd{wM30Y-48$I!S38VC_%lTnqjW5ED{`7|Y8v4P)Fw`%*9f^#h zuTsC!t%16WmUnuj0r)g;Oth#Y5%gy-DDTNw&&dKdq6f{FtIvO>L1GI(57Mx|#P{Ua z8|$NP;J$2qdc$lzu!3 z;$`MCZ@f=tm2faV1|4l=^K+9spRsRWUhv@59mep>=g~r_Q;2@ay-rf>t$O6-VdAIz zweCNSH$sm6TQ2>61S&cs>+tXB#&q$Z!xR74tx#Gjid!8iLGWuJr)U%f!~&BBKCsqZsFc5Z9LeA>37>fc^8KY&r)19~PJz-r4jazS1! zS1;IY0**;wc>2}D%W&2u*c81FNXt%!uPohG#_dRi!p453ZMO)~Y;A*YotcP9DepSt zP}MTL%0TXJqH<7sPT54XvmQ%VJek>cA|(Xnbqhcq-*7sUu)$hxYqsjns($hH*$xMY z*YJLy;M#Ohz>-}um5Jtor%Zkh7!o%z|yjCrPYJF4~8c0nsKD92Eh zP2zcZFGFY(KN zGLVBDyLcB7Si(akOiaW~C;FXXnCG8%`M=ry9m?81@&CNG(bCr)7q1OhBZE-#)5^rC z%Zcs{34>gKvr1e85~cT=A%~ZQ3KOulPewjvs2=D8=cxcqA+82V{?K7kssp>9&wK{Lc3g({c~`&4e4Lbg);=L^v#&$Hv?-M*(IJ=W*KVWc$8KjcPnrRzJ_9^`#m(rqRU1{#^RQ^VA3x{R(-`t`VK zxFP@;(%sPIfK7R*#@O=66l`|=41d_b#IsKo)3!6}%rONwr{L;PKpiQ+;`+ z-g3JM&LF#44Y^<{Gq3!2szA@Hx*5KUsc|FK8K8HA7_4wY8=#Zkd6C4qLm~L9;=lM5 zT<(I?W>Y_Q-DfK&o%S=Unvy^qVDE+U90n=?TOqNO$)A=X0qtx!$&n=GHmiRhf>&&snAb17?fcnISu63uv zX`$$pWLCF5gG$R+c@KBRd$;C!CjSnvo%{6-vkJXOk-%2FWO;fgzy44&b6~0 zXk<0AbUR~P4={2L4)9fo5jeK1Fm}63ys7HlLxaLN?SOnciKiAk&a-U-?1tjyXQ;Mo zli{uP?mts#Z^;QP^kzIrbX1SF;;RecHp>^Djvo;ZJ3nkjTk7!tvG<=rO|EVKD7;h< zL|dCXfUOfqmv) zg5K|1&;Q-CXMfo<&j)8RAtvW_oJTLeLxs=jrpshF!i4bBnj5*hIF_2?&KA;}y&gor zX)Z&SPjr=RV##CwoqQ^$2qZ2P6F9BMvW>A%js0kUNYu*ugA4 z>vDZ_*Rv6!no(PebCI2UikR97A;Bb+3A8)&EGRCb-rir-EAi;4(CVcU(%>*u8uL!` z=jZ%3S1?=w%F;8%?r83dbow`~hcp&Pc4R2B$<9jh-947NBN0-(iexRh$fb#NUvaTc_dqj1xML4To>&ax(KXNw zdAnr9udma=!wXZ5Db|A_MjHv$&>1-aD+{koc+%Olwa{KVGE(Hb}&BpoXj2ge2H$)5Gw4slr44;$B4iSyt#Z6%D8 zsEFS6h2@$A(*-iQ(j5F+u!m*KU(WRXetWLs`<{f&g)+D?){z7E#RZD_@dxdBvG^v7<0 zuQsyXuzj~zsI(_CF;Xh9cu8<~_{>5U!@F+qOHYSBPoCAr7_W++(z4~-mZgQ}?A=(- z7`q-Wq$!8Hs_+jW263*76Rv3wy^nst|JKfDh%;%DmvRngJjo3}(*Gl!>Zh6pq$h5s zmcn-&>8PCSb~WfkZvJq|S91Y9ncb#YRotEx2aB=ADYUUW$ z(=Sr^xiE?f1GOW= zmMB}9@XtY-1EOUp>uOZNaIW7DyBbT3vB8y5RM4V15)MGeVuoE;0x8j18rGQeE!bXe z!drhzMnml*Kem2O$yeXb`I?ng{ISE4>XSBM)2O5kH1IaqhMneM zyYn-ATNK)1W{4&zO>#CzJ&U*b8#i?mb0$<<)3eavmaY<3gI6|ssg>GO+TPJ`M9Rj&KKpMy3Y#@4wG&E;0F7}R;`{Adcfw$O(e^Z3@v zoNZ+4HoTI}@i}d0@)`m1lHt|G8`nb?``4-qMa4GVX}c{8P_G^yAVdGY#sB-sQDpO& zHck(BXS`|-(o@?!)f^##6t3BPlXr36#|*L8I0NES#i~0uF3Z-9AYy;E>=|J)^C~v9 zT2s}s9%P6oIAP-pJ+CwNi@gh&7AK9Hq9J@$5=A{VYJ7MxC{PuL%%!^e0~L z;MTrYoh)9#cR0+?HD*3e9uZ8}4}LkO{q)3vKmPj;s9%HnG3l|-PY+HF3L26C-$Il` z=KNva>Z&A=sx$9p=+E4+TkoZ%<#l)t`bzm*a@ARs*+zxR-+~q!5*_Z;i=BqMW%-Zr z`XBnVwq#yRj9(*R(+2>HtwwBxj+__m%plE76ouy|jn?;TQfr zOaqxc0T|<}aOTpPgOmD*_Yy0U9**UQl}Qai&Z5l*{GiSLMwghw-lzNSOGS5cak%Nl zjaw;0e#DuXruG=cuFdZ>Uvcczm;X~7rCvBX*J=3H?dK2Kbz^1xNPTcv-G;Xbd_y34f;B_B-lh=&QH_G4K0Vg8P zbV-Mn|B?zHLy)EHK~P={`9l`|_fUWx{+Cj}dGhzG|K%wX>i?BKqPF~BJ>~QNZRH8c z+_iDqoUW4LOFeWSAZ!%I-ggugMhcz{Uwgjx{Xy5uUuFnWe&TZ^uzSmuMfJaZzkj|G zzjyL;fVf#fc^SAbq}!r(o1L9$Qg^0dk^cfZZv`H0=JPXkdwa^%)h;oWFeN@ zo~BYC8fDF*m-rog`9BIl;1A+~pug$*J%#gsz+gdSz~ryQ`iH}PDFDzH80th; z{$F@4|Mo?H9o^#s&$s8b`Aqom0eujl5P*wfmJt3$5HKnRp6?6eQ%1)jlmkERX3kp+3)e3M=~P= zpOT&7diPL#T2n&&ZRX}!*qDR|jcWDvZV75misz^Aza8^jqiV$f1p~VWPfC3q_}#}$Smw!oz4O`q$etu$C)KTs2M*UC zk8+r|Sdye74gS9{tCM{y{b00j`F6x~AFh zX{*0~$(I2S7U1px`Uvwy)ED_JF5<$7jD^Q6~T{X#W4pWFMZd zq553yW+vt*K2hl1Ala1iW>uYVm|j&?BFmH*T8YTpQj%N zw*RdW{}u;GG%?7TIVOYqtrw3<0%;OIbu8(Z#Q1Iz1544@yr}=PU~+>`IWSxN7F7s! z9Ee|9{09!V>uwQS{Zk<3f5K57NXOyv0%yRJ7;XR6eSp{l){Xd31N=bhM)7OEE(9V9 z)MO}C+E@PnBelYaPd5^H`439oBf>QEGGPz11Kwbh$?7$k_`|UI&pSUj)YB?rBqr+P z#CRyM5C{M@Ap3mt+d<|zkP4i8s`0vG$R9uV2Q~iY3Gr#-Jqe`#35Oa+T&<-LNI3X^ zJ}Ka<@h8Gf|LrIT@A*C?5hn@rATLT~;o)S;a9WXVu&(V#T{I@SSNLH=c zAEm$DobT-(<}M#)2I7DtkMB9WOOnZ|KP%q<@ClBPlmeZ;a_m4ULDopGQhd}!u8!qI zf#LdZk7h6oK&R9K!%${T8g>uO^7uwBC0IqmASwnLYpr4*7^C zN44zL!n6JNMkd49q|C4RYdF^>@=xQ2@XiBY=_gnu>Lqg7iks$8_Z+AtYH^meEQW86H)36Fo^|d@e<$8ziO9vM?p$iOyEgkmoqW-pq$`$XAJIv~8*-F-ba=a|M7Uu&1&u>`{PVevko>3x z_jp(AvmWHvp-3uy#GWthJ!5W&J6h)noK^3@`GdcGrU9H~Z^BabM~dx$B8S%?KG8`e z)=Gu#_z@j!2UR%KQ_Fp(X^zFc*m(9GK*Vdspg$HKKO2aIu3yMGRKETvdI7-M(F(SQ2CKDfuu=XXHZ;IN_ACS~HoBXbs1ETQL3V*IvP}`I5 zg+1{3hlAy?EWs6<u9Ky1{5g$4(hKx5M-a=x|*;V-EK7 z3|5#7Q_}b|nlZY9mE9huJ4lOJj}jIc$bAZ$wEq`x@)s4k>hVyj@-wUx-jDmDC5 zp2kN!U?V>73j?s5x1d60r>tTh4hRFdHM?)~yuS$kSqad)$Z=$z|KJ9G3;yp{6SlsU z4E}Ij(q*;nZ3@vr;!ea6U6*h)>iMz|fU+H{blc|8`7m!On`audI)FXan#327LnkA# zMPGxJ8J9>~vWyjR`H(Ib*KLq>@U_0|6Nx#{VxQ`g!nwYCciL+<23i(9$~!!IiLBQV z^yrX&^hBvJlDvE<6$dx=TeyJ`_>1dNeKKo}mwj+*X~8jnZ`Tum!3i{3*Mkw=)mmq? zRwZr!iBB~}MF-9Dx2Wwpdkd*>G%sB{udJeP*S&FmIe{s{ZRlEgVW8YrphffIgD}ej zU-BUnHCU^%!*gEeKR>^c{Fh;{C(g>8_RmusoD6_hQlr3iT{*tF$l|^+y)MK;9Gt$P z>sqZQR+xa$D-lxCZ{)t5T3}UPNema!;jMVIGu}g77cA_$8lgHRR)jmffKpV;t7DT5 zzo(X|@Wi%)0)-%J_w@^y>5bzHS15}UGOyZw3WGK)Cx<_F-UCo;GIb(V03P^WU2e(W5M;h?OcS&YxWpo8Dj^w9guiQg zb-r{|1m{_9t(|Vy3?zCr_U-XQN@SlPaiA8hp2W@fhsINvQHw+BPPm=)_w=s2m!Sly za!1?VBLv$J0*9V5{yPtTX7Jsj_D87foiNv{cd3*RL>!&&7#fNv zl)LQ43*C)cqxfnDD=fQ27od;tReJQlOcX3>eOou0ugO!dCA|LDDHJ`?Nq+7nYC-@{K}1%6_I{Ntu3L`x`UBp zZ^~TNDvf}Mx+=&VJCFzQUXr7?$X99aV)Sseylla^^H0FdtZ0ZUJ~ZN~6ZK1c@^fHa zaJrTof0Ukdj(?#e8T}^r{J)z<2qnem8K6pa#>pwe?wvT-)*IBg95+(%M6Ys_F0TGr zl}_2>A`P5$60UNxBGbX z>?mY6J%`V@0|EE2>>lf{(&uB>OwC5FfO=NMwSf4^y+Gu+&bKF9lm+p31zJBgN~cS& zzMp)K>y$gaSjXC>=`~ap^h-niRBE7o1nfV|+V$vm?v!QC#>2Kf;z%i@^rOSL=PC_Y zWn0TeQ_ZgQc=oaWg^EjA0vdU9Cgod$*}h&nG1xfpr1k zrwDHAw~_4nk}k``jNIqNydG1vM)L<3W)9Us=sh-L^0W)%Fd6`TeP%8>Y|{d}&5zy> zrx$ijSl*FtzQZW9pw_hP{0o-+v%nF<_s)jHNuqU&TSF@%(Zz6}t?|uaT8Hv3Lwb~* z+!lCGD8Fk|ZHav)Gg%#Z%t*@rT!O2jc)$eaEScY0;mMPDog!mW_bf^4)+E7-!DlZf z3Q!;7BM-jlztt*)o5-lUfghn6`Z0B;1M71gtOPI&9$+_&)$fMSUn*Qi|Lit7M_D1cj6v@zlm1M(h5rH8ELOY=JC>9GRyeyHEtV2PP(vrBloY(zXkjIf*M#S1Hh zS>>6Yp7|*)mw>UY!^@g_*EK{s&%>Y9uY?P#KPV~OhuM9To-%3gr#B@`2FT0eFN7$@ zG5#(3p9Ork!SE#j4_j(?=_DWj<(&*_|8mCll`keXU$~(r=2&@7)>Spb8YxOH`7Ji*g_WB zgy}k$)|6}YG!F7MqfYw9^L;8c3l|XOZ?zd^+mHnXIvPnbrmG4z$yEX24hJI_ny(l)5upQrlxo>?u zYlG{oePP)C!-TV=JS+J+b4Nn2z>MI=g?1FD{u5W^`+i-7{C07utV;ZC(|QDB80jQ{ zx+?(wy%D462UoM!2;nSb;mMx4uMxKO`%PBuaeN7<7KS?_!6hlu^+h=-Lrt*@j4|AX zVN{Qti%=LYERVK#XMSkhR?hC;oLHvb!uGn_YLdXRNA(BT9K(XZ0L=udq)C~9zr|YY z*F^AVvLj}1^$ihbk70QpKDG_ZWLqLQX%{PF{FL7+vibpt=j7ry4m`)&i{$>0S{iQ@ zVrSnG9>>Y&9sA_V7m9{+-l4b(j=#u97=4E&wi4b~uxvU6E=cK7eDX{kJ5|T71*scz zTx}bkv5cqn$)~QxEV^Ggz|KAJtbQ2^Y5|fT9@O3))#E1J*Gj%VMr!1w_U3HA)I)kG zUaqj~pkzEB`GD>96^J4C!`TYRZc}hfx;dS8;7GOnrqz7s%!3c;)f7zmYufJi`6q>M zcfjaCy7vB=&fLs=yiV;iCU$MCJ-dEtHj6^+w*n<&iLEm9eq@lk-u{Iws>E9+8kKhv zwK|28CAOFd)d0h1aHMlQS4vsX5t}%uwE@?h>EqyuEq1IKqYm>M6Epw%7yt>v*gTO* ze|xmPm0S~vWuuM{T+!7@C7*PMMGT?GHv7H1jB0+l94;G+|G$Ys2(K@m-XC}BBOE}M zU=`Ma)L6p(Wk>gaa2KPW5c7)@0!J+Gxb;mGsL)W?FU8-2a+6J#CGnj?n7xE=hsB&| z#4l&2GW~1t0UYc?vPmgP$@C{ZlxICQGFjx4H8Nk*bBq(Vg1^S{jh|8oSwzTSv<){c zsmBD{-8CMFg;A6#YkzdI+}>DKQ1y<0_cfO0L`;aU^>$^bppMhzXL=op9D}d;p z^ij0pm{Y(cSOvkbjPr~1pKG5@MC+#o{DAhcd{<3l%L>N8#RHL*O$|3?aPt4jS zUo1QD34AY|907)cj|9Op(GYZOea|1)QAYhd^yga#5H@bfH&A@aL+LIu-v0q6>o)`& z*CG-WqeXPqFKJpn05WgbTlz;j_s?vHFB8!VezYB7?l_%g*iW|xI;GkZ-A~SZ{;6&X ztBpS`S`cqNn!*PhVkaHZXQO?k2spq~M~4H81q7Cf>AeHrZH;g?d5)lS*TR|^*_X@r zGk1g+!BkFL3f*2<0mnNV3@SGroJjg~u@)Ct)*x9@W}F4Y!>WbL@&bUEfl!-gi*vRUTqb&T49^&S8mWTlbQc&o%O6o+Uk zJG$0y`S{d%}VOxQ5_2c9f2+HghGey zA0|EJl;dJMcI@hJ7O3bSqzJnD5;bg1(34};@>r`vok^hjv>L{9y>`Cpn2#34lyMM5 zt$4OKX39{`PZl@3W|~t_an=B0>a8!bI(9vY1b`VQv!UVRkB5j(PKI|1`qT=$4zC$6x&~}q1yj%i_;~)DYbGgf zpAAqi0K|6-#`2@7E|?xO8Sr7SxU+N56owmvb?3ncenus8@?b3L&9S4h)ZTr#Jd-bp z+nMyxcCYbddRE(&L{t6!En75YoRDLoj&h-CGvl|C5`t^D3+sW4^~9GL#rwQBTH@?w za-Ms=>Jc!O8FN`vA7R?mPBq$3jPD13IMdiJcf?S=NR!*4nCZt@wVB5%^`G zei8c!6Sd*)nh6|+nM_8OjY`)pN2Q|Y`RPleoM>&{vB)HKYy^A4`n8d zW^$kB(RwJCy|3+=Q9M{@)#S>H%VTv(B08mUYE3h{8SU%iv({&G_XtWnLLWpwC|_)i zV@}YikAT#!Xa`V{4U|k>lZ}6v?6vu%HR4*#(lA`SzYyP^D7EjF{?)I$+@0Ao>FD3l z<Cf=v0p#I4*0!4TMvlbW1)3ynHE#(-}3H8Vezr zB=F|m1_7Hl2tIfK^)fbMqK`Nq2hhHSgFcCTjrgYc2mgcX3Lyb{a}TUh?ykIlZMar$ z#w-In+_&ruSk<`^zK4{35R&jTrBPQRpnmwa$~Ml4oYl)zHRWN?N!JGxR$0Kv zfuQ|*q|-8VAUR~rWMwdoPbfWKa&@rUJyVC*mZRQVp>sTb zU;rD)M?!&@;zRUOgvbqDjp+qlw`*CHq8^@QQEHec@(Runv?)?)rf6zQDyD>|D?04s zT|zw`lTP+z$Z_%wck%*w8FGV|zmq2aAP{7Q{4>yp7<6~Gj@HXhmYN4cu@v0uiH=*CmX`3)ygn!JX*r88h5>`nM)B|ylrK&h-SRs zR|M9RoPjM7OA^658?OwU60|%`kBRMqdfM1(`t8Fir`yHKao{~JTGH71`?sNw<;7Ee z$@0m;oq57}l$_)>NMZAx7w8fuRKs3kV0aWo0GZK`PY|=dr`;M8^7SvtNm2hn|hx@HpD_R{bAU1EZCqT?E}g= z_H+5ao{5@@*muwYdTi+54>>3GW@I9t3JI?pEJY4!;uf!us1qipImmg~U(S_U)0BFy6)9Tksnj<8;Hvkumr^IW+neiwhq2OPyBsTv3e zhwX?x=N}5M5Y;{`LH@pJ2sNC@EaOFUZ3n>&pgP!MIR=j|Jw!vecbyuf6iVW|-ddTL z4BaW|{~B#B*7X6!%=x-N;L)oYE(>j4q;=V*?#*uoS=4Sv$DOtxon=)p7Q!~|@R+nw zsV^%%DD5h|$Fv6mCjlcRYSF#X9lcveZ_yBU$dQm{gn?}bUrK{KJ@l;5tI(!NV*n-g zW5bhmbM}UyE4v2o@hXOF=MO6PAdgAjq?``RUpR(iKsqgFvM4~pd3BbyHY6HOS)CU; z3-T2&Mlx}qmERkF>P-9#2on)_-eW@Na1(ON7Ivip~16zZ?@f!CK9zSsR#8|gNj(WToN zt&qbPkKQ+FzPun@zdu?AIKXT5HaQ%uKOQG0~@O^pFgj?F-Sqe;P0oaiBg_y!! zr3*Yd@88I?F(1c9gRIzt?E$>%PoRD|sjZ&aTdwN{MRZllFO~xe!J-7EgnPW{ z_E2g%Rc$OIRkWpmjhxD>d*M){v)?C2Q64MNoF4@;I?!<$T1}_U(07WGq;lPK#n950 zY^6`PyuDYqs0xv5#_aEAdl8yR-K+0e&<@?qf_=rOQMHJ9BIE|2<`_@otCp61)wV2A zH0F0!-g#WtpPuS>&oCaF_U$fvI36m_h<`}+m6XV&KZz^6MCy-Jdi|EAe`sA|Zw3{g z1z%c$m1rrem0ze@#N6dpUFGV4*%P1e4HWM>&b%099=a$VL%mv4oZcuY8+SXye#>)w zyxU$Wrp8i|dTeXVoZ5g4-M_mNV4ml^@kK57-av)98|4Jp&^;570_RUi;LqVPVc?k2 zd4cg{0W&q`3qw5^#cR;&(l9S~hw(H%$M~8Xc0apmz>?>D_q(jp*riD&+{TbIuMt+8 z<3qFxTPWQHn|j1)p#ZJouattjgDD6I<0R~lg=oKi(`}YE@`qi7^Y-MQN zo06UzLQWBu+J*8M3!ZjP9x&oYGvy#7lYIqAfT&=t2Ekeb+&Avk)=NEK+KXBXvW6q!;=JI?uN%f&oRYcR`GYE4Hd%Rj<=Ys!@% zO>;V-FDA?-W>w2#LXrM9s6Y?*6yzM$>ZU01TNvhGndcwoFw4uhlA9ji?c+9j1bbev zY4XZ7UgR6ySQ#NlvU-M+CTz@hMQen!P~2o;yxQD3^7f=Qrcv+&Yl6qKPgS2^7P(?$ z{u5M*FA>c;{>y8IL%aSTE4|e~ch2KLN7)l~|3JfD&Ay?zK#kGBamI}M%d_%}(>bHq zv;MAAL0!>_>4#vE8z}`&^ZiT75T?Do)y^C_w4B zJf1y()&3L7x@tt?R46u+}nwTNBw z+er5n#dg-I{l^>!!iX{QcmQ&5Lv(mg0Wqw~FX)?>hH}j*c-KP0s3)V*@zW(*JLW~q;3X}`J zL{P^vjGHZ8?wzu^ER?oy=AgTdRn=! z^G%vCH1C1AK8_vs$|-ZDt_MG}`9d8G$V<8}`ATOP7oIhJxjK3c*T$o(9ZLK2P&^7jwW56>j>9QcAK#0>kR0!sVZ>~7Feur~0ppHs?Bv}>KTtvTUQ2T7T*f4@sHuFnHx(-#9yVRXrvU^@P z2D{B79A31`jVZX#m_EzCBI2mxJ%kOWH8N~|I2^$umpXQiN}krao=tNGPGuc_8h@u{ zyS?y3xTZ20J7b&fqfx$~^zj5_-xeEx4K``u+PV7ufykuQGw8eZwYh@iseUX8o)l`Qf@qYEp{EgRC~67wt7{o z-dteiJv%vkj<@vhJiQOmNv*T~uTEGE;LJ#CzVbLQ`3>}dAzjH5 z%Us?l=t4(fX-3vo$nR=gpssX5`#TVrfG&LfpfqoK8}&-;w}kh_(+CsZz2Os+*wZkx z4<}Bb>CL!P!iqSXWkVN7HN|g-=~oPvMi);7qAkDqsgbr0RHw0IQ4D!E8Gx` zq+Nw`2u(cOS^0uxGd2+#R9m`PLHwCOc<0%%O2>oA3;arg?#+8b- zW|Us5=S(2$$Tp1Ij4I4%To^`KN9%UJE$=4dW-9cxhm3>)Txs0ih636Sj~NW7$?)te z#Ck+AAzpeumLl7h?%S4OP@xiZ)7=v;y+71x#Uo%lda6Q~Yy#8B!rCWwaxB_tE;IqZ z^m+T7X$2)g*?O9y)-w-LkyPjOXpCXw4Px}xxDxf)z&10;ty~uI%3m5P=_(b>wiRei zWKUy9Gg|f&*m_L8&HF7i;F`G_9{po;Tq7EMw+$=pRJ$`Ap-Yn^u84MRgno>rMLB-+ z`@B4`hkBK6w;nLTVS2^svaCA1sJX2qJzI^PGkuWSzG#=g&QlX$oI1wI+%MHx!5HO5 z*l|8lcZLJ3qB^Cmx&Nl9Gw$a5Ra)O9vg2fA?HTRbPj4G!DW`-@v#ZJEMy}MO^Rdo` zU=8>(X826L(3rt$y0Vrh(EBOsFiHwJW>ri{=iGK~J?qu!FQ-q)O?lkya)d+jn?ai$ zh>`S=F8p6mG(EJFJcf0-D{20!dsg38Uu-No-oF28dX}$=ag2$H$pN&J$RZdXzv?8| z>m!>bdp+-Ha)4DRfjcgsF<^uC$Fdi3^LT;NAtvlH$>;G9OD@NPUuFHYeuD)H}zYD8zU*M~y{;ir^vm zWQ#j{zfKj-U3rN^C)GSPK`8BweDZAx|M8m>A|`=CF)N0dr9rIrS7T-f6rU`c)#vn4 z^nE-wRhPZgxpQv?2SxQT<<3l*NwFBITpi{pp5oT8$=Ta=mFG?WNf-+T$X^3-bQTUi z_s&N0THrK|VP#lp43u{8^_rHasyj4iEP}QVn$0#Figgh9!ZMUJ@fXp4`Fets&t7tb zqvUUEc9*zbjgpch=D?@$1yeL*G^ri2Lo_WSaLn`GOWT>KkKHA39@Y0YXHfN5lB>M5 zNJn}Ss*`<YHqQ534S`|yWkYjLO}YB>s@nH)GfDPku+EW{<;TRe2tSgPNfT~J*9=J8** zF^(1bMXVsM_+;ODc{;votokoA>OG|pF?A;V2(AA3`Nw6eUQo}Od=_oKALw`01|qBS zZ~*f1%{hKCpdx@f5jN;zTvHm`-&Q5touPECCte|{w^hBy8mIT2DyTI1yOt9j$t2Wy zg-~4&Qr3^1Cnc$5vdTNYHqAFY4X{CidL{+A$ydlRyp!`uwZ)s0ZL9F`j)d}-F&G(( z#)~BosjxK#pAWmn0DB1b3VgJ?4&9-Q?lJ3>3$Z}7o$QLHs@MVQ!rS9lp->`zFs* zt0k{SQM89K=hb4@gVko=tJyR!4b~XG3~_-%_)FnDI<=dYgCFc8JpP&p&&fg{zgQw# zN@&`u=r0b+IpHU-AQ<$Ah#gs@N=0%ZtyE$KQR*CP1s7q8I>jc@Lbj;4S1r7dH_%ZZ z8R52(pp0-?45sk$Icd66&QpuTWP3VS+h*a4kW2WYr4ghqhlswG+ zXqKvu#_y}+q-npSZf$IA7n`76N|aOh%nGdl+&6t|bO&K$$ih%8w#aRPgC@}x=J?76 zp;7cyZ6&8U!}#Y_E&h$gD+pg?O=}|0#kScSx*Y_);(My*J+(#phU4{O`3B7Udlib@ zh7Pba`Od__X=F9O2`-ti-j|xeCL{7ed!SLK|KsRL;QwG~> zVHusf7wMgnHfD})yKZRV4f^M}AHMVHMQlVHf1TLbh}^i;wT?}yG3j+f`Ke0nVC+sN zqU~Jqg-A_mLcE4Yd1|d{!K>lP_PT8YurAW1v!~wdVo8@SeAs!xWawLpbpKOYXY@BY zhdBqR-9W08K$Xi(wFUxPY}_8MbK!=e`8YgBEi3%&{gVl3xm*T=AS*$K&FoD&UOAhW zBezp*tLy5N&*j>AcBi)*Z+@%GJkf=l7NgsbC`0dkSnUDH(Doh^Mr|YYK{PwY?8ij_ zq51MO-7>+0i;+u5%^+V-m6V`2o^7=HTJlMOH-ZbV)tLyjM0FO54!4Q^jO`+ecM_|^ z|1=xSAB*Vs)WRW+gFdMhc?q^g44YW-&wGWQUKnm1X{6g5wHgo2kh`G0`o7|;AM4Io zQpK}4S~^%QL$FyJG{^g5trfqS!RmvA>dATUP9gl6h%QgU+UKu=8^-&-jabYK0OW)1 zA=&q>hna$VR~f7yk4m|)GQK#Wp4tAgexGJ;^_hO(u;+bSS8q!6c$1-bnGAMCEZsHk zenrpPXneRT!3I~pHRaf$Ta!F2s_!hv28Rt<84T9HL%rMCEi*HjuRWL9xw}HjqCNXz z-`i7`blUluX4iGY?K3#CpH8tJKe{7$Z94={P})N!;%>m6vTKV()oP0RO}Agb(t;hM zoZY73?Pm4}vJuIeMP`NEO}nHlt)|WHrM|B=F}Ie7)p;=qe2c?P;Vl-AtB4if$&-fG@Nylc3#amcHk8-d&oOU-JP?<2=#)qrxLzqju|gYS$!o61nx^4)Gej3QS9W8W z;G>|`ytlQ=DwxOM@eM?kCuaP#d0P`^ko%N9lV~Q#(Dw31)RIf^^?fB9zs+DS{{igE6`o3NWEP39YrpfLmjC+o~ui#2|PStRWJ?C)Gj zLuKqq<2wqsp%eqY9&>j?cOfyG362Vz_|G0Z3q?ymmF~EwKi(p{;(T{MZM)f|W40aZ zHjl@@-(^q73opj^;#GJ&d#lhnj%c#KHVJ7Lp|e36mqqS}`gNm+-ro(f4@c)u6{1hIR;sqiUa^;4~I7nnfR-(pRQ) zVJCfOXR}*xbex~FQdAojg3QVG_S%?p-(E?PU!Qx!ISb7}5WsOV^jag;@+805QRR*V z-zwVq0;(+_KRQ87@a*174LES?K_>tYu+?kLfeZ)P^((}(V)4@{xg5RR@SSEA)TyV| zrsWYGWcw!tJ}D;Ka@j(Eik0Yhz46{&WTjc`H0y^RYZeL=k$%zB!$O5`-eflG6PsWh zVHxZ;jrdXZs)}?vfSZv!?~b=+N)Lcn7s!21naU{asu!L|>nq&3V%3(0e)IOMO^M}r zc-mZO>m;9ax?Q>C*~qJ2T1Cr6TPM20V!H||O3 z8+wlZ#IQ`8K9uHo^KxfD(Q9^TCtV9g?=1>$N%5Hd=e}C5={gxy(;>~d!)T8Sc$)n}6+Vv4 zw+^H!4mF__SB5wWOiNWuR5Bwe5oH3+BKp~hP|wW_%b4laWBakoHKEI4I+mux%R2=V zZG3UIoRt~R`QB^my^morM+V&Abs781vITggFdg}5gxdc87riGvUoc?UtwhJsWElaE zwH_5Ni{9P-^c*TQBINPuTComk0PMzTjljw!_0Xzio6(%C6PL9;F|0Iq5Yy$#2nF>y zy|m7pV}8(@$`yoNl*pa=y4OitITY(QKL^g0=SmeJYPV{U5?$dVXUl3I zN~6oPI`Sy%I+H2}tP>n(TASNt(ss&{60UOL=y$`e+_k&6KHJJ3KuPZ`T<`pjvbi(4 zqJ{G7qhX9?_q1$=A@0rO(Rd*)QO}j5W||ca2^SPcKCN~BPL# zBg&v;bVEANG(@w$v5vJl-rPX!F2E!mFst9Y5CGbqYF=a{SufzTbm#iR1u9m_kWMh@ z6nvw+&<7TA+ z|6J|x#|rJl6lE=t9)-DP53nA>#~YFSgk3Ln$!E;B+ugT78}{O(xyaK1pWM3JmF zK_DlqaGd5rFCC9T^;IO27uL3)zE>bhVM8YF1zCu?E4*~VWg8t zr1qx0>h_g>s&KE{mCBW5H+C^qvrF}e;$+h8zOh1_d5r*X`*4>()hvpxeB3|kq^Uex zNzJsQWv*4tGMo7D%rci6A!pMDFQf?N$`hz!&~O}nucek7*XsOiH2yyR6-AHOuzOZc z|NiKbT8_e7Sj#K{l|Ct>t@1Dcei6&{c(vyRPUEdMS*)Sk^|DjhDLvLH6*Tnut81-H z=iXq;4Q(y4a$oEu^v%R->0^Y|&&GYfl`^a_FH~toI?2crspdWI&N}j*wD;H%8u^T9 z<$CBjo0KsXUD9D)B*TT5jP;xTU$^%1PXuMjB#Xbvq5TAn5lB(GHcV+CaKya}<^j7m zbw#c4-T*upE&g-(7_RLMe`5tAV`{&+Wj~pc&SbVGdF;Hb z<$s3IeLJZ0(V~k{!p*XOZB8yLT4MFZD5Fp$lF;JJ9?w^`B;wIA(%dOJUrecuP@xK( ze_P|YZPB4ai;Y50QqQ^MiR}0#s!}!zn;98yJ~`5w{2eo{)_)i6h_+R`>(+)J@EQ>c zjnq2gYWbaSyh-Hb=Wcfo$vb(WZY zN`L50&@6$;pV!1!ipqKVlo?Gmenn5VJVGIZoJ@s~MzkJiU_YvauYOFShsOl?Jy? zD_G3(xMOeK&G*CO+0qoVN>XRc*MX9R&bP+0dL@A`{9m@nOBob(Er6{m_~DL4h%TjV z8G8QtoYkNQ+j8|ebbHXoll^D<3m-40Z{)3@rcQk#%b*A+EvVNhs4~4&bJpv=MMyz` zspzSbw@ULz^m3t>FFuq0yM_+HZF zt?CPe?X^{3#alvmiv5Ur!<Em+BIy#B1u0E^&@jTjZ7%=T) z&X0C-=R3E_#@~qPwylebxTcfdX6kS^Vo&@l1L=V1Vq^Op2Ps+-;7AYR!&B z2d%&*l+i$Af$AmnxLi*mB56*9s2EewSE6M_r7ehs0R+qozUU6cbLL9L{t_Iu@ZzNj%EjgQ=3m z5t&~>NLHW8aM=yaS!5Q_aMnjKdhhi`j(bhHc+!|sp`l#iFtqKc` zr~n48Ck#v9`O@LreNSuZli&BnZnipoZKopa)nuBQ8dfbzZ$A>8zu&c07{d~DRz94i zumR*q5V!?<_eW5x)TI5vN*&Rpy{ZL7a+Bx04@31U9j_NANk_b(!kZDwBDtoWb>?-n zk4HY|WZsyQI;Q10W>8>{S`FTDwiwd9kR698uI>!UuMAtb;JrX{ zG1z8Y{GAj|$cyoNDtugDb8Zwq@z(3MMQYAaouWGCtu(i6dS{G6k~U8#F5LiSim1ni zPpjdE_WS}mV79Q}kJ;}DEY23FQ`I{=WT)w@yL$CEI#4rDocBj7?WOv&X=k)^$yhee zCIzU1#y?p$FOiNbSyJ#OxSKpq&a;@!r zd@n5*6Pdoab7#LoQ&hVo2pzYDS$*jZ72?!JeH~6-Es8Pp`dHg+lT8*A(swSBD3aoV4KR^k$&E=j~Db|GNTUSQ~T>s&stojdMuePpQE5?`-(<+L-V zYCAhV3`*QeP_c77jAxFg4AHm2*EBYnK}{?L$iY_x3UP1FgG_0hVa)w@tyr?oO;Lhi zo|tVxQ%Z`ZRzeu^&ZVMp5?v(esB*Hl#I`}9U7i{BqY5-9xtLAsiMZ`X9mmB*<>Re% z7RTM_XZ0&)!cQf;>cM6!O|PY-J;$(XTsPp+y189;?fF-(1=F;h_h$SlSpA1GUia&t zW51$3*wE|rF0t61zK#nvS{`|)=g@12Hwik6m^<`uymdPmlC(Pt>#+5~yoxWr7c~65 zsJ}j3($6SF%T@MriSMH|y@Jh#hbgg(6na#l)jnM@jy(p z2aoYhzVv|XiS3sA*G3r5e=Bi9PgqatwHnt@6jda6wIpZkpw8UaJDr=Nbx9knd7U+| z&{}g~?P-}*|Cq`sYIl4pg*!|xK5uZcBwhKc*q)=rP9E(5W<8c-Ce$){I1&@e<1Na} ze4$eEqpetk(%AI`1DI~qWkf-YabkRiL62`k zICJTaC2V!*I3-6qSU3BAo_5PJhq|_}i%P?8d%If#^ot$cw(lXOEg8VfX~R-I&-kO} zzEhcQ=0qsAoma~6o*U5^54Tf`-kbHfn~N=3l<>EBLUlEN+(q`0vQR$NIp~<8VLR80 zJ_Rf&_k=MSCg0@yjJ8bfVPD$26$RT3>wd>W>5c6(FiojujD{(n*S-r;Pve;@GONomn4s%mtgHCt*6-K9oagxXcB zc5Bv5Yg89%HTIqfiJcOprL{+`#7aVKL9B>`@LulczJI^xJ>KV^_dl+~Ay=;RyuRmm ze8xiFt*z;}djvkyp@u+lz{RUkJZuOpcoe}>Zmy^NLIbm@{Ml)?F4OReUKWuLsTs4) z)&$+B9a~Uj7S&oYrH?WMI3H<#dE_fpD`s7y;ewSVBWQ>{-(xPX8flv~m>Xia+Sh-h zkXVC&)BLUGgU$a6E&(zCfY!&}xxTyqG>YF}2Iioax3_^}Dd7YqNt9{Drj^D z#D+UK=K{6fHB7NdESC8LbCZtNQ{s%1O{n?$u8g%{1RFbgla-zP?qJyV4lm@KnTt0G zwp->fQSJq(m{!L&l;Cd~=hpQ{%=)vQ2M2oLYY8(kAm}LBIGZWymj7r0Rswl-d{V$T z2<)7AiamZPbu%MbkBAJiJ%_vWuNDC2dhWu__E@6A-NFnWtH_z$ir2>{F;IH;ZVj3C zGBRZRWgNqdvFm^l3VRgR#1H^#OKXG&-?{Q2yKgQS!zn&txRjtZM{0FIlV7~u4+`>S z>4se}aY=NBEwNgOuEwfa`{Ss57*^xQkbp`4=&I86Y-yQk7S)&evNdjteMqxMydSYX z^)=J(d(Q2G`l_i*RG!Uiy7cc0Vg1iMhtXU~QL(%4$v|C5_vysP2F#EjT$*eQWXQ?` z)7{E(CC01nOu$p46fmjm_!480i8fa9d9#xHwt;Ym&@+tZX+wNK1Me=lLtts@>dKo>NtN(?H^unisM{m6U$*&s$fnD5(@3hV#oi zBy{wel+`&}K_Q0*!Hzu&GmXN&WsbLek6+e-$y}5sc})0@X9_5GqLdM%JpsV``o`CE<)L;lDa$ZpF;8@V-*BtN0-JF<-)AY zw;0K}ILOE}XBWg!<+FXk)`G^GzDs#_J#kM0{Kz8h06-*0EqJzMo~c^2Qxfo3maiVZ z*}P}1zp@VPIx1qZU%g+I%y=t*+Jp3LwHKQTSpD>Nh4lV*)@9h8|2PrJ^B3(Ti+QAS z*lLEn6Im6vR;T`IlS}QEhZ`;JbVyc8v1M-p%2AL)R9Ui-2(IV&={~xfnWq)oA6`eT zdCyBW7gfKz-7qLM1qivy+l)`R{K&zc${IIBMhg2YgjO`kFWiz|h#On15e=!~Fj6aSBX`NpI z)br1IZHM@f*ZHH8iFzY+p^c!w?m$CsdXT#Ja4d62xEK?peTKP4#v+5MoaM*2aVwS> z@;_&0rEo^(!@A5y4GMJM&=OyWB*U{E_E9;F$G+DpYkH|`8m3XzJN*Cz%AQc==<;9# z_5_(sGGx#atH+~y6%cOLsLl{Iai)f*yRWQqWvQcKCDOd}BoAB=3M7Bbi93@HEnB?d z_}g=iYPCXI!SwQ*{Rm$Ugvthk429(_pEdacI9Xm;?+Qb*col?rJ<_mxW#%ms1T<@x zY@IoY;+SYlB^!H!Z%m5mOtttxGE;zNl_-|WnY4~R!Ukt?wS>0RaBf{ycFcfkb?1PJ zAwtx^Yb9|U8^#p$Sa<{i{>+bNhTLxy*c!S4)N}rZubju~4dgu5%fgJDLfOb*tRT*> z=lAkSuE68FL*_;qhLsa!m;O)k6DYjT_r{%*CQtoBXuEMb$GWJ0PEK(=uAIY8B!Qr$ z@@M{mt5$}xG4TES_+8(t)*egy^e5b_xE~bZNM1SWW{fDgCwb(Dmz}N@MLo5Th-B~C zneyWKCX*G44|ruYCNFB(S@uAs+n3zRDd9R!^&NiwG3Da!k1_dhC<2LST_!uPnTZwe z)bG%}t&ayy)(KM}Ttt6M<9$T&NVm^ZtP%0-y7rhl&OzR5uaW!5^TtD2@r>hd#yCCa z+3NOEk8CBeh&d9V2v} z)w;+%6gc8lz7-fE@O_^_xxraKzeAQ(V23{L2;a(HOg`jPB*f7#7GSDW@!tRHEm%6G z0Rs|a1WEnf3w?V~qL#Hr68~3;%5hMl+NRBm{#&9R0xlp$cIQ1>I`~RwwOR&|(7cR) z5D`L?<*QkQgt1Rft#BnlVV!^~wO^$J}pJt=w#a69ww*vF^vUN7mOF z(*1yZ^D(2(Qgd*^EqVA^eV+CT+ph8v%G6>^Y>vvu;mG;qeHfEj?SH+QE`8khSj?Ru z)+Vw}wEnq0_s9t!=&oQm-YrDVt)^r@I7}ubU%Od#efu(}IbwE*pXR-ZgdrU|FvM|RlG20I6vK-$5G^IZ;53|8QCul;4q0dw5&l+(LqjFvY> zase+$YK!qcR@NxvztkmezxJ5K9a)@=ZVY>nG>bv7JYGkzJlT9+P38~J%n}+O&JvpV zJ?An~5}P0;v=#XM!nr8n9xwCC4C`9o?yd>s*nVb;4})Eyxl!K8HDa#PzEcXQz{`7Q z5!kvrMq#U>ANdk%q|cAJnMK$*JJ6QZe2YCa+GNwms>>7~ZZFn*r2o&EQ8Al3G>Fz! zw*`Pwr6FYgCTmFL2-$uzF0mqL2JQTAchI?DSHss5@$yNqd-d{oXGuUH_9?Z*WG=8y zzx4cKs!WF&X32z^)xX9q-i^NWl#|?bAeFv78_1m@tQ3@Gi8no`>KVE;Tx*`jsi{J; zO&?W&H63GD<~&V4H@7napSVdLu1QX*Ci18^3b>igZEh`+y>KHj^Fc&&PTALu__R4+ zjW>3|1uR zCg3j44zVpRnie}2J+Agr@mzdoWb)RsLH9+hxzM$FRUChd<<1iu`S`jGWf$GF8r)t4 zHY4$eGtRR?xp@E=;(_X65jc*8S!O^g~65$h|z)=MqYGzg%(wtWLTJ0w5HsY)Zk&;8qB0f-Mz=*G!F1^{z zbY@80g3~VP6dL-JQ{c6Lp(6;fbTV7kAbV+v-;$Q+l?FX`MlrY1XFNtgRtwf^!TCta zgi@)CCOQ5CcFo(f$ZkATpsnv;QAGZsLskavSK}UUxbVPDwEB~uObwh{%S~2FeN4_q z#^78PQM2zvQ)E%{bZ50+lnowT*+B+J2;ZG^?7Ki+Z9Re(-uni#%|XTd&Da(xJE%0a zwtuVr7m_l3z>)C%%-!&hbmd#+13^Y9R%^u4Sby2`Bl7S$l+~B>mq|Bg650*m{7Akc z&%>3(XOCvJ4^k2e@}|88}EopOH_I|Vw-r=yZRdm3^3r}=Y)h^x2$?rCKgf;aQ6jrDohJE~a~Tq@UWV$9Uii7ePdo-x9^8=6rfYw0OK{02xxc|=?bPe> zKrPT|^x7+y?bU-pzJU|6nwxWZ9PN}*40vl<^D3XLWo#!5)t280>pF>Ja9Q8p&OV}% zaB`tJ(&W5?R|J}jFhylHg2>e{03!3zT{G>F0T7MHYd3O^d<}5$ZkUhc@_tc$udfL} zp|B7fT5vm;*=dWa0Gw<6HlnBQDh0O(cD>MaOByLQGWtR+k*O&c zNu!KtSJ->)3(iUZ4q5$evmpPQX9brH5dAA7)R-+%`^jz$jij5tMf4J%Y zvfw88V+8cYxmSvGk>H$nv$(G)sNhn_8>DzPTD~u?^@n`KG|sY-`dH!8)o! zsh{zW={xAKc4JJ&q{ZBhk#OGKasaOo|K~m%f(7*GkwI(Mjz-;5O|4UE&+J|$Q3+Q< z`ZWl8itRD)uYCyhverMRGW@wO%f=Sy3^EM>9kcug)v)E7AgYN;+2NVkZq|M0#79<+ z6=PaHI=#i(K*~-GJwdR>FRx;w_1uFv_oqqcko~cpH-|f$eZ#+X4?>x!9uN<^{0Cyf%j_!zlqJE!aL|XN;A7NkL538HQLVeS^8Ryolyvb8Y{$O=hp9~E;|_d z@hBq$)LxF;4IUc#MuBg$$&!zh?5tTGos?^$&ZfKk z<)A&+cHI$ra~;};D|q?g{jT7}af+$$yWb7RCe!EqgmzGJt|d~`9PI7e(D+cR%@Rtf zxS%2+9>YSaZJII%Y$v0aESIO#7SToGerp&$e+4^FR6WUstHQ;tg+Jvnm!jmP@~0Y6lJ5n zF?5xEw8zj|a3zJ>Gxx*F*N9nEXHgsSt5b-k2gF~Gz!2DqS3~bRL<|EtrnL2=Q&9@` zM6xxBIxG~9H3T|+vyT`F$cgU8_?`T|1hJZUM;oWs`v>UY%8|_5+ec5FdCxKO&~I&* z_l>yFYYSDA<#UsuLg9NrxuC4#w~#ir@W&7s_q@JmvmMA2if4Wk0D3C!JILVa`V@uj z) z6h7bayZf${pxXC+a-HV%K(ZIrM$X~qU)ss)wbugoUi4+BT3|daGf+?Uog1j{=2l*k zjky{vMp~yER22wY0FV~z&J>pOd56e!kD@;oI4kBTNR+DE+RrvDkrH`Ki7{3#8=wxUWrN6c698kLvY(snXZw_=#2ON-=>Sj zg=6cgi3#^USa6bSA*&V7BYY$ip#2vd8#GHt+NC{$Nwe*VhAHq!sm;4efOAgVVfx0X zv>8Xnk6_D-@Y}DgZ_|qh=Gy8=dvO?^a9R-9jL$w8sS1cgC==n_eF!v*jyhS5G>{uf zgBe}3gwC#(%+8 zW2txVzTT;i`8bmUaH&T=dQh~vh+}hQD?9y(r1n3elH-h*Xbt=({gV0Kxc~i>7|j>( z?&UP=bw{wo0R(G-lz!vK1^~@QcqpGBeME}{p^_ik%~)}Rji69;hv&M* z`})y9bg^oYm9PJb4Q1bL*niQ6?ld6U9h=eF zC(%IZa($hGdLKP3obLn2P0@K#kE-j0YiEAf^&260Kj<7LayL{cpXRyWzRX5S51|WrY>-a(fqimP9(eD>czJ1su zcy)YM^5#m(+)0N(_NEVyA_(XX?6FMbW~uEse@dqtHia` zZh6u}-^FBkY$fJ(JCs;^)k9Y0T5E5edskIYVKbd7F_RD?&5i+Se;Ib^68QwO^ZSY1+9YJMF5}h7@dIOwfxygN^!f|c&S3%8Uq20W=K~+(#Iq}6xozoy& zQ{m5~r}*!GL!#c$0A2%)Sf1_pUr3Y_Km!gM<(CZj_>cC4(*b$Tj?0R3CJC3DXOPU& zU9{)3>an0O=VQtkg)PD&ks^O~NV+gtw6x-wE#N9X<)en%WFkuoLhGxSg1TcxBQ5F| z*`Dl%tjf~uDZLI}L1C-8W-(tn?1ki7!y_WK39qQ>-D4Z~H;9)it#^M~`5K33ZgdnH zD4W<-kJHUTEQfp)`~}N68hNEA@t5tkX2IMRr}v)yzPNC+;S+&#PIqnh*K|Bmei_1z z*Nq#()`sE%;x%EtUmP{OV=Dh7Y}63nAlVsWlGC4!;+3J&!h)#n>kmYJ#xujk24f?TqpiRxB6yFb%$v+s77QrjWXLIrSR@ap(;)N zV(=o?1{LCLapc4)PKC~%nqCW4No(e5|EbX6Hz_wLE>w*t9$YH^lL;l;KKWji$+A!# z6(_6O@f*Z;D6BX5z>$zve>yZsSI< zT6=!qh)TxfYfXj%nz6p^DGIk9M5nvzCyoPr^v%mPBK0#~04IPxg<4+k?Xg^+%N)}i ze(Et`v$A&i_@i65-Yh{9zdb~J91l3Jx~g!^>Sn$Nn?CqC}h!xl3QauNEB6W=Iw{v+GK_y;72OB%XiDXDXCuT;}Oj zEa0lHX@&`ifJl$=hl5=+)ib`A_p`leJ?jmQ7OD4P0OFc$!ca=W@|a&6l7N)x3K&cG zOMqFSG(47h$q1gj?o3(KX9z`}EcWq4ZlqSmc&*U{5n^c_`GeB1!@;kxCx~WSGoJby zhGLm~h#-?N`&mHC+tg}t{6}378edlSm`9ZcJ4|1{7Z@5k7VUhooM*DM1E>iS6O{t$ z&sCdN*1u?Hqc4*63Q!b1Z^HChlc(bc&^f=j#xK)*jp&w}iaO-tqdke+BGlA_RR|5F)(c&kYy@$9-9N}@Mk7-%hW5-%6ZzAOGKK!)dD z%-$6)LQyLaO*;XuXeIm9!AQ7*rcyOBGb2#|n)ztGS%wZZq zUFs zn2$fd_=krNkKiJWu+PkcY@Cp{bME;y}OD8480x+h=O>%|t?F5RdjvN`?%ogR*o( zZmgd;+M{vhK3B-Qx*Id@oJs{xN!OO0=VDC+cy2PayBz!LzWh8oxo?oTi5V}-(=7D6 zcFp`|0A#;xm2AMI?sqwDDl4HT>lXGLLRXD5r_{u7d-Z>#IbQ>^+OH{s0s;SVjS>0c z3sB?>)5~)T|A}{BT7Y611BgLM2c_y#Hj(^8WC`gTo1K&E4jVQOD(aUkyss+XiTF-E zYsl63a^;TP+)Ue>a*of?f;+7zlK%*Vdzj|edY20FUl6Ev{p>p4cx|-!xF+b9(6#;d z4Md<+aEy>GA`k9mVX0dVD~KGf9F3 zlQDYa;GN;_H+a|D8tW6^?BgHgZMmbGE{{f(?FLHR^w29mHi_S(C)QVc z=2Rn?qt(8APNw-04NJh7aIwMS@UA6Ri{!s)?eAmk7l=(w?gQJtuq;iiLXH|#Y&({x zF6P0TgwV!X^~-Dn`)VQ35S z+cT<8Y>6QxJeO03^7P0KTLCUDY<4Y5tk@|{o^P2;Y~wLPL~AzQHV*}RtW|A*9^l|D zW?FeP>95x>SM3)@9Lq*liaY3lHnjaJt4U_OJZt7~$WL)r59|7rKH%&8nOR@{D%`#3 zF6(2=!=QXAE4%Lh1Tu{?Ec7kSHye87K%_VGTL|~>D_nar?p5^LU)S}Ju4CZk$juMD zOH8|-!}Jg`_|Jv?-@mW>7e^v8E!8w7IYBQoH@t83wDR;1v4k|(R7&kWPj+fH6!oyQ zfzG#~yP}Tv{C0l6bJo_;!?o;A=@U&Q+^co~tJ{-p9TwWthLP8Z4P%K(8KUT)chgYo z30Cj6u0JQ$z~E3wWnlZd$G=Ymd4?;9-T%b<^${mgRtHx4^`jr*BFJiG370lALB72m zVJ#Dn!j6=R&V&vOk_goV)aX)2kXT_{ z>Atb&Y?lmoD@Tm)XDgeC(N020>gEbPZlv`*xM5=98b9uaAL7P}kp1U$N+Pkr>Z_-{ zWnzAOUH8wB-j*TG;O-2GCUd1gWvuLGV~^ACRNJ8eHT_zMt)DyXX6BjO+tUoI4R25e zPR}&JVAfsLMdBGaI6tWh+Q}0VeN75Jwt6;v|CM9e=Jv#Um0u=swt$tZklF*83! zWs60QQ8!-Ar6*0{(dk2DrO)VY5l5y!uvB#JAVTvPf^X%@UF?eR&Yc5@%5^5}NT*i$ z;l8jg81jA_59EYuD%%@J_O=%(PHMi`C5;fhj3AN7`jHi0u-+%g2WH|sdOp(@Syn$} z9sH*7(J1lp>AiY*kJUqn6F$C@QEi$XvZN_=skCK9{*(6UaJ0L_d+BMS^QOiKRPJZ# zdCX8|l9gf3r68?&AOCM?Nf8h{__Y7-UfK;GaUwbW;$|L4lbYJAUEQKh`uzLdz|9oX zhF2D*JIv*!yFgBLo3AO4_ytt3d?MXFHpOr5Wd^a!9L)Oi_$2pE^57{kz=|rzy{Rt~ z0y-5^@a1Mv+E|wwbm@swRTV#)RKnPdiZehak)dJY_kl4{#&FBeTH%eq!wvC(2d35< zVfEiBOP(|^RB>{tpO^k;%93XTTv!2EUJ2}Vy@|UcmKPsd+7rMLuQ#iF@SJjXeb?nt z(4PRlZY=HXWV0iOPAlhlvq^w32;}b}*q)*Cy9-6yi%P?K|LAH}VzrjCev(WD%wx_0 zVyZOj$3y)zAwOoQG0MfXI>I}6uz?K&V9Z^kYWm9^0viy?0lj2N?D9Urtjb#F?A42j zk2IfsjmcEINZLma5qZZxAKkc2P6D+#Kr)MPaDep9ra zerrHWwYN;KCvLy%gZI$_;E{t^+4XZCY)zz+SKaHJonH!BMuZR{=F;)FdRtj@d12oW z>yEP$w$hIVMd~*ejt33~PDjL8*U+jR(v$?}eSkW*ii4s>&zb2)GAjE#N98>yTLN*a zR#=U!jE|2cpsy+yQO#pB*xdbMt~-i5PV52Aw0-x)viH|3W2m{#BNB2s_wAH>=tqw? z!~k@CB??QI@ zp__FQl;4LuhRM{fTCcf$N>;3x_tMEbnB}wfw$uL4#!AzXM4Z zFJGB2}@Ccr|2<{$HWatY?d^4T#np|1ti4HBa=*A>NlGr z7vCMXbOIr-6^E@df`X4n1NvP<)y3k+PA(A)BVNnb`b^DP?|(_(Uy1wveF@Oewyc!a zxQ&|<*YC)hoa)-ElklUg$Z8kIAt`NPkugcDs1RZDHRm#g8^Lm?ey!V5>rPbw%(TkL z!c^TjIabt>J^;9XV=iEIFAyNWopA^4yc+9W?^~if5R#NwwBGI_pXHC~3-N9`?oWRh zFsQS;G2%B-roK1d++TN}yRr~Dz8_adzhu#{NO^NNGDbVn(Y&pB2>z}5Y^F$1V&n=; zMsTc7u#m4rVpC55AI|`5@Va5BgC_{sx50t;$5xzvy}cW0|1(M3!EkP?G~bzrmipj1 zLsxaI!u@dn&}u1d9*o_U8E_17@Fnkg^40Dj*?R7F=}B3`X<@zyD{A%&+v%E(sEEDh z2c5<7w`ZBE0Z*YPx_C{O&Gyq(j6s$*F#+fAAuUv9mTw?ac|+*SV1Hp+v&vKib9 z;_DKS2{%I_XyjrmLhH=J18_3^n}#isD;&H;bG5)198w2#y?05n>wP8uEv*BAV@`rp zwO<#`Hpe6YBSk|}7cFHEbLC<5(iyH9O6)nAYom+Jg?I8su9{5Lj!PdByG}lt{ZQ>z zDG;yyjw3Stub?+w&907S*RgGWkhk&lOD2I6=EgfXk%r0A3Nb+wP}+$sGAb?%pAv8D z5a?>-kA7>&e(eBn=BgePb3s!;j3K|iMiw2yM%uk3cf2FXw1M|66_wu9KgPZpQv|c? z@28$>B7dQ=AoP<51{;+M$07YZ6E!{LDr>o-&*O02N~%fga)H)W+wD7&IRXwcWP z9s`Pr@phm3HFpjt|1F%Z1py*{aYt|ae}$8R?g1KAF(1&NS#ZLMY9CQtV;QW!NbK-vUlFJZJ%fp&;7io!gt?qx?k}vm9hUss4s`20` z6Rc#^u1}(_mD(@jo>oMPTv7X9)c0F$*8PD%)N+W*6++4GrdJ!<|3XPdy@iTF7a*v} z!Q$<3w;abDKt(rkf$BRKs4q+YxLYTO_3}JES-mk)aAdjr_K(}^OBlg>q_k&xF_02z zroh&dPssA8`!ppvU;670X65^t&1!i&g`m>rE|)p8-gfyRvoZ2zyIHdknO5`O0Eg?Mo z5^~N(!@_iN5NR*bOi_E;wrP)Id$s5^oQ2OHmPRK%F6L<&3J#&>&vgY@W^a?D%$N2S zY}1%v*Xsu8DJRc&q1^p3Nq5k~P*v-iz3obsMC9t;7q;ddx_z9>#fx$-J$L8}j`-Jl z7u0kiA-<4~@e1n0WRPR5$^(-CcVb@SiO|f)ht_-q8WX!oZt`QB+cnh3KS%>UqJSRz zhM#6z;f95h^MZiTftsr5N1ntlzaK*u_E;n2L{UszDQSgtD~!nRhiX7EXV79YM>e}p-ILqW(nE@4sBad-_cHw zD4|QWUYJPYbM_|lx1PaMO_G+G@$2xoZyPP|r&*g06dbI`N?n_#r16b3wq`zQgP(gI zsYx2~TleAo2oi8chM4{OqSTst<2MBp-CJYCvraKm^0wn)ZidGUMX@p*4uPTsZaiKu zrXYa9MW9wB(;WQ*(2|FpWhUd4I$g)>hi>-4{cZhoeD&l}9#_xb93SFr$jN1ECK;CJ z)ZJ(S?~SaXx^`U~EgFKl>%s3hm)tb{FC%-Cl#T)!B1iyV?IaEcV2QcIZ)}hPKcqSX z6Z7}<=^MSv>+OZ?T~Q(dsY(aAWJ68Rkg93pqU6Teq{*;Aoyj{Z7mb1&){--xBw}fB zwh7#(XV`B4ekK9 zeR4=G{9Ou3CN5=#D2L4iSoU&qbaJX$DrqguBsO~hjf34y*ZP9kjhOb2ZhT#hdWh)+ z#h6*c#7X&15(}zr`j@!<&JTpcSbbyuv@uT&e!v}aAHy`?(fi9f(mfLXefx<0k(ob< zFX<1Cp-FrFfEhjHdS%y>Ryv9GVJnZw*1M0ZJ*n*O?POUGkoBMi8rg$=Ho%`;t+5UT z14aH`(+sPV#&cAfiEqI9qNdfDqQ%MNgb^ClI$?4nTEMZPTOe#zrzRt@JTr5Uz#T0A zOEPM`bNPN{TjFn4QwG!2f3D@!X`i8iBkx*QH~;cljS|)dpg($wMvxIOd~qq6DOc31 zf1(f(1%qZ;#SHYcEtG(m8vt*`rfv6uI2|6(z_(mLr3 zerMRZs5$&d#YP*%tGVqb@_zm!`qCnne!tOhN_BO>Hs|hS01%C+p3gPn8Q&(oVmcNx zEl8q_f{)@x5g~%k=)IB41tPos!aeOt5#rw5>u>6uW@a87o93ADUcL9ei*)vi4%9nX z^Clughf#Iab!$NL40kxbRn$S(A;iZ_!-CsRif>a|T9obnzMmvP@o#6cTZ@3u9T*hY z^Iyrm2{iyW9PB7nd*+|<{_25jZeQ4H2tU&fk~y=;=b60Jrxvmo&bC{}g0*FDBD2k9 zHmd8e&3!E4D3W|3tRAez^xRCDf97wz-?o@sD3kHkv7n~97xqP@UZZV<4|BTp^9ppu zwf4#Xsh_o$IjKH|2Y#GAu1 zX&LAo(|Hjl>S!#O@cCpJStJA&Q`eC_5V9C~bd(F|NCR|}yB zYGLuEmd}HSD03&`Nw9t2#nOx#bZ_#0gsZ_JXBW?{>=%7ap40CUW-ktQ zRxS6yLCkNB$^4RinN%U|vs1UYLEk{$in0oso|NccN#+f}Wuz&I4@(Nl`w>j70*ai6 zjL(ggw@9w9$sAH0JRnrVaz=$j#y5zqR!>^en0J=+DSAvV>E8TO<{uo8g0lg_^{yn-ZYi( z>fJvU!wU;}o{p(EtN?O5K}z#(qpQqoQ{x#|o>_(wk(!jLho|~0-3?uDZ;yEKb$q2< z_W0S)!m)@qe3bYSvVPFq)rf>#Gw&s9TaS5D6OMv@Z6@Co@Yn_NR(xvcpy7q%F&8{C z%qGW4Dx|vmwh&VikC`7ZsSe4n8+lP&w{W#U+hmHe|G*C|e(N>EJvw0b4UwQ%L^Kn| zJ9Y^G)Em?XdPWp^)EkolP+XaPwyk#CMIYQrzXBQ*2uOG0Z>Y6DL%)3M10%(NXOimA zz0u$Rvb-lusxZA=oa^qX(^d7dvhIXqx#VDJjCE;K1>NX{V!^cnk)Wgiv&VBESoE&q zRE)w)I{;BpwYOS0a_VyECw zh5e1D(;f~APwa+0~d3Stk>0fuPnvxtEnLW zDhCb|>ai$89tt1gBMmow0WoGI5&?`|YPN&k&kZ#rl#-|-Yi81oayX8>vGpR2G+vfF zm;ATHTRkvis;7NE2KtY65SJDPh-(8Y!+l==#5FBj;6g6TB$X@L9b_ORVc0IzO~XM& zKJBA_TH)k-R{aORmlHGeAZD3}P^?PDw}I10xN2V$53>)DA;OyYEI6!xvL0aYvFr*a$lp$QsS&M z#W();;1OH45v&y!rtSkK6k*0XF)@_bF*cI3Y@j)pmEV#>RBM%xXkv%@3u{}z*bCN^ zbfYI_#3%m#AT6IGsVBh3p-Mm?ZFxz(!mtBzqoSbb_Uned=JW?xX@@o>zFsb3^}Hazr477>WKqj=|l=WltYs3(H7Rs?pB_YkVi=lpi}zAMj5GA zx4DR1VkTTf_!&#`Bz%2Ienshpa;v_{X{&nI#NA@JTBa($o#MjF7sYYEfW|4yOziog zG|bpWYqTJY<`5#$8eIp!xOl3r&T9wPJwvj@=(RyiEQQ7?3Oh;P9PStmZK$v}Cb;eI z3`qwSHTOlGSHH+LVM21Pl7x~4B|tAxRZb&D0aQWfE$B={v?mF~)4dA? z7#9FsH-+9l^-ybTX{4)uqm>=Cr8tM@89_jzb}=2b;6Zx$)@3aSc}4AYf|G;RK0fc> zTAM%)AZINr58+v`l(2ImSY-JPa&({96op6x?qx6nnjPpQFm^a`mwF?Do$m{VQg=7Q z-H2rL6;J*nA9YUc3C9KXOMo)B0T0p~Bbaj}=k9}gdtB6c9GHzqYN93>mv(HOfXlp6 zu!g_Aoe8CPmu50RG_5F!{<0s2am_cx*7#-7@rVeo#e%m4iMDh#KmBV~o4hAjS-?<~mL zpXlc2w)2rL4VqLQ|!~HQz z-wmO?GZf%?C)q&M?pCJI7UP@4W|WvU{3RV=GuTrYeSsdp$W9|3*8|KDZ#Ss+=lzJs zId8x<`6Tf{m&g3}=6t#D{J2BU9~#gz^4Vf@-)VKU>BbA#|4Q+3KeQt?ayZW{p9PY* zE#|Hy{&ac4W$+>cyGO^bMDDrEK*!wHe@?|{9itAi&YcC~F?FhPueSFVWG1w3(sTQ`X!mc$-0O*|U|3$@l>{JcyZ%6rh+5wn=;3 zjH$03b!&afiwwWl7Jk3gep!~U+rPxp=S^*=Aw9?PQ*|(T>lr<$Q^oM&7~DNwGcwai zCKHl@GQah4+H0S(ZB4WI3M&wji$n?dq)r>y_|Mrv&?tWsUmwF;(@>rH;2uPIk(aGh z8f2~XFONYu(HO&tUCTTHn5tPlPQkx~3YY9tLyNmgX#~X)9?zQSotHz`=(-V@QgT08 z7^{R?mzMHjsU>oAs}e3LK&ZZ;Up{UD!er)+?c_WCSrt1wQ_y|Oucy7DlNV3z`-zk;Pd%r4IHLal4f`u zWO$FVsMyV5RMiXX2+z8+#!<1yK55QmN0+qpLku2Zq#0U!+2bkQYoCYz2JKt%p8I*s+FjCM7NP+rIVsrx^Dd{Ll}k7)@b{#XAxPw$|Q|G)e@^KQT0 zJbwUUaD^m~LJoPfgFB}QD3X^zgPg=;rj0mW)zSyE@{Qr1SWN8^_T9VHan>^etGGOT zn)1ArI?-cp3_?;d2@#m!bm=A`GhF&(r>BqZdBiX!h96jH??%Ziq{VO;y zwj$0toi<&|4L8_=o;kn5W?!;^>E{qZ(QQlX8=T{ORClWHC(h&It%4 zys=hOaYvk|26u=tgL+6Tx&EsAiYarQa~4O;nvQ^E@^7VF&&!R`&_i zCfKq@nDMX$nI_WOBjlx~FFW0K=Blk()Ypo|;KiWDP+~#mnXwixv^IBs3~bJ) zSLaz$YYZ+y?Ote`RC|E(OA84e?t-^pPr)pJjOS#}48mk)e5BIg+vAHdhjhShi;u=y z!D$I4rWIONJSm?%e!T*_nUb5DnjjG~@@D`x#WD72Kg4U-Wo|09itTG;xJ5jGhWO>O zU(_l-=i!Ssq{%zx>i7g_rWCRKd3CXnTr+qoxaVPS7gHQidkOfuVyv9{xI#aTb zA{adN{m0#hUG|F--KLfIunv-4=8YDn{<*E6Zp2-IWR`Hx-=YIRW>FX1U+oF)AVAGE zJ^{rD5(mQ{buCg(Gc&Y5_Szb3$vT@ew7FoYZ-?K!9jSOgNOSEBNJGJ|hd3C5NKCAM z+@HLNVy<+g{yQy29)oTavK>}1bvs+ksN^nljruK_teH>ba|~@BjK}Tv z=jDfR3M!wzq*CD^XX8yZ`G(h>gby;Z*SF=@T6rkoR>OP28V&PYWdwatQ z*>RclDJ6{!{|jL4)hd*i<#Ww=RV`uoo}+(s2|pG3K=egFxJ)@qqxb%x9_z^AKQzHF zS=`FL`fW+`E^Lp7cBkasqYbbBE$;s14v>W!{%-?t*6jm>?@S>pnScGe>$OE1j!zl} zob=xQ=9+Due`_W546;d2EqdjPA)h|UTgmhIYy#XKF}%E6Mt)d3Q6%Nz))s#mCAFwX zGD`p(u?8X8!dh?g^cX(v;H`&L9JNcleEXptjP0TMTSij!2dCgck=<3<2T7LI?Sx;F zAF-L(3(CQ_qSI8)8*=yI__`pECBA;s>9J71ynkWj^zwty7c8$QKc?Q;>YR6kllZz! zA2-D4C$E(^Eq0Qgo&j9Zv|sH=qjf$6{h9$VR3pKz=zA7^kYukF?y-7~>t7|F8g!hM zvJG|J(WjU|=o~G+@5gG!N4z=-za-$Px|gPDVbG3Wl6-`k)|#wS6^@%e+PN84*>iu?BV>jv}T6K|g$ zE70G6y0>kMX#fL#(5_uz7PM6dyx$$)8yn6TJsHzT;{UsLQF^OrOY6*-S^*P0$KD?M z`x@%ZxqE1oO9;9tXrLr42d0Sj?&wM`4l6;JTimKB8u=BWsg5 zx-4+#=p+H;?)*Wde)2La)R1zCpqWsu{bNVQC0bhj zf9tIPBC2ACxL+#+h~&^;HEOqC=icSgs8KpG#-)|?{U8bt_f2_D&)=v;?xCLcy&LPR@7@&<_mzm z>zt2m9vJuOv-$V)M5F=aB=}SAgUo+6p@`nWh4|*#@&VAIFRYaU2Zlk8PsT1o{a--d zTQkSbC!>8DTW#aLX68V%I>cKhT2LlQz;0=Mt=+%Qo3u3t;=M{4Z<$138^4~ABNvYU{sm>wot<*Q(GV>r+iGllTDNcz0Ceh0DHXoJ}5dcf@BQ zHGK)=lB-h>V!oW@x40i)e-WUE(5xz)sb<>U*F!!Z>HUvAmIyzv?Zl5r=>4PO_t)VJ z;6k8*jUezAWdT2?q7U7Et#RuAeulz_tN}yRCl)o^UbQ}H0b}!X#6E0TTf~EabRf1bpY*n3K-x7>0Ad)X8yk)^ajPg4Gi$S7Yr&q z>CXY*Uze(ET`jJ?T{?T5f(vguaF|y4p21v_Qn%+NCsGJN}$px^SSLhZ}R92GHS@JRH(_xbYx@f(yd%<``Rj@Dryo7H}! z#SIvBrUJ{`{p}3kM2S{1`9J-V-iRM8p_0@1=nGVt`$ufTH`&*7*%hE6i8*oKX+>xVoO^{yIW8eJf*k zolNn?uGCeAs^iPU)CpDP!EL`MFWU_*R-efWUBD%PBW9_CXd1achdzLFKEci&PiNPF zm1?>JCh|Zu2bAh`opB`>!RmWSgF)ZkHJ}s*$W)Dr84M-gHqv)pt2@$uU5zq|&!!by zqmU=h4};nGB8c49rlL)w%+L8_lYn;ZqrLkrJT!8d>&#chIdn2V48z zr{I6zp0}x7S_h-b3gnq=N!9Ro@3aTX^|~P2bL-d-h{?ymTPbn^z%&YU0lg;66IbuP zJL!;0O$lGT`G2n6Ux$mYP9Sb?|3B=#Wmr|+)<3MYNQojHA|c%%Afbq~C@HlODanl> z-BJQlBDqQFZjeTh?v`$l+;qo(qN3jC+|PZ^b6xL;_rvqWYq9rst+~b+bL4N1K^Hs` zT77z|rsuHj-F-5ffc+7<8I%s-qX#+n`fGN|8D;A%&QA84qlU;~^cyk7Vj6o>V&tvi zdDjzxNWZ^rA9s8(xSLuoo@1F%dOx+$?d8`T2@DK#2hC_JE7Sce5F~C?fMssz0kPI$EXXIE)$V4CL^kk zFvtyW+V9$lgl0X{H8Hm*<$fLb`sLt2W^Xme%9wSWL#d7Th+u~Y_Bf1QcEEPDR5E6H zyTVwG2j|=Rp{&r!^Z>Sz7-_{I8E}AAGTN4LEw#TG(TkJ!ZXdkim0SFb`eL}v?w4v; z0@9zinS1TmcHleT4*C*OU^|>ACEH!VcF5b~3cDCIdcmyk8%mfFp$beV-}<86llzLJ zdkRYGe`?r{-W)cl7_A-*pj$!?`)@f726_~@25)USIoeCjwS>$9_bUHYZwb4cqSj#V zOhA7qcooXBaB=luFroBlk>Y?qm_{5A&rryGHg$Wgx08ur;-e^s?SaQalJIAQd=-x> z(9hS)pz2lKA-ipl+HM*yk-L!PQX!||u)06FkQTy$h{D`!(r-eE*?Tjxf+6)%p^?QY z)k2C(_omR}mJyUgGWt1h7LhwDGz5^wGj275^UAYhdxC+y7Wm=X+}WB8Ns=KS0t0cb zD29yN?=CSdR}VdZN0VXHFAvb-`(wFwM$`QBHkDQKjk}xMy)ub9-qV}Q(CSu&{c5iz zVq2n13DOk*<3)-EmRS61VWH;9g>=rSCaG!G^Ic*9HdgCS?lP_wowT`bD|JCx=S^p4 zp)Dzr4BU#HNY2T^KW1xPsLFHaGL31n_5Qvm=8|_ozWrF4-kcnrW>u8qyO?g*j|-Zu>%b9mskE}v3Kd!1I6>;7@7E1VV2#`*LBU#7^C)c*|VxA}WO zxiyYT$Ki*voIIc(w@h(7?dp1h7Z=MknC;PYJzy{t9j|3%1hP6VrIZC>_-W{e030oZ zL3z&KXD`B1hKT2b1hkwnEBZ}GR1k_^47aNzscJ0SiH1?I8W>DIF+$ugt&O)m*uT4+ zD!JLKC~9kDmu!N1@M`2tdfj;~gQ6*ExeEN$-+C*yFl_Trr-TaDV`Yn!M2*)d6 zT3zfDwq3D{;bpjVfrt0^(YnuAxphvTA1v2s0$-EYp9Fxko57)E2;~Dxs$7Nigp8!$Xy z!*E*0RO;WbGEr-V)p2%}wnxJRI{4kz%TG^6vfr7m zIt&V6NA4c(If|4CP7JAmjf_6&POw{wq+^|7Ax2j(D$_- z=%PmYYZb`sFPqHdo6iKurYxt%r z6k8>2VHl@`Uszz8J?bP`=Tgmc8+b5SwL)9#96d`@1Jm#YT0#z7_v(leJTQ^&Px=3O zTj1I+-a;eKe$xfICxm7R8FPqitTtH6eYAPC0f1Bg_QV~3s4?n-n<2(*P9ARkJgRAWfkgnT!^R1ZMm%RL60%Pg3``gRg} zh|>LiX|Mq{l%{bm?9atc`R*n6_z2qp5MbV2p!e^A;o)gzr_Ym6bxxy(HU~z6bq>8a z>Qyw3w--aW9c?~aRn^-9+Cg`8zhu_8Udk_ai@<~@*IBdltjazr zgBl!wg54Htr7aK)E@v30YYlanvWhA)+y2Iu1}P|3iQ>@@RmhSXYE!$vpWSKw{?yT4 z?VP|uFortMfY(=RbB^pPE7s+B>|;OaIi2akW?Moyvd6yu@vJDHK3h_pdXYQd(8zLZkIb2!jIXmIHd0~^*_>+~Y6YqwE7#5#Z_4>( z729CoW03*0HF-xa^Vw65Sa6;SfT6_Q9?8OHXsZMuJ~v+}hL!MV=+%A*INgIwFI`E) zzGqf_Q1WWQYW~-@{{#$LBlQ=n3pt|jMv?;qa>}z2*AD|?wuOp4JUyF1rem&1)-KXA zB0oL7E`lS{*6F)VqnyO36-$L#F?+yFfiEc=vX(e2kp<3!EPPp@ky?C^^krhA^R&_F zvX)7QCBa_>>pDyFS&lWCJvxaOv43(#QDbc^llj$5od?XP*adS_p~DAYqp*^ zqv*6WrWHZKWIcLxqNQj?&h;X^&^B4Iuf01|o*ylTzpq$BZ0N9#+cw@ve{pcZ@1>)3 z6S`wv%^{TgYsi58&NrH|@=9Z#Yic;B3t$Au&8*=_N9V&=zO$XN6M`L)D*FfcJ7}OPRvESh+a+xnK84LgbgCTtP?!Xd;1LEt``7 z?RY+noOB?x+R<(McAkrzTKiuc{j-4BwIM8m<{TKGh0eZg$t}e2P3NS4K4&?h;tuh5 zJGFJE>)wS%_^awnj1#uIS{p0me4i|1>jM&UbY2J$lGe_Jp;oAjgV?wu9PxTBq;y4! zPBA^PQl^t}-yvw5ob^kl+Ry0JbVPv3Mg+%c)V*ZRZIK+rdA7h6sz1+Amt?PvmCJ@Q zs%Z^au5L!nHU)_&x=heGA8(}Oy&Op{I^fY>xNo_{9jH;hCqBZRr85hwTaf{^z!j5{ zma-y-)g8Ne5OJBGn?6g@5g`Tayv7gv$?>D>Y?n3MSd&R zX};_rZjZP!Q6j-pF7?U%p}Z~42u|BbI{m^)3B9l?%(s z(KLR`8j(vR^b`aoA_KL)gjw1(q0%3ZRP}9u$7Xzo6KsE3V*J z@G|rArHY;OjM`-(;6X|}`^t52n3JPBe;A=xej1OsVb>DK=BBgW`h5Q2{8m&M!hFCm zt(V6{UACzlJ=o~jd{|H0$6|~?MN!kKumf7*+VS7A$*TFq?Q=bIvC~}OR}vd3D#%3j zpYR*vK%i%f3z~_#%lbQ+jS^m|)lU>>K|XTvm3?0bfXNR!4i~S|8jKY?_N@)Iy1KgN zo1N;LFOcH3m9g%)FX*+aMHns(JU$!_AmcYh*s;?nN)MR>w3VE1i^$T@=}m^tl`^vo zx7z9L=icu*9V*s~Hr)#`70?|k+c7=5i{Ic!NXMkcX^}1Y8VAC&`FWd=-L&>$$!^qD zfU>$j@4zQzo|eA;qxN@h^dA8KidNy>$uG_O5Zy)!z(F*@0O)?9*B&13sjLp+B-jp zxcA?(5m32ov)rFKi~5UVzi9`q_TRbWpiHGyaXcmJbEa>2Rc5!7&S}`AZnZjo8(^Of zYseKtwna4=gP=Un(6~5`WF`%!lPRM9IbKEH3h}90wqkP2I(Y#7KDm#=Dmv|%$C$}g?xD)QY= zb5{1}JI}gai6s|usx>D19?>rVF~n2ORu3U0cVwI}J$USNvcpplfHke015qUEvzH{| zPSqdab-p(^hm4m$k=Ze%a=2q#SK8d>%&i}TLC8%R>0LAfCY9Qs+++g3ti3oI7= zdsO@k9H7QsjcmW-lE*;xKgRrRJauk}w^LM9G3qPY2$090eoM#Jx;2GMc8&g;;puur z-DdTf(|b~}vmlWa?B;{kh2W>|$>;3%s_|p30IaY80c8Caei(Ym=q>Cqh=k7^Q!>I? z1eN?F=Z)5==Hq-yZIm4+B%CD9S8qZ7?%(v&U;wV-7rABP+P(jmFKky zIQ&xfMa(ZJT?24{hcBYFY${3O>)`YnM9G?Ux2f*kYXfG{W)C0`-&{F=SJlf$_7Yqf zv5(`cy|zEl1;?gGuAzG`>uQkFwgZzmj{Z4A*-?YA&+;IRwF0YJUm?dXpT0&d$6!r$ zk@fHQ$w}Bk=v!x|4{2!;a5!N|7#DPK+=P(0ebG*86%ZirPbhX)$8C z$F;U`T??7ws;yt6jUJxnWjCmX6(dgb9j}#9I+Au6c^izEIYBOT-!Eo-7vvfq@WMy) zpUf366#XRxItIWu-Adkz7%5TvGB`ZG9RHT+u3s66O45T z)`1+1F-lclQG;_=8f|6l?E7?O8*?wpt$$Px&IQ-u97_m)N3HxjT>~T+6m2_WIDY2t zh54?pH&9e?A+dSLX01Neyk6$qH=}Ysj<(PxW!<_D9=5+SBNG_|ni!wqmN^@nN?wHC^%o5mf-m$dgO=m^U3ydEu04` z{wu5d7iFGk33QQxBhsSKFTrx*K?8j`JfFs4U+hI@q#j8eK@AKPeDA;f=YQq;$8)ZK zS?K#sf5G#>Uf=Wm)eEd86d#l=R&PH%`xip~{JQa`VUo$CXeAiw|?B8;-%d*-*A_;r=|f5Pv~K zG~pkG9B5wJ!25|EO?>kXu!+%d^M$!zE-Vm%{cs=x~-pinsfJO4T2LHK3CeAg?k#-y2&%~ii7#b(pph5?P zkCx$(IqXt3Bcz9+>{meY*NyP=S%L2n25Yk=>t6X$$v!t#8L*a+jES`P<-dVajQshI zHN0C&@bkgzdajTHNSNmCDgDfdU$DK@=t8?j67=8K{%g|%NpGfFKl(h8*s1S0=gPj< zn&=Ou?Nxbs^N&6MH!tIGxT?R|%|k9VW*_@u+|N$R#>PEt``x3;2)Iu^@QKg2VR6(ldY}YgR8xV`eS+$%?Q-Ni_)CmAx%Gi?g9tS)7}Z&Jo-;l z4uCqIEJU{^j*C7kc)&+K1-N^9*gu8;vvPM4Hovler2GC=YXi)1{rp^}h#JP@Z24~H zUux_U!Xij|I+PLUF@Hyw02!$eovcZ7>oHYD4O*rgTeb|Q35gUZ(6WINk{$k|R8aZ4 z#G-#@IWb6+RuNF7a%+uiHQA@+27Lp>&lf1uPbCS?Aa}weslPv9H}}~VyZ>nmw#j$T({4) zwtzq1;Kc#0H0Jfi5{uacs9vN~U*(>J?Ws5OdwwIbE7$?zJ5}9($kngM{Qk@phTrso zE+MMgTuJcw@V3g)k#h29<^apk-GF=cTYkOgNP}O@M!Rn_j&$+2F#VPAmURXN-Tmck ztZ%{&t3E|Z;Aj(-(R?QED-DxRc}yW?O$LAOv|X6BX@BLkzC0yV zI+ZwgxCV*xA6bHbx6aQHP#=>9LeGX6t5=USc4NLW=(9u@aBp06DEnh!za9bdbe>3? zNb{lW9xY9dN$?P{wbV3(b?V+H@J0GR==g8#&i?*pzsUZ_1HOND4lmi_zR-Ms!_?pG z{r4;Ja~;;(XV(4q@!&rR=Z~AcS+79-&%gTHRrP!Tt|BLw`s+WH-QWK5?;qeoaS#6Q zZ}eXg!jnf^!LxW5KSE68h)f*a=|H5(s=v}1_={gyQ7%-glN%V_!|vy z%h?`pUnrRWbhk^Fb8vy0Rjj+O|MRC5n84YR9llSA7F!ro_`74&|H${h`vPQ!0icgC z{8w7xe@UPBNDOO~`Q$g8b6iDSLc!at>c!l68sLQ=wx1!$14aiikq=JX%KMp#2doQ>-sKf-F=J%rcx~=po1$tcf52y^nAN{YiYmXn$POmxo z)bIRI?ys(X%e6N%%RRJ^06p^Nt!fr~Q*;NjHS(bog1qLlxg`8n(j8H}xAB?PpJ*Ml zrb{}c4oyrDiQ?DjR13pDFl(r`gfg9jMy-u_C@6&U^|2M4^L??J zf8%iet6hx~2G;n={bxm9m=C8b-kS8~;AI=5p3l4F&d2wi4(9L=PWC3|y5dC!l3{jC zkcv;E_VCKh8boaI5PC&X)P*bw)M-l>h}X=IlfM5ojowNC+b7MQ`NMWmkfNzq+9`HK z3$Z9OYv$$bdXYN?ViI#Vg)+fb>EXK#R59JR_Q8kG3$JPVTl%_L@iK&kMWakXDN8j%CuHJ!OBzG{ zV9tbaB&Wd(Wp8W@j1Tv0%|}|Vm<0bS2mNQE18$9j{CSqhpW!E@R`?y5?u@)|q2Dn) zvd|8r^}XrH3^<=Vm^6Pb1xcd5p*_hm#O1`LTk}Ep(jVOW-X7vV8RXw2P(An75P!a& zxA3!P;ZgT&C4u1SW!nG<8@v53Bd^o(CuQ^-%<9khUVpn|xzxvYauu9>eS>`UZES3; z@^0AU|CL=%i~_Rsp!lM?jb%Qb&q|+MUr3++)cg+H4Tf=1Y_1eL4wVQ$O2sx7b7J0G%`wf$L-IXOQ zCy1IJ32s2InVpHT)Z5-mGmsPrXmaDtJm(LwOZ?!1^*Q#gtz0&wl>TTB9;{X8*&Hw8 zCAYLqXC5H$Fs!=1Vv|gul(1GSQMcs#;KrY&6TxId$PuD6f}4P zq=I8>C!>Cp30!Ayd~!M|i@N)&N8!yxkx^KBlwgg5yKIL=veHNy<#FoMge2pk{A6~A z_y@?;041H)AI)h!4k~6+4{bh;!NGUu=4EBhz3tnc2GG+Zw}4hQQ`7r}PWn%trE4J- zz|`=V9Dm?6+WxS3+O#3ByYRhN_@oiv(Os9b6`Y4{1hQ~Hj)kpx!3Ew&@-7HB;w~Yt(Blvg+(CFBOBXSTib1gc8Rc`vOLl_s#zs_j z4u_Tx{!M|m-0a@w1fe;oCf)_dWRi;_xwPri6|+9BiY1wCOj9dls0?<^eVx(_V!NH= zb=iae{I++5FQY&C@q~Z8g!{P&PHK>&IZyTA*&IpBceuH(zq02gL?bUbjf(!h!aL7- zTv6}ex7l6~!aW`@fs*eZ9gUc*?B5H_$Y7w6iNqkh0sqi&#UMI*X3PAx`p6=OemB4e zCAFU%VSgBWz1Z^tzD8V2H+D0hqB`Lb)vtmEN!Lz;ZbwBD^(oJK*EqO7+0Iqc9-CBz|&Cx-OS)E0mJsygVQakdov5MTM2F!!JMiG-3Z`xlk ze{HmyK4)z+BpFKaBC;fT{Vf^;Uw(FXqn%ZHy_Ikh7WpI1brfWaR@`@e@ z7nd@>dO~4*V~-6bPD%`$RnBCnq0LICGuJUoU+*l@sp*%d}ftLj+Nn1zvGku@AxUOG2TEFvy2o#X3NCI~5Uf6FQKC*Exgic_6 z{`@(w7=nt53P6!pfh#Bm@T@_1o>Ja(J|)S~tZrzB@q}x!omC@pwGt)Da(9E8oc$TX?h~~J|Ws!jQSns;GY|wp2BS$I)1F2 zM#k576`&lrOT!gk9>S-WPwPk=%JQOw;KzXc~<} z?79Kmh=y<7+8=avTSM{K?+R8ySs)b;KX>{Ng%0Jp(E`kCYX4m4*%d3}vBXs|g}@n; ziM*=|ow1GWY9*6Yrf5xJJlOP~oT*JAO=++FBqa?~4=s_UIGDja{aSwqTM4WgT@V zG=431^xAaVJKMV%BWUlL$;{lu7$Zs1E_LIqcG^z%)D|6_RP0YS!-Eg`;EjSg2Ukgi zoDjSMNb1w%^#i$#l=89>_KWWe%r#sus%s5n+31~Ty6e1Fv0j#5cr@+%@y@9z+iZKV znn`kHi-mfTA?Em!;M;7KN_qhG^U6@pnfjJDu`x1r+lIO0exx!^HJn;1wCk|7R)uY_ z$`RFUs)zcX<0ex57j;qJ8%iC$-QC??1Ec=i<5iAfHbMt<@Xik(WvvE{eeC^+`U5F# z%`L`0ENdIAZbGNvw2vnNqt@}s3l}<~uiI|Uy58flE%0~Zx7a0le0mTj$EEd=CC1xA zv1D_uCGJH%dQ0{(V-$p(^_s znXZ(x!AALY?STCfYlwpvmsKr>i57Bc7HIa~ZsiMn9IOZ5 zO&J>vPQUQ})v)a6Cp6cHH*59tX{X9iCUTAinD(g-P_2=iZ~@Sz|gss ze7*+PJ%}*MM$4s+v2tte`WRg=C^WA$?0aT6-N?dTy{CKd12m zd@$626KQWLpfz*T4P{qk+#z-;gUzDhhCk4i?0IXgWf|0ZO{YP#=>!Jt>nmf)FSTR`J^JAhmWYp+VR+LQ~- z1oF(ZFp?hpGA%avvtzRck< z5kT_L6Qe=cj-&mV=s!6u`>Dv8nC{9!<_j&ZP6GV56kD3EucSIow=y73Y41uNGvNyP z9Xk1NH;*pPOomf>6n7&0(Bd@^ghT1oAmr5Lk;o2ml~8NPiKgL5|8~b^Bmy*O^mVnL z_F_831!K0-@yI_vdA=Ry?K0GrFMv;nC|>@05dj+;o2n!*o|kV+qtXtqK)_>i9?<>;ebQMxBhvg`U8!vQrejWo{GCw2m-q4yrC zgK@7CA;C27L{9813whg-gRPf`f%ta}42LtkBjd$5<$=B!KNg+GGP2xkMmADDaZyN< zGqB8`q>Z8JDik#;_coA=p+Jr9^hBc$0GAoXtUJe%E_dr;}cjs9++(l5n zJhSKqMouRQ`M}7t{61xJBo_O13HC=k_N#0-oFt?A zON^JrjqD&p@W$!n@K%;bTL(MG_Js49pD#Yk13f$E{q^|wOsXhtDUYLxKu{qJ4grg< z=fdjeK(vBzTg?Braeu_vh>IN)t$Bh~sTw^GMBamI~liRtHhk)4Ll3b@~tJ~v;Di$?EF zmVS^G*GW6`|n zi|?Vg7F+WCS`|6Zt@mW{Ub3gJZgg}!Rw~lH{ZCOxO(XqtBD9GVG1MCjxVx{XQ#xa2 z`W7w1OI{+6f5X-7PgB9eT=sBGiQ5ZaH62}V9IH7JkGgw>T~{1+&{$<#=0mHh@kDkp&#s%wGi4Z=!3vpj;ga}mUs*A z#Z|l(u_M|f#`;q5q)K3CU9@=e?ekfgC?A33vF4_z~u9E7whKq|pIEhb5 zVlijtA-|MM^=Q-K1Ug=8vhjL4lQJ#*wM-y|YiWED-=f3Je07TqW|~~`J2L@0xDHnH zWFTV%yZ=4>GiT_n0-KN41DO#)R1&x*4vRbc8`srO8?oi}JRQ#joVK6OMYa^vec7wJ z2zz#Kuwt6?w0|Y*e$}fjg0d~_>R?+q7eU5Ds%F&mrX9G4B@uy$KV!NV%9A1caq|ON zUD)@CxC~Q~yoSTwW`{3&0T>};VPUOrY}{{;w^xA9t(&SfTNFgY*IU41U4=w>mqfLBw4KdWr_w9_2 zqS1iXdZ5-Yy$;R}TnvkThy|Qx@WMS~NFfdO*jZLec-Bqn$ydsuw_?MnUijq+i_o{F zi5#9PdfJtGFJ)BhQ_r+s{YcCnZcO(ar2_v?;NWeY4g1TX0u$n<%sFdBgk0w{mi1`E zHl!S@w~`dc)zgSM`I?Gy<@7l*CIVna)209&wBoMw)W=VB&vUh8Hjg`E_)NdycXxo3 zu-&<@EJUvJHYiM7xudw`EY3X7XTMg_^rjjeB(Zo3<)M!QL-4<-rVpm;-(|MenY6Oi zVrr~#m?LkC6u=iU8Gi9{VGTm&i{IA2#E4ug8_7wgshmmS#By2}u*`F3Pcm3V6ulnw`yZS>@#YKZBHL8?WN?o?SuIdf$FM5yURt!+qJ=iT6hFoiSz z?9rXP{VrA0&;e#>D1qSXxA~y+9bK{Mi0LQShtd1T5kmF7OeB38b2}O6nlqkfGFhh2 z&ji2qu^sZbyhP0Vs~h$&iGcmLNR(IImFXTw^9Edzp9d$x=L2tC)ssfUcdvxk+{%2m z7mNnM#xzsrPqcjO*0HMQLs3v+h^N=C=~E{rd*{f!!h?g>u(7Z9t2f6NYnk{7`%=Bu zcW^OJYE_>Wb@O{h2vgW3_uk+c@%VhvW9g^9lOyq-=8<1|$c722_WGGX(DSrGmuspc za52o<-Dvw)>+9sI(TuZgL70{0)(=&KsRT*w9li8r9|Q~778|}Jo-Qwm-6P}N&Vm%- zMsj5)WI8WJ)vMYM-3e-HnW*%o^}#=B>k?aQy0lZ@s-!)UBUVe=IydhF8RK(T>)bB& zeYLVD6tYY0yFJ22h&~YzOBKXC|bUh`-CxuUJJf@>>fmxjd5b!B~6}6|J6M~Js{cQlJ zpDJV2tRkuN!p2sH8i@k4(+zTxiPDR2F$mee@~OYuhJGzX_CLFJI29FD!}<1;trRwi zpt&?`Bth{i208b`kucOL0H$_ zP>`UNEl(RjCwq%kLG>#kn>j`(v!>QvKK+5$4)RI)@sG#1H)M8(r%%>W`b#!m(jka333;IO*2G!sSP=}RoL9tO%#3OA*yA$;{ujsSXSJc zVNSKr%4Rkt0c32}KIdscsOMg8h1QGS2J(Iy;MB04@>!~mgz?BRYTP5=g!(%L^W5r7 z?-kbc(3QIRQGKXKGP<;=b#HEvP95Scz`0nc93C;l{~ZI7Y-AU@dS)u0?D^J(g)1W-CUlR=OFbi80Tz zc{b;p4N_wY%m}eai=)CN$D>$H`>Z{UR}Y2vF&2C2nnSi_5xQh4+)gzL^e!V#?a(v+ zaIB24esdcD_YGazSuW^nkgKd>Y2Cf{oT9>!gctp1dPE6n+^4ZmquNO%qO~r83l#WP z*>egCIOaQ|eL?1!xM6E+t1j@Ch}mq@H4(?x);f9%+$O_uAk80WvBaaPsVN<6Dp>xh zN<*O~dwSyyIHs;x2F}9?s6V$;K=J}pl>SZs}nHIIDA3@e? ztMQC;^xbsb=1|mZNTsKw;=733e7>0jh!3rdEJjv%x28C{!=K(KN2^xV={YksdlqWP z>nRvM{JL$xWOzzB$OKX?N{s`VVELFIJ7KAis@~rAF#lUh~HacVX;r7KSB zx+9)LgKff8$xaBLh3k44OAMR3=8#!q$Jn=?tQ4&%OCzPYef)Y1%gz{qx)jlO)s2S~ zjuxM1-B7kg8?ja~W}2dXG?h6~9AFoIeR)+)CutRL#52 z-1JIj$}nQcNTEppVl`RM(l(E0xSD;6|0?SfoqV3L*>*fUdcp|?_Wde{(tOWk40j$g zv!A~3-x~)UCf8J?S{O=piOCQe-k69`H4i3x*!{6JU8$)Zqm9`mN7PrE_p+2s*m10^?EWJkCqhDYLN@oIKoK!Ui%$B z=j|{DZx&{Yej5!xkQzPV<ekZ4ViKJWpmlM%v-y0)fiU>T%)d61}zA^d8`0BWsl zH1~D+#E;8yI}oC=AWru)R)3HK+x$M#ZK$nknog<(4%1{u3_24L9PY2zzSQ=;Ab znZ^nkdR#iKkupX(ENwCTn$^DPiW$#$-=zUfE^Kphq%^Ss`Ykmn=f15%zdGOkD9LB- zztiHv3J6oN0?z*aW2gpOeQ5`nL<3hb)%PO!P*Y@MmHnG*wx5a|8bx%*zJ>c7+WLJi zppBn!wRtPN*#bvs#%IxVt3-Ve%w76Pr{*C%e)v35ReP!_QVnl{jbPTHdeOZ1SyR7; zTpNIYc~Af0WaX$%?&<8S+mkCd4o?z~*FJW}J+;~YmS?W6C%H#Te)z>0XpaQ1UD3xS zB$6Xx({x`>s))08LR2=w+O|aBAYJ(|a|i^*LC9d`3H4V^GWC-gJ!3qT$^NW|8=c(+ z7HYc6f(O=Lzg_EzS5k3AJ*Ym7`H88s+>l}Hs6mB0iy|MW!DjT0q2x<$Py;f$kT6xB4fNP~+ zjq!kt@AVy}90=*yt*oUmNgdVo-Q7p+ky-&K!K9TH6%=mkt3`^L&m@`+V;NQ0b(&vj z%$wP_lMdZgOdq&Gj(d{873@0CkxYcxGWVH#mCwLLC*Cw#bv=4Eq2FYz7;((r)9Gc3 z64{|HOTJGdTI2}j)jMQ-4pdIhu34SzxVFzXjFg;Rx;NsbC#4)9MvDVMk?GXTB@pu` z<3KXc(2le(;>I}z$;La14P&fd+7_!)A?@K@qg^#^{-v+BUFY@6YB;}^83aEf;HTL} z6zCP+2YJ=!i_^$co^1GN(8YQ8RFv8r%@U$|A0oGgIRua!U=q-+*}j5sxYxQHUEWos zmSeV{ZBoA=imp9#$R%lE0aueU3L`b2XB&T6Cv0QXKji-p6@)uDT<2=OHAt-l?!hxH z@zwmOOpdqEbl?eH37C+~5>u_v;%INBkRQOn4|n>;@ga;}1w(`1=8jmlcM3W-Dd{fh z$O@gdD9z4?7B0HXT${pd147 zJ6suxkAk5_MyongJ#)^icIq!TDN1v&?+wN#jLR2qm9JsvJk|0p^~758aIm##7HKJd z)ykZCW&o25jhJ0{d;pzB)Sw-pPv;QM`(qI zH?^3C*XZa^)q12pxwRq^htC3ulS`8P1WvU|w0>J{!43e$x^#!PvrCaOI)36Lm98tJ z%OI!swCGXv%K*FG?JG=2pUcR3p4U*!Tu11L=JmeO91J5p%i4d=wnyuM(YPH^bc9Lp zZikrLu<@$15*m#AG<6krvrRWV0C?>c2QSY3))nGia_nS_XTv4jIOFg`;MVA1*lCv8 z2)2YV7#X}Q*m0V6>q8Imn_-L+BF1C*1T$e_Smt`kNXBog$`46V!j#^KPF->S=DI#V z4kL2M^Ef^b`V8WP51}CQfbdS{AFH#kC~!iA_ra-E?{|}0grFx;j`uKE`ebu(%C}Jl zM&ijAO%?uS9NCI)RWSGj*iTN=DW0SkwYJ2hp8&6#K4GVl80T#jC9I;Xd1blh@z^fuU3byeLHcqaDRqsH zpAS?d85WogJ|OGT#NIPM6X6fgleBlFaG0t^W{cCOH6c;y3TPL$LBzEFP63Kgg)J{$9C>vZz7 zCfkcIXt0zIm#2xjjfLYMXkG3`#Uc^*!Xl**68Fw(Z*IOq%wy^fgTc6y)Rn89P6{2> zuJ&Z8mu`x)kg?t2=x+2fnVwAx=QVm<7{+eWLe-b9qOn+QJ>gWi+*QvyFwj0de3Mh} zixRPV^((|Ryy~acApyl3!iBqkB+z*WjxM)y1 zBby=iq(BDc$9Iz?67E%PoSqz==cTpmj1Z^h;IdO6HHi-W3D!IYm|(<9lp}2T?*tsa zJ+~<|HOR{_@Z2pX%VO?cF9O0OQzC zmefr$?ua3r!XrBUN36n^v~#N*aAu1luh%v>kfDcL{rVui$%ZVHP~GTx{Bt7VN~=0< zH*YZVfd- zb9Q#_%TmL?fB(L!hB2Pt$+mo^W;NuBi{ire_VyRRoqi1% zcG6V<0QWM6Hmb;lE^p1Vu#@-#-C}@ZhaV>Wb+b2My1`@HlZ{3F)7zz@CBh|8jR}iB zw>bn|>AD>6NXSB8ePK02;!mF8gbomBKk{Q85jh<{&1qBi)Skdx@xoda2r0uBz&ULj zvPjj~Q>#0Ge?7duxzMSfTA=6Ux;}6x5A&ZY1Qff-A92xy^rzTK_#D*LX20Y?%&tvL z2sj>MMB91N;J!zNab^2*K29sL%E!UoM1j@^dj``;(Z?V}BWXH4eaN5P>A>ukf!`ke z?wU5u^zaT$5(d#2iMPqCl^hB!1sOjf$6bhizvWaL&k;^Y`QyvE3*y~(S@rdq2ihL3*2{9DpGrBXn?GqI^lB6Ge2zz6DoWC zGAX3L78DeM?v=Xb!)|Htfg-Ow9S05sZ<^tU5z@cx=u*ZNF|5l9QiIB78$G&ng*s2{ z-T3+`+t;(PKbCyp6%^#J#5U5VI6g89`Sj>#0D=R6Qz$x@ioC>(n&K?gia9Ckc7tw< z#p!8;YIB(k-vA>-3Ur^r4*dx^P(WZI$SjLByRhH|>Quv=$-_ha{dgeW7JaZ0uk24U z)W3As{^TTJ=H+N#`27mC!H(3Y?cqrp)iqawlV8L(`(17fAbr5M5&yo_c)>%6!5ze~ zlm>Fz^Q$-;Q_qdCWl02IR15%Zs)TctNQu<$`My&T%=#e7BFC z&rZBL1zBA6J7Ww^bQ{*WryWhpn4T3PjZ>;sIss{O*a{`{+c^AUrqf*0oeq;tu-RD(=~1ZSA$dHFEtPKkBS>eR4IM;5 z)kd9{88wUU-)^CZnP1@-WMow*?=oQ~P+xu--!)-`VLE~J#DDkOv_TVJ)l1;(*zHBT z8RLwdixkX1t@CR(qxkV%t7BQ|*ve;P_3uloDC2~leiGhq!E32a5WSrh&G;i|E+{Gk zYRdcFR=h;g_QH#tck9XbkFM&;xYU)?9vb@QKFQUmIF{Cg-c5R}t3{*q6F&BlB@X*Z zK(qAfp)CO#`N=ZgQ(2&566>^odM-AUI3y%*U_8+2@k;xbqhI&q#a@Sg1&!plS4WhW z%dFv$LMRl&VF0T^?v{4LJ4uxNS;}?AG1U@NKQ~nD^m6XXY?T6~X4V)cHSQo=!91W} zhHdwox}Oz0Bx!|P=s%|PSR2Jet?#J!4SSNMry;Tv{6hzAZ|R zwQRFfX#Wrv>1j2dAQq?6+aNg>StRtL>$x_KE#Vy@Xa`2!UC=vBEQ=2lW9BX&=n56w zSU9-`^b#b-0WNp~fEigRUYN^+aW8|qG8vXG6LL-@s$7i{w0NBL0JKf*DiBlF6$0Cw z!AS#iF#Et@t2yNw3GZv+Um?)n=T?MWn~*8d-LbK-8s?mj%{Kc8S4JIcSf7PqWj!sM z#aA73H?VlMr0{JIX|ab-`}o^uGc)mI(uqP79x$jjHD6O)M{CD;Oo5-~D!&!H(T9sT zgs)M--)5JJ^O}5woMxN_zcz02w+_6g6vtl-w$=yJ3R;fY87c$zaCBP2?+-bj9wXaf zk@(|xYSJGOawIj*=GA@3H+W>D!kY?Jc2H%FH9tOg)i`WE3@H0ozCOJ%7k+&0I>>5{ zaz`IgfNVMA$JsE^?n-UqBu3{ZBvW~PmN#eBB_GO$`>Q)LkirlA@wMh)?yAjELlx}@HgcEa@RcK34ibOO+NI8UNpDpvOcIDT%(#$s;X=#0R z(DT$%TuDkUE^ba*RpghkLHVnl^L0J67& zjz`X3n8eOK)?WIsfjer+AKqU(1{N`#&m?)M6d1TO0LD>Bmt#`jJ?s<}wgQJ6f`;mm z#Q>-pJkb z?nfUs5ZlbciAIuTSWpHCH4c)V0H2w!C{-uy~;k#V|!QFX@r^kAzuK)RK`D7L! zyhrft|Hs-}hgI3F+r!dbf`lmDU4nFnh!WD>Aq`T}4N{_{G>9VI-Q9?Ev*>04()CSV zb$_S!yZ8BB$N#*yU_y2~|K$SMpG)BeVykL5&LBWcXBo*F~&_`t*7%+d~Jva9jO?!^~f z16y2R|1I%5d}j2-jjedSs72B5;=7aZI6u=gs2+VtV)lh{Y&1C*wMY5Ne za+iar_$m`EMsx@`B6J$1#?|^u8~y8>Qw`R8{Kd8e4#q=;qltZ}w&qth(uiG2h;NgC<)t5RtTJkv*xJ;C z>fd&@9KrJfldWo0W*ZnCE7A<3-|37K+l4guvo6e3Lay|*jHwp!;Njl35e0&&Nc0#z zS{rSU=gDkhA(YdaT;HN{_3A8NfR7Yrjv+N0(ix51 zMyr|CW;c}4qr%gGi~+$X$m&_2T0vK#U{lVkM9*XM17-#_x}wKrU><?mykYjdO8GMfQZF00vD1?z{%tXQf1-Y^kKmEmlJ^ODBS zP$C{~Z3R~lL07YIO+?PF5W+AlIpz{KI57c2l|xzj3AV0tr0n9gi~mKIcwJ=>?^p`Z zBTRo4zqe%*P_)_?>$|^#BEwkPv0W7fp4-heG=&e2Q*yE9i578G93kHde_QYmb{yo} zrH<#yK-gHD=B-l%dVm`Bne5+9(u$d&hhvQ8+dmcz;Sz~-Or2N{Ent+LS%X2L##9-L zO5G&vrLK^S!9ojLYs5Y(fweW?O^HTgmBn1e#KWG8nhme2vOfbX2NBFm!amIdsJKj^U zSV!B9*BviC#e?yN=h~$y&uN^PW z&ocn43I|czG9pXSd^i(9D&k@GaWa@)m=2{D>`GTD0DhZX>Q0#P5_>rCS&@5e+{b7* zEzNgB0>ZWi4-F<=JbP8N3V(FM+frp-ac-AH6OZMK?^s`r>(*Iy^`}Zn*^I_#?Iv;z z15pVNb-V$#8aEA}-sY*0%gGyM5M}vdPcHVGs@PoSJ(%=LG(lRJYOR^&Ud2*PjO_!4 zK1zA96T*=s5*I%=F`S~o`X0jvn zp3!d0q6Kbi^|01J;Sn|ZRc`B<<>sQ`@{7H>x^(#=sPFBz*Q^hWe~|fZX`N{s>!tVj z$#Gztbfn?t;e$zL8w$3no}>MVuLOXBgS}^`JkG+u-f#5C=)jNH&gGV2wpza3NU3Uf zGE^mReDP~JW(aFxP8DKM`S@nbpdTO1v(OV9tVfgV5*9O9L>YObo zd%}PIl>=sJ**Q6NfT42qQ+vbecOr9Go*zX|UG`@?{L{(0v4R|oT>|UlhUZoy2s!9( z0MuWGxpcy;FK)m}-{no$#OcSDs|iUTH`s%WFL~-Na$Y&~oPG<9p2PrRsKOC!OmJk< z^XQnE`g7Ey^-8w*siBnQpuNjKM1I*I6OJVp(Up= z70^TRS)igAhNZkfR#V+0a;vUr*`?3_tLFcw85s5h7|<_OB=WcZ!<7gjg7E}0dUw1Y zJ$Yk4TZKB!ekPwRDZqX?lL?*t&5*!8h4q>!T5hVk4bA^?+^%yHl?j$PtKER~EOOcN1pc3rs=l@VaR0kl`d{>hF)KuN>B6TVi>2_YRA`hI<61W6K!PiMOwHM zCDa?VNm0op+_T|zNFZ4g`n#UaA9|9nI)Of!NB&1gvESW|F<1f?9p9>ZWq#U_JoyR{p?M^=cZutAjuF7PY z85THZHaacmA_24JD2w<$&=4k~HG9dk8v7?S{C%3$p4|*u0AytYxmtIq!mnm|P9DDx ztw3gjFm5~V8TBtuvni4SO^*(oexNDEF5hni(#%W zjz>-?8Nna~ArSS!k53q+LXQBuVdC!06_y0x*5#p}{@i>8_^OjE;+)v?Ea8s={%$t) zKNQa*$pBy_G{_1=@cSP_w3~;9ux_uBlnX8}z%nGtefB?0ygjPYAj(Oqq9?$SoJk@Ze<% zU$zPgE$(j{il5IAXijG$3jOK44Te5BB%>A5IsAOBQrj1u4QEw6!me zIU0qZ0ZM_r)m4-3;`uT& zd2nMo4Aa1VWHH3*yd?B@qz*1c!8Ftcg5n)PB<>@%bYr#MlhX6F3_ytm;w2r zV0SEIymLPIZB5rl9;$0Ve_90tm<#!fhqT=l4!@f_|D>hxKbW`gsTwfg{r!sN-oGq( zk@EigsYMgjGXTeL?z0SeVaCE)1rGLzUg@(!r`ydH@11eb%BLngRx|ZK1u%h9@fyS3k+7v!>;?Ox+d<1j&eJyBQ?!D-M6-=8X|H4_yk2JTXcmMf|%_0-%a9 zauiktH8pw1TyFDsqeB^0^S|f9!oexlIGBWKKq`XR)!!#B09 zH}JJ}bV`hSl&Itquz4MfqxRT#r2o^!{&!yRfdn|z!HSq2(XYGb+YP@rkI@WnD*JD$ ztHu#>RG$wtf|zHiY)EKi8iY+uI3&!`6Qi=l#YG>$?qZyp9&_{b^knzEc;htjMUBXK zAy3WrvrIH>B8N%21(Wk%e6at`(R{1#@{(CSxWeDos2|(>kyhSrf8O(vz!{@dIH7?` zTFArI!IZmOspQu-<_w-m7g!xJHmC0Xn8)T3faZT5Z~vHQK2ZM72ggD!!ypQGBE=Xy%+`+R6wYVdw^@NHNyOZ8w9jw zpdTL1Wua8C3vxL&+j)Vriwg>{J0q{E;ysHFU-%>5`+#%JgYE>73+1oJWZ^Fsct&SR zm$$aw0U8H`H~`l^M!}^iH+lO1VsjLCHs?TIjsDx_Dqw-gzw5R|=Uvdsh_;(6)UNT1e9XV}7eaqZ zDH@wi4n0eP%{^ zmziaSwPgQx<@pZ)tC!klp)rdKC>#-iHkTkzqGSWe%{CUAgg?-{cs-O6TLNOoaF#R{ zbzS`&5&>Lob+vxSyVzc^F0-z=rG=d43nbw`KA(TKL71vR6v@g(9_=N zergj=Dm*w2=|722oCBOqKij73aTb_^Dx0}h&<7~Cn&N;*KuAm(A&CW1Lk!3uEpBXt z09lnuBbmtG4(s1K!Y`+CAafV5HVp0Pf5)pFWOyW`<+TCtSQhc%w-6|#O>vVXyVByv z=avaQ?+9>&OxJ^@_e9GDb0Ede&$nLgjF@cl78Vl|%in!i;QWU$-#Arj67fSb>j^pk zA0dN3ZtS1d|9^f!B>%2d8t>sH|MiLisZunG1Oo9V15PB%^V0c!%K>`>@6)I0>LrHV zNj$8xv$JGnowP4LDW8K-Sro%`?6N=K0fe=n>uKIgAdIs0O zDD!cgXz=CB5)M|{zaI(x57Eqb@&WL}J8vDif4h)51e>{93?iRK>ltpjRDqto(Hz`;ngrl)^ zO=)$T@#qcclK}=(Gzc(Hv+cz+&DKxXMz3sLI*J)gD8{Hx$7M*P$VqxF&pB~45vuO4E*bBQ#?RW!7VUDA|tx=h&|ZhAIZPz;<9T?%R0DY{#HljC_u~4z&9Xz4>p4{%x0rl4OdR0kR5LbInN8r*K3gyp%?K zU)2-SKK@6;T@(i?Sqc&Jl7LLy?HlS^XCq8oDTe zf0aoU;N>vsrQ^tHBBdX_zG(>iI;2at{~{iJ^Zh*UnDnvG^`!iX?sr@qY;QdSebck3 z>$&Emum9|vKoTKJljvjL6#Fm3RVN2*2gdi4zneKwA(VR0OMJlJgen;m$0tIh!wF%$ zy>Cx9@hEiiz#j))4%~Wi2P6PSzyTRZc)6a3IaF2v@u+m6saJ^*o|LpuW;S1x6cBo! z(@?!={R-1Q5WATFcg6yG*D_3g5YjC_$Wg6{Wa;N;hGvKA$*Z8STH^2BuC_b4&!T&w zB{0k@2>Rweu;)iD;2z5Nuc*J?}z+RQuY=D8rtZ`Tk$Ld>pNQZsbb%DsJ{|+*)w+O~;_4SsTra?{KfGwIV zga~S?x3}_g$)X_jl=$M#+8quBkE@>Z#uLRao?!7EvY$7-9|T;r)+a|@?rOJ42d3o* zRNxLWM2Wsgq~O)&_*`Mn<@`5q99(&_MzrpJXcg=p;{w*_mzyCw8bbN|?^pnwuX2|k zGMG|l%G&Xr9B#NX>)8D$#;lAM2a6FXz!0cdjw-?~j;&T$&BJF&VXC=g9rL|8lcC0L zlse8$U>{H;1`lA|2O$qZ7gQP^2w~>CMY=+yw%)>fvk&A2y)W?SUwsJx%=qa~V`6XS z3HC?M^VThP^Qf_&9|J82=W(>|q*CB4$bhJ|ObTDc@#85a-0I)#zMcKqMxt|}p1j9K zzOV<+?qcrdYdhxOcEOL@-8ByiKl!V!4Nen4=J6;o>BHwA#J&{&} zzb0!=juayHXw;s1PsLv5ljL}zzU3z_X1BTZV)aepgBzooEdlwzN$bug5$@B-hBOrg z2@RZ!_oDtr+)CEf^E%d&gKT{1viI_-+o60xV`MjrP=`xq<&fVl?C;qQ9o#@NJ3=f& zH6<95absE0>rW^(4B&cwhGPN%I7YPs(WG(a;Zn&m*Szlq(zaGJjksTUq;k;i49%0* z`&gjBvtq^I+x)cs)ZklR`7k^+8|;UI4uqOqwy!;?)}i;w;Kx5-j5-Fq4M)&&%gbmPFGgh3+U1vRWL zg&yZ9RnV~Log=ZZB(iyK8ROk%CRBQyd#<{O2Z4^apzK`Gv86(}jKS_xv`Uolhn)Ce z8C%!aKvey{FBp{=U$(9PdR~Q$VWjCIatN*7KFYoH8W(~k%C{MO?W=A;>A@Ma+wa7( z45fjUlz?R)zvHz%19WGK6D*u$u!NqTcGbH%+MhuoPjx<4d<6CJ!^eDY0)T`1L2cBD zZujO*d8K>(>(kSUB-h7}C`cR=alXC!5~60gY`7$~crdbFZO?eGM^ZI1pJj8|6esy_ z!Gj{c>JQ$_M>O2*>wK2myT_vzub#qBLvvK;_4omUib_!GUx>HBxMN35#L+ z@6$quD2u+nr|-FSG^t+3bemV=?ETRKSmJ_fJi1GQ6{}t{pS?URnPkaDVnW!_%#GSQ z8@9MZ9NDLQXazaBwo4-V9?01hYj1{YBDdPXG)po!hJL`y<@#h-Yt;f7Fp*DD)>$Qf zR7@1P9J$jsa#|tSC;m_G*kkS z7{>kVgb?B4pYDd=gmTxZ1XoO1eABOH$fNkb$QYcsbo|ghyry1YJ@H!5{3Ji#=S@r^ zjG4lj>3m@{eq&wkLRNox`m})ae&V@1laa$p6YaNGTBo*t7d$``ZU5q^(naHDp#E0m z%Ilps=RnHGWL5@WD63whM9A4K4}Hgiq$Q7KQMOk-_W6FL>$Dsg7BdxUEu^kSi358f z?Q&YvQ2Ld!#@x2i`$JZ}WxB1wEJnTd4a(kZ4KAybwzl^7+6tsxTA_gu~P-L1&;kmM`F53Kr6)y;e%rg;Vur+QFh%+fu$bxg>YBXUry8h88i0B7lKW6y zZ^;`9yZVjww6-#9%d4W5_@I-m!I~oiHj^#%ovAwc0ws=I&;vyjtaEunEWiMFwZ^I* z9vBu+*V0gG1Q|kwFrF5DPuG3C(oJ%s*}HG*`SoQNkfDkRKQE2a)~n&jW;fUOHtvm4 zR3)lhrhRF65w7vn{*bkMQMRVXYTsHwGpjK{dhYG%V;}!uuaR+z?bS(#?a>m!>55~8 zUa;X2-%+on`Nne?I&&eLa?=IiagtK!CUmreI;1kTz!7MB0;Hxrw5pAczHqyL^VQEb zOOtbDpS07Rh~kKGbv)TvC$-c!LSrrzcN!w4q^gTc9Zj(`hqaL;PvqR1lJD>5u`her zT}9#cvRgo%R%Hp-o=jSOf1_Nq3|jKTKiXaqxvy8gNZ_zWbl$tru(wrcBB8kmSBdN) zMZ&!DE0}Q!U9JPWZtFFSe(sIIHNN^nfBlKeCcw+{?Q74^>yasS!y7Y*gx{N@REvbY ziR4%omTBd0zCDWfogpdE%G8RFG+oH!Xm%g_R7@HGfv+$D)gc z^E4NmuJFV%vWdJK;*nwlPjn(K1^9-qt>f1GZ1x{mZO#^ygSrz(Fu;y2DsI(E|MtsV-LM<_rOXY%yuWtn&1_;oeRGOe7_^rOkT zzOAI1m4zf;$8`vRPi(!R)5)cJJ)RT!Wyv#h0SINp@7Iv>SmA{}jar088_BAj?;8?D zLrYW7PGxWA*sC<$nU6h(-rbTe@|z%@>@E~3_nlg-zT4|c1QUpA{qAe<$L4PGwNzc@dg) zcp-EPMr)NHc6Cz4x8nxbJtnaabP}BOf_fx$fnBJ*^uCOXpi?08e)>++X^XMt!!H7d zINS#l%EV{Ri;Eeu+&}G7=2d_&6q7deD*L&2mS~o#CU(As9#K1w%b;Afaj)@Hu1U+M zV6DAn&MOyfZ4QH@PEw4-B7Z-UmaKxWKc3qho6%OC6A%3l*rKGpw@ZLL?S z=npY530r%?_IAZ?wYpw((Fd7}yQ&A^7*95GALX7oJ<+aP3&v?{02+4FUdcle{V7%0 zd#inh2bt&72A=vy;TDZJ<3+l4mw`tH}o4XIcOM^awc<6!OpN7j~WT#w|RqvbS<=@#01Qn`&0dA?nR20FcQ5YkH%ME&$lRttnxvZ_@@$U~h^wqv5>(1OBM53!JfF!i_2WT@ z+L(2Xr@a^U7IF+O=eypo&u$Xtu&Xyja0w^k*WLcgioP*{l)_&zp=OX@Z6G}_XJRpO zi%|{0`wS}I3b|5*U1(MoNJa{Ph6(jq{p&G?1ZEp(u_@qNabBD}HZS>xet1gqL=2@c zJ_4ZTUCuw1}c_`Mzz3%E$yAc&Ua8bYA&v+}>=*5%M)RBbs zb65Sw?yf5g(bY;N`rQi4^zmp{Q~;>oTdLmcG_MO301OEPD$^Za>~76qdJQ^G5&#)g zar=t-Hman|@u9Ua&3aTs2~Zrwg1`+-LNCCkVP%s-8{|5R!Vy@kP{ z^y=8x3xtTD&&!xD#`?Z?_(&8&9G2cw)Ei%x;Hz#y=QRnqOx71n)O}@-^0JY!Z6~2x z#smwzB!VY(_s?=oX9bIG)As^# zpfBCBW*|jS2gTUq@}%GcO11Unpe(5;*)ehSobEgxSs7tpx=NaTR;`(xEE!&_i^wE4By+lfYi9yE~O{9KD&^70CZwGHGe)Sd1Xn{(fDIu>%PA%GV-^Gigc+u+y z_frWNoaRdaEF>!{zkLhkF$w=A&5^-F)cua6;X=P}{BbY)xo&|@9wLXCvu}ShQ`wI^ zqG}U3Aa68ws!=q5eLFO-aZk$rjr|owlP(aS-a13Z&IMyH`zf9a#>?VGe81=RF2Q)G zi!X=84k+|8MCkEozuV3~D(kmCF=lm-A6T`XnVCxDu{HJLtc$0)b?f7SNhfomnDOyJ zDm+eLH-wg}v%2~^av02-<4^GmV_^rks*w}gOjJIILLXu*!O~hLfc3x|8cG(54T}f+ z$_FX-Y*n~*!ZA<-G=Yqf8uySGGAp{ms7_kJ@cz`J)TeH!G$nMm6nrOOlQ=m~U3>>5 z^isOqWSgl7VK!$Z!;`<#q8lkwIt=sMEAN4Y@ZrjoG>Q+>qTco=k2-mjtQHzyxj9#u zkHaIv|5W)@=wsSOsDbql*Mxh#?3j&(-n?Z{7@k^`r0W#8?%X?D}>g_+TFOO9zQo5oDh9FYR z$ASw*MZ%+32S~1$dyn0j88wYXS$06keV+dB2mg<79LcNc4up;N9|%)o`}41_fA!x_F}g`2%{>X`MH zpRwFV;84$JDy-Y)Q$P5#-(rmC&)u`sFm@5>EOD9HNLmSYlRLpE%{=%v4 zc(B~DzR)5b6tTEq;(A-C)1d9Pww>~P2fnb2^!4bP_Q;nPaB!EWFtc`!>wAI;Jl;6@ zT1_wmXQx)@4km7&gfdBB?zc)e@P5U?g`p6E54neUktj#-1Plt z5HEWhVS6nZc|DP-|HWS)O$ICBgN*(Ad;aTRnmOSMuOT{Ou6TW8W=owjE_<3V64Cn4 z7CtCsD@j6o5?XT6S&iA#QbVA0Z~8w7@|*@|is8NPQ`{J4MO9~io6l4q+i}!nr4Bb$ zW>T>=;!2~Q+6sT-n#^ZU*eLS>9%`GLSf(-kdMc9$o=q{^BOTf(+V(}ja@em%p~f7enzzZlnjfvhdHBp z)*NQyDwoKI(J5aZ0y##7uz&JkbDt5ZcTKx-z5&A1;Ic~F{Ikrbi&IRPv7Sd0hnLv9+j5#-?q>j*;f zUa1_4^hcxDSzI~M0Ldw3u2=S>bkLadoai(4yBoI)IYRzHjxS;B@|-0O0=hc&7_hA6 zsZDJ`QGRY{`1V>^0noGj_M0WohBw0kj4zN27Z`*`7q>@-is0Vt8@jdmw-@f*X=cPOs5y_$setY`$st4=Fb_q&t6dGX)ojd6;^WGvorT^7C@;1=fP+u|~Z zrhtEmc!}DPQ9nHDg$KL#?h2JXj;(F%N-{4V*h@H`K zO>rszx^QH$mB@=P^{)Z}%!CyW_sXOfv_2NW2*wUDbJ{36j8;yyYaB%9>)kRGGFnxK z=W%J|?uFi`R?4KJOF8%ufGzM5d;U4iXu3g&lc<=hzt0EaKOg+huUoW_PZbIThgWJ@ zx4~2}U%iGQoQ$MTv7HwryWP>0{JMv2D&+M{$^zB4JRa9h3=k#WTl-7ytG4+MDQ8FR zvc`}#W=+QyZT|}=gB2+Gy8;>gn3?DCEJuBEOc&fz^-?fK#}A!c=X;Eh++v&!SO>>) zCsv}vFHiPRtTn4vrr?k;d^?!l@xMD=8%S))G+M`9`(&np9_nX_heZ=YsyCuK$Q6dI zi|^B$yz&BV#5)!J0b|U^I$KWW_9nuCng=sJWTemISky9GftS(^^jbrGV(AUFH zYomX~t_=3{fg9LK`8IDV>|G9JPs|g7CN2;)7mzr= zyzoU`ZIjw3dGZ$B9j;5Dxr(emjVGyUcdKj|ncM?ftD?G~F-cxxx+-)SS&B8QtXS@9 z;q}d{&Zl^oZzA^+*IP>lD0-!vowCdVdFik3X}!r)Rc-CC9fOnf?Ks1;#7%`QTGem9 zAUi#t)pv}yHVx~6)ifRWkVr6dzctJqRq`0THs!jd3BQl^&-!A4;7Bc;_l)R~X@r5k zNE4g&?i4XB^OTkV%2~D&o0lJ8CD|=We6x z8U9Ej%TmJMn95ltS9F*p*>lEA3|C*LGE09H8tl}KlI$TsdZ+w6mZ1m!;5PqmEWSIFom*GJs5bHGY8I=5_9Wjj+N4Ccb27*gGyI9O8TB3%=zAXoH zbuo7`rUhbq4d!nud?_!|zH89?3SA=x;4q_6lFP*hLJJ48m;xs5$M;3AS)<66=utc@ z(3Z$efX1@B-fS zeyimi+Ebc?P6Rprw^E^IX;%W}1ndn;V_SAWYkkjUG$dmZURWFAuzTS|ra93DAGR%pLfGo@VD=e{NSfyM<#YtGzz4y& zZ6f075r`nX4D!C9<1jo$vSW3U~N1+?uAPqEPYZdK{!iLv+0(JYg?^ldUhV z2F~lm!$BIJ$Ss^GkO4k%oDtR%78ztx_N#9y|Jnv|M4OWWp|4X3aUMw);L^&kN;Wo9 z>)#|N3hd@$du>3Y*}Pi}o6Uz?wkL9uMuBHFyb;4cSoFZpn475j{3V6YaT2P?28wW} z%C#p_4GXpO>h%0}P`2>LJgJhgO*tzDu#%~CSG!YLNh&`zji2zDGv#Yk`J$by%F;dB zU8t(MKcEqlD2W_(jWeA59FG+hEI@$hfQhL_G4qAZK2v*?qnUj0k8F{O4nZrr zr6g5=#042*jbh}oI}cYdaNnk`zCK|0Rq5ESxBV(N} zw9*|XNa}ZbvmJdzpB#MWj0f`dL16rN@OIE+a++yLPNy_<7MK)i&3e>Z`QOg4*$95V z0WGk~cA@xozI0M&or3raVV;q?8)8Pp?{G`)YJ}>CM z0h(c%At<{%V6U}MZ$+8dnJC<)co|JZ%v%a}4+*mVlqPJC=Xn0-N2Xy`@d5Smthbo4 zxi{UO$pX&r5u~5r(b9m7bCSpwG-<7%M?rPw85EmST%AVrDfUZTqTIq`D?e2 zIQ+EMsE!O2Gq(!B_5xB0ixV9bvJWyfoXh~-4h9A;7;0691}XVSM<#T=O$3mpcstW& z@l6(D;q-KN@T1>!UTIGYppgqa)NXQ6O9F5N+70v++%3(QntNYRo1mm^G zWF5ixM?8d!Qg9x)&7Tj9xCHKnm?^)Dpo#n3RWHOJkD4Vr)%}<OV&= z3htA{M`+l>iaUQO%!1r#>ZwS2W*Ta)ZF~PQV5+n|p)HL_HOm5N!yz$dsmTEPYHwU1 z;EFVxrBMe+qd^zj9jADc4gM_D$ktE`kT5vUeo73&N7iJ;-+L<);4f5Q3 z_!IgOTVpFUG3DwDLPA1!IWTqFeZ}B=D3Gi{?Gc&Kzrs1F<9v~6$nJt})M6-`#*v{8 z2>8ah;&Xr8=0C6a7Nub4sRGgtkKO=u1W$J?B~rPq8P12-8A(D$-0%^Phr(F&1mNtK z+I^eS1bpky!=UvYXzOAnnQ`UXdLbevEqQ@TJLv-jTu@>i@BCWrvnYO?*)!u*rcxS2wq0oR&pHjTPP7w#9lOk;v0G)e#))urF@gDH1C%ZF4 z^Rk%xqKoI~1Dc$+@naaYLUA8IaiAM5(&~D*w>?obrD^lIK7)({8B&1Q%wkt2d8BIX z7IOxc#k6j%J|CqqC^RAW=uX{*&Q&c0l_c`p!ho^Nr@2$T<{qQq7>ANnL8n_fjOQrs zr?8(4J|C1U)4-zU^B@*|d_65cZeR(an$iDWZZA$KKt~qN{;rtQxOebqOw3iSTK~Ib z8WV4gu%PWK@<>S(QDkZa2<~AK`kI#bO)iH73}5I2x-yn(+mFvFaea-rZeatNZMk`W zR|Z}L5ppU=^7l?;W0T&46(}-`D*RsJsg=j) z>Wi~=`}PbO$RhMoT*pED#PC>_AX=pXD)27-Gc}ex-n z*!Mk$BE1qQ z+xT@Zp$rO6&3qixsX3jz7<_waj(PG>Emw6loXd1{x$Mvp&+$*4T4$uhc z9_OWOj?;J@5nuO7vkmAC_H{+d-oN*>aung$H9G$hsF0Js>PD@-cyyogLl$AJY&L4| zkqqcG?4ImRqXO(Jj>?(aVT=Wv!H2*=P2_!s_^l_ek0J=q>JtUs5!;O&nMlbmPujp_ zZqX5ctj;9Nw7z@HEbGk1VmKcK9eIP1$B1WlBb5MlAAqn$zvnTC&5 zbUdSZ;Nd9uZl$MSy4HS~Cf|%W8fre0%NtgDo2bI8BY^=6-0{OBxh$%GblT)ZM37UT zA=~=(uqhJ!)s73L39J?SorwHRJ8l5J{u%P^dz5C$0L}0(;nCw^oH%d%&JNtTjQvw(fD+x70gbB#k zqaF$dh4QZ(+~+H|grgg6?}o;EH^Ji4zF1WpEwMU{WQJa!TCk&d&o*MHo9vuDgC3O=E!_v(cL#Y4^$~4s zWMVlG2y*IPAxBxy)tw6G=$1t~Ea<1s#)H+cj_gcp+c>RZIiu?P>-4zkF-5HH5yaA= zhqj-wZ$MLsNvx83%c&$Y(7;N8j7t9I0rvn;Sh?|O4`AB+m*HfBM|{zInWx(@4Ry*@ zwu^2UfDNoXt8X!aTuoVx$grn&V#zNE%eE=C&Nuj&o}GlfJtjU}$0)2gFoMgl%kX>8 zW+3L3hV@uT$?c}sktHnF-oFiC@yJ=VNaa(n;6Qd~G_)Hn%Q_08Mvc{MdB+y-OzHqa z<@MSCF|sMK^Tki%=9!p|?Lr}9$~Q&@$w~ot4uGbuNM70qv=m^kJ$L^7DF;*sa$xs_~Vwk|p)u zI%I%JW1&Ap1DWcB7Y7^rpz?UdO^{ygcK#voJHut#{auMrp;zs744lm` zSgbWGihCnNF1vE!c7UmqS&<5yp?T%2s6i+XR~;1=k#kK z`T4~W^LjRZYh5pb=THJ~lC=;a>n9Y)P7JZ<_{=`-Obh}fcVF9%+vPs*As_{}P&)zU0V@_Afqdds-t&T8!$O_Tl!Wp+;) z)iOxKVBIf%78#IbnpaqF$9>P=q;09QDmSx5mP>d#C7b>(I!~J2z>W1*NIPc^%)r1} z4DG_sEaa#pHk3GKz3h;V;d_BGgcj8f+3!?OmS82FMCIt1P9Frm0YyuYFU&Rhvk&d9 z%Dhdk=l8*Hh;@lYQkomQq>0Bmo_kIiRbHs99~#!A-#%JF}1>Ww1VA(uT1>c{>`RW^|c{iJSxhs)!o+b*7{1ugEMQo9>F-kbi+;9=` z(2x5Iy_JDSyi3I5Ged4aHpKBSeeNQR6Q1ATgz{>K<+-Dn^aUaEoHXn}mx4(Tn!ppR^Sow>9JL zNEduO{|sEUdEO!}hX4Pn1^l{dG|9=xJ0H{uK#tY?Ed9&Bu0UK9{^~A^;njfcB>s}2 z4RC;*az~|kjoKI3;Nb%im` zP6?N2L$K&S&`xj+_qTZ&!mqJ5$>B>Ce^Y_UTD(IifJyn_C4u1EH_V)Q0y{Hp(pYXy zV5y$D%Gwsf;J{;iC&I3ztGpsE$)>`KxNT+-XxGGWl<(3REsw78CZpPH=hgI|0vz0u%6}umBKqVn+_USyckkp2N5@a zQ9^NkBES!UoN9aEckpSJn>#Z_(7(ODcyhJx?F~DRr002@ZS%Dh+sg>br%T%9gHsIz z4-oXe%69OeJ$2amcG>m+vG>+dRjyy#sDKh8B`F;$C?z1B(gp&ebT`raXS|HsG(u?lJxtF@#{?7M)?-^&DKh8L3ynpRIaMd%P`OJI9bzM_lIO##h>LBC& z>r(PIQ(vtq6dj=>3?oly9CBVih{mGQs5xA+WS5ijiKX!K#}uluoULzJB}&ce%cHM@ zO1h|_lr*BW9pSkOIJc*4F(p)&&QgtOO&Kc|@8Q&LJSXb)PM5gpZ%Wd+1XifRq7Luz z)?~rV@68Lsn2~r=zq%y}9`xNiy96725n8dhA(xRK$_Y6nZy6=h@w46+5rwd(sFRW+ zrWil-Wo;H@QP>Mnh;8q4@ncCu=#@HnA$R^?gkJ6Y$nHXr)|yCmM|KXopfl0y=>$mC zI2l9l2DO3v1YAR&Ve$#PmNfJ!5dC9ksr2r50a5qoL?MD65(dXdVxb;k@;}tLEOXGE zti?K$c=xq;zY~(T)YtJN+Dq1$w+Sj^^5uM`4BOG6xk@YwK8p+&LVYtY`s?+Kgt&n| zqt{3Img(;}rM$!7c*qYk+zJC7*V3cu#IxaA2~+fO_wa_TrC2P}=+z+Zkcf8C>&;@o z`{&c7?EZ4@kKP`L5u+wr8H4siME#x7s6UAN*9%1c>1vuF-)*!bCA@PRsS@wwdB3^w z^vcO`?kEjp(7x@8&^|<}{npyaIK7@zXA~R7iA(KD{K1p!0U|zlU|!>smgP4sPOU%d zHdbfEYBpWrEin!@4B0Hn1r$i1TdXHcPV2TFvjb=a`r;UgWO833J#mn#OIXl%2{cmf zJRC9qTeo8^cnfi6(5IbG80$Gc$#&4DWG&IEMg(4}Rk^U~_kK{(cRU0~W`Qg##=M3R z@l3+W=RkvQrDo9Hn_uVbh8~t0zVNj%==K@q9HvLsltLW3+?kK5|BVEBzaoKQa$3Mz z?9WI*0v~$#$20!11a%HB`b@2N-LrGj&U9%4amB({Gg0W$9;RI;X)h(%kY^xNH`5^c zjzoMlOG8kI(fXU*S50NJNPP#wwOn6`+F!&Oj$&C*!*)iR3a|*BLOssloYe+Nma6Dhr;zbXPV&NM%T&$& z&u=i76K(^~c0mlSWtkWx{wf?RZ{vCy-EOWSwEbku^0P5BJtXh{zdA;Q*R}U_+!k;G z3gz5ELjp0UjB83kg3EYF9Qd%$=lyq(%vtY4Ld_z+C*O(9;52OCxhCbxl`|E;)5$y7 zerGso@^a3pk**|P&!~}yL#gk?y1hLTihR47pMy>xKOSBE>j{~9l|9z0`?A>aF6>V} z+zq5tIL6x803DI}_Fgrs)fvG6VtN++z9_u<(-kt%QTNj^=UMy#C_7f?M-*QjH;4pW z&sbf~&EHnc)1%yd7S$QQu4%9j1aly;QLzAGZWfm-SJgtPf)*t+4lwNj(-rG03| z&DKT3aUt!O$fD?Yl=Go@LbUHgH3t;L#kqsgxpZG@nD7gjev3ZJ#>RtP zM-n{t8Q(_Yz_4oLAYNt4kwlHT^SV&AiBTfvX!|Lf8=k~e471y;6FwEkw0}h$e`FYt z^4zP0;m3E-2@c(tC<|tSD zaMu%d6?U+3?z1@^Ddh(!_U#jtG(GVXWo(+=Xa|${7>oBh^jlSquHQ z#S+#i9|c`gx~wg4lMbrX&ObJ5KYz9I@g8pQvqgu;sgk>FEqsrDgCsa}qhVu)!_br||_8if@rrf!YrcAE({4D_W~Iv)<1iw~7MB}>KF z<@v{6*w*u}i1+qWm>Y-0`oL6QGlc!QjGwmnK?lwc*V@j6zy8dE4;~y2^mzn^kt*&Q zN!ZEm!Z^hsV!^iXQ5dMKaIwJRGSkt*YOBtEydv#^+tLXKm}5ouZ=xK_LD(_GZw7l| zku)1zF$>2pIzkD_bX&{ZH1wD%RA@dwloiY}Y{15H#c@Ro^`XDIK+$rGCXmEl#{}9f z9L|+r+`;3F`}5v4N+C<4(*(otI8 z9MunwX!Qrgw%a`yYsGe%tcE#s?GMlNpW|i`x$?HipS(LBXF^sd@K~ft@1LmF3a*zm zFJH9SuDIa8-RyjRau^C2cupHZj&kRCrzw1i%R=w$YS)d(&j!x!NSkLFEWV|hl825P z)pD5Uy2h_Li%h$p@_u5meYa(-{uaBWA)p>-x!yQ)c!Suyr{1r;eV#$qEwtM7rNDrn zQ_@`@CHyqgKM&Z?i~E=g#ubMg8Ki%%^%|a4AUMYopP*~i>%RHw)Vs$wg)fC8!cOF? zu_4mrvSL;hoeuUKJ+UE+D4k^=D3gszn&Numb=_)y^J!*E#`X7GrTSMxeID?-G>Lt~ z&>TmJs-JgaXW9Whra;{&c30_Jh zMnm;iH2r0Mk17ljnpcmSrNrewoNogY+@rQG7v9dJ-?zEzV`C(&erPV6w45U=&&s}s z>#AgHUG|yooDX-=euXLMhI!zeFI9X7otk_HzELTN182qMefpvE%X;nSw|!%p5-ax8 zOeB(zj&_4irLI+scnFdyoy6%-#BB);*;=j`vBn9OPk@ z=Q~}j+o&DaPO#7x92V>c{K$|uuQr7*!yVJHssvwvJV3^$;6G!$hZ|@+jzWXq{<^H6 zm1pBUkRzF)(tthE8QL>m5-#Mw&sO8mwf=d!{_LlinP%U8ieX~q54*~fcrG_ovfpG0 zL34gLkXh{GEc%NckPF&_@L!^Xr4NmH^JO1?>CDj4dcGqjMx#~l^2nlxA(N?uzP0*M zBCZ>>-n+%qAF(R|x$b?0^O zi}~oxl!AWr*o4K&crd2a7|t)k--t7JUT?k7OUdV)3l7=jVE@!>PsP{$;!?fEXcx^y6_by{3jd7_@m88b4yj9^_*$=v;rHQz9$$?m%I#S)p8H1i{IpJ+729HUmmfMp z>>?gel5wkX!L=u{EBa$2{A*);wH#%F1T1;GO=mT17XrD4_5tRL6R-OI@8lE&Uc?PJ z1Na+FVU8S-U;XI>#DaKSddCOmeB>w6J+gJmqsnD@6Y28YIP`&W4K%3dJx6AP93>I3-hWm9nuUam5>cVy=)@U_fMyNT|!Y9^BgDoi;lm0 z+{)2XE11PBJxo5bQ4BOu59MDpU4)uTLr!3L7f0uBAiEd7AG9hJIT^Itm8wE-1=3#$ z_;$w@ii|MD1I{{>RxI{Q*q;X$5u5@Deg`Lt^SNLWapr@cx8*ZCqm&n0#^inXZJLjo z-A*f<0(R3J2?57BMBjP$1$M;?d+#?ALT{JCc3a8i$WqY*PCLleOdIbc%2Luvb(tNf zaSRPXALr4vHm$Rt0tiVnrnAw+{-=;L{+LIBr>7n! zu=owwHbZ_AAI&s6uqrCmd(5xXVJIc5dztg|;ibRcEyIS~n&!SxZ03QogotsT&P5!{ z*&*A4Hia>tN@gkBv`TeBS!FF#p~ExInmRmS)z`sq1S=)WXk7T60obARQ)lR3$0%@> zF+_DXX_oEt*Ps1vKZ?YqL{rRqLdWB{KtQz^uh2maME4XI>#pXI2w2ulqy351xC;Dx z;yF~7yG0Kz8=hhz5N=&s^-edal*Nw`JsD7iI>LiuS?8y!_B}+`y>9I7mjJv5G8`yh zJ{D;bq})-dQ_6#%A5&2)&3Fm0f`q!rYKAE#l#&j$HHG>__Gi-*~!a0IkDZ(WCG6=w{**pd8Ls^m!ZIy?X$5lb`{%=*k+urcn{;@PX4nb!--l?rEINuVRe*1+a?2I#xEq-UOo%zQv(L@Gxod1#Pcy<=yl z4lMvRoVq^$@bu=Kut!|6t_XNbKAE19jr(F-P)611-52+|{>kld+t7cXTRO4lp{{a4 zJZ6X8!%h_!@q$`JHO3f8Ju~j>shmh)OUGRS+qa)sH#8-C2t!wo8Al0r>26NAvVZnp zf78Xk8GfOTj-%j@wf6W?;x#2Jn*bBg27?;OwR=k0}hDhLogw9ZaPJgBptAt0GWJege2Wxz1mF1OZ z5x@~@?le~^ybvL$vj2sYdhKFz&qA|CKl&1Aio`v#W@l2b9ePN``cNt78oaKWj5J^9 z1sxwpYjr@6=!qWv%a+YkqM6hEWv>cmQrnxAW0F4ep7@T{qAFU_$F8KxCAwGw&PPeB z_kcR?{To*%)2o6k}|_avbiK{c0z%4nAs zC9hqu_-Jd0k>*r z49%!j}L1igVY#xTm`Ez5O;OlblOocS``#9|H>x`s@`Xz z>v%u2=zq>vPgwx-x(d9J`R|EKJEw?9gNSG@#!FY{I#dr&;(%r?hU++Wpv2;=8jtImB8T|)aD76juR2MGtol6+Rm(4fxr#NvLo{GQc>6$XBUs~qhg!TsT+{^LQ9M|VWay8m?YP`mGTz2 zh@q)|piX>nWX^xm6iQKZayUNN5tq#!@#?-0jFwL9?44ZsGlR|qPRx&1buUb`0tgw2 zj4|dx28_*ZG2_+We=L{#(si3w;dAgKry=x|Pyj3hJ+b&KHXDZ>H-^tz9fhR;Z3g({ zUgW$bE#+AUT%Pw{omKlG1AH2OK2+nY@ zHk>+hECqLFSm+zxEZYZu$#=0hDodfKs=B8~8qWLKCl$2q-Ean(-}itz=XSWnP02_7 zbe|uk*TK#n>0U&b_c+CW3#Rq{+A(wD<3V2%)-pq49vxU5WaID%CQg=qC7C?zDp{<1 zm_zv0nJPvx7IQw!Sc~YjZVLa`_Z8}bQm9A;xq3mthjJcwYEAfe_zF1HjpWzlW;u@9 zEx2}z>gxF(a-6$n-@8Qwam31T_lxXU?>UQf#n9AMcXADMO5y1ip8Rr$ zXRuy#*0co0w!1Sn#%6@>I{Sv=YN*;*)1Q8a9=$?{mfdz4bSY zqUBx75K>_?V6Mq5%}c+gNO!nSqUi0CfE}@@qi`jDVfWn56Cft*{n9*vkqfLh@$g7{ za#2mLrxM1N>HQ~>M1L+q^uULCICU{EPPD=3cnJ7Ts1Nn_T9r|4js$Cjx!Z8m(&-3k zkW=mYnNvY;8OzY3jnV9o`4?YLCz9$(#c3tyX3nzSD7NHiV-+G1U9IVFoB~WeERAo% zWCU5)Z3IURlkgpH;JIzu&6*sqWaC= zO8V~2OD5__frD%okna4cwyxn7;ZoBG!0%r5`bu! zJVkG?87BQPXnBpk$hb+JK{F3s7W$JWQ-W^&yAmOS8$?vGJrkg0R(tI2;YKuYy4>9$ zmYE~GcR@K_Ba(%^1yObB6g{Ey7p0ciTT1U-C)cEL0l-}CE2JPfr=0?GFw1vPgzqkZ zfmb-A<&%^vy~DSUbdn5$b#4j9(Tqsw6ErtCUHVB_YoYmvrMwQ1K_^8`y%!8E1enOF z6Kn1Dv55D{Ny>E=bY`=J&86$3U$70Ht8l4watSQBH=>`Mf5WF_b31Dtc(&Uoc&@XY zpzSvONynT)zO3k#U*N!ER`akax1(l%01`~%12ozO(7a$rCTs7@iyW?WX8Oh=P(Tuo zzkhH@R;fKS_CP@6lbmMzHUSm1474Rhg@`uC9dKjv%fvf&rt9eRrLP1lvvB^HvN6Gf z=~WSn_13fJ>YEAl&>{A>7l!zbcO+z27K4AZW!q0LOHqc6s77~({P0#JlRO#T5Mw^F zSyiv5jrq<`Nw+zC`5JS?rzqbkdM}~(o37zwD|&@q>>r(vORp)JlFT?!y)W)Em8#qB zZ=dX7<@xGlA1x~`9M*cc-+}~M`J1C=p_#VFi#<(3!et31ZLF00`%Ul_e z(J}P=&*gcbi?bkrXX^VBb^n&Pv}{N;5UxmE0Q}~cD~VbTTXb`y_~=ahy!F^GB(Z6p zu7es(6N`*i6@8r+KQlwb?RedHKNvg$%ms^5`@@@^k^EV@VHh+%vWX)T#4+EbAU-gg z){{?98$2RAn7kn44briN%F%)8g)Nn{pZr{%9o}!E^9R}K7}B@|{}@P9mhZlge*JJ; zgFyfS!?l>LLED+}d$>7KVRq|uf8`YdObd6Z`e+{1{wzL;@gw@9q)4Tf5`&^hBt) zvLD)jDI0`f2*Qrmnk^ezTECP+IPw!NulwFyT3#F8i2vbkgnsvB5_+}WTP$y>;KfrU z*J{xO%ZB;2Z-Xn%RhSJ(HmO36kHw9Q`2KkaaXpKEHK^#!paPL}$AY!YIc=}!4~Yj3 zhntBlNm2!7RTt&Mcm-^=xppI+%(j1|(odj6lLD&O-E!JG-|8f3`<_Aq7&@9$8Hh&> z`4d$3afJUt;AbfSW(QAHn(I4tX4qctS>6g*3Ie}H(G9y_c&4II9AiX>!`B* zoKkALX~FNv@%~4pv4R^sw}H$j#3jbnB%eb{*rBstApS_a0d!*VdX{*J;X@muOZ>JK zcCx$*qY7MoeIZUvs-{zlsmbOoimpm~840*A+xJ#i{F6;QG(kK zv)?Co!vG|1@fv_D_7Q>$Mpwl(Xwv{FW@jPJ_9T{;d-6>_yKU6ON;B{47`Gea2Ro<@ zgGt^C-uJ<+Ckgsq2bmYhWk01A5QPMPmM`keVpw4CtF!~zdF88jF^Amp>Ec~>;&(yQ zu@raa4!|05c_BO6s>OGSp6nSs7oAQJR*nt|Vu7Q~x9t;Ouq$rH(en4G5KZS!l5soU zSFd*Q9gTl%$fFjv3pt~n>@U{deVAe~@KuQ5LDFnjAv2h*oD-0e-rV9-n@RA9m{ogx z8dyp8=rJzf$Dn6EBk%+YWuC?jZ!>LRdgbq@B$o`nZK6zHPADJ7viva)C`EoYH5V1a z1QI`-hccT6W|TX9BJ%qv0}A*S$DL2YHwe#bsA)5u$QDe8TfZXNd@pz8HR97Nw`Edw9bH&wBTMb(7a=^{>pkzN-V1$6 zwOdGey_!*Xa;IaMecQx{d_@R!mQj1v$`?nwV_AGeZwN~ugMHViQDC+nd*eo^2T)=Z z0sXo8wjg4;s#iBo9J^W{*{)#F)J_#Xq9}?iwmrWuoyw9-fq6_EXom+roFrdh-Uy?K zUW`v9q9(wQ1>aT?eYQe@6>>H2tv;+%tr(Ic_eK3d+xAf5s`lBY??tsq2A^E5G))ST zdt(#)YVyHNc8|D=aXq9+Jha1w?u>xP{-%13b&!-h757_Q0i%eG8nVN4&X_hd@u+jlS6Mz!E}cq*Zwq+2c}waL$H~>-$L}FtHaVmIbJ{omuCR*)_g{og zIhe_ZNk3)A9k@N@WJs`WB4L=}ALQ($7p!DS9;wN`ZkMYx_GvhU-{oz&o_IgIIm zI-5zk`E`Eg2FcZU2@p?)QaXo>v6sTo#qT1OPWiK=oTfyj8tvKM-fDAwE6Ze7_`+iF zmdP$Z%VyBXu}9AhbA2vIi2n#=xL#j)ojoWv;5=$|Ohb4P%oyk}-u8Y`K6jdoaE&+L zwq^=lk<_d)F0{kZuCHs6Cp%iNR$Va%<>~X-7LO-l<~OH7M4=pQOLD~2Mv!!CP4tYX zI&VMf7(Ns|{PR2@>hv;1M?4X8`LD5MqZOh$=c2jT7NqIVk#QE9F?deY2*e0=wq1fg zo&Zi4=&vy`68Q$5mIRF8d{iBN)<0fI7nLg)9UDf@MnF20t7i~IoPyh&OV7j!Y8<>`3$?^nw=b(?ejilt>!bnyjBFr_qDD-6cp{|s z%Wei*i)C!Vqxo}o5qWsoQ%_v)I>9GIFxfl8w?k6}ZUn{VFWdcY0pn?8a{Nj8KGJIY zNH?KVZb(X*=Z%KpeRd;Kk9&|ee4Brj%>}$+L{{EIQLR?D|wgqQlN6014(F+^}f9fUXBh5 zy_9&P`1KwesxIjzqx2KX@cn*8bWa1FLfHI^_#)0h6+xo>9b^@zFjr!z+l%g53`o9E zc;TiHk)rMu+hL?YQCN8@n0X(3id2i-*k zu4eehe=qT8B5RclLV$*m6z5;3$;Ti>UMqus1pq&yp;Iy0V2#mP)7e~z)&2_eZw3QR z4~ah()s1~3rUg8~vy`>Fv;d9LRFHnl7EC@yI zx7xl+!Ee&+38`b(F>NosaHVdiJytMPx2clG^c1kzEVdJG516ZHPp8+?e6H)d_JvBm zV!g`24&Ec+=Ik2T-OAAzT@`)_c?$OP=OHgW*cet8~#}82c&|@N*hq z%dy{*?D4>{32^Eys|>1-iRJ^a{{(#x^sf=X4W`j!)&DA=MG*8gFiO{>0=x{2m#J_< zIdS;O?riO(o;PxuV(o9BQr_=)C|BiT zB9jE9%zcPiwP5VQs8q=kim^Dn$gYEuc5|J@gI3GycGBUGkY11rigSFg1^g*_*V_@x zL4@CMI%HFQD+_0vWe?wqtbJ&UD!{=c#{rB9G>~L^auBmLZ^T+QRzh@)hdx&5iedaQ z0yPs7o%l`36R^q*y42~~L3D`DIYg_inu`}_!(c#ebEw>suCM~NxBj2<$To&kA=4Pa|UsX&26*V&8=UGjCM3NW2m1XGpJ1v zKRb)@Asx>3E$#>w-kJ3WfWYRa7O5zBrc5F#=%=FtwJ?7$Ey;)<-UUJ`fW5I+1X;Uz zkLp}89m1Ba)xbjB&Ugd+HVP`1C&1W?Oiz?IHYhnTn%23tG?>f8J-%@SfGWA81-4q9 zRgeMN{wxPp>68mb#w#pRE(80~j}C>Da#{8=x$LzkKpQ0vimXM*sz`7@xb*sAL`a3% zB=%r@FVMJzaKV}T0e~6v(;EyxrlC)KxJH<(*wPhk3i(Le)5_Zg_6VOGZ({S>uE=dN z{M}omLkt?F6PAMhbTX)s?SD=%d|@Zu1%MW)dIG8C~YAZ(@V zE`791yjG%9=+-j#P57#j`-?(za*g|5`pspRw3k?0T3%p0MmC(~7!=bqpOHn4ioexx z1@k;MI)Gi1K9jheIhk0wV`!;6rfV>msb|`m!WsGBx_UCAu#bgsB$A?E56U5ZTW>(%Ms|z!GL&$CS=6mYnv+Wg>?v{W?w4cYLn@Rtid1H?`NbWb z%YPg}x{zK??YwyJZ zuuQe(Z^XHZOr95~dtEAXv)tDd3pI0>zBtMzJR)+-R+M;qAi0l;GGJ1bMw=2Bzj&S4*fomZf86~XodmUKR_ zF}ci~;b<)Z-K)35Ip13QzNCTXI1?&C_wo8P)kuLFv{3$*7NT>WD$MGljl2XD7!XuD z$Q`#}GJJGSQ=bq>Aqj*5^q^i7HPfvpGtpi=Bf3M0LlCms6HQTK^bI}r-V5|f<+6jV znxn>=%CuSm=L)9&OtI%Fitc7p6_{S`#WQn{Y^1eLy4o4}0sCX15YLtGaDz8a$T{tM z@7>MA0zFApOVZPQ==mek3FXi<4%AAkCUYa16Gre0zh*PNs#H-T0>- zj3e0W`I_ZkCKBltBffa2x*XTe-oL*!1SP8_-Yp`uOCok^@{ivZ^Hd-z?S}GNr{3hZ zCW6XTrBuIAm+~uhB9B-*XR37>)*Ic+g7ZFg-FZ7a&=vGdeR;S0I3gj_j2O>E_MjKv zJbTgf{_a9}G1w$b&9Z)WkJ_a*6U-gO+8sdNHy~6LJSRp9?|t zyNz$Y>%1jKzH^7B5Z8-F?n}5_23Ht)D&S9(j5Xi@@tVXNEmdfM((47t1RwA8BnjQ< zie>B`JQ;{S#hm$i2eu;t6e)aP#Jec>C$AQnbSwUtw4|<#oDB_LG@k5}4dkIB;6Z+0 zDh(qE1j{9onY=EQHM1m2??0mc$STW!R&>v2bduY#6e$${>b}4Gd2^*;rSWu%ZSI>= z)L<;qFYms+p}E+_r%ywW$Z;wL>ov%Zh*oyy1)MXsvKy%>X(%POfs_I!U}1$$Y5;}p zt~c3TgJ6e~PiWrDJmuoKvkjMWWI;zI@7cD}RU0D4;8F@-*G z6jXnc#TCwO#yiDxh-n}wZ-B_J90@;fhU?^rj^>jGvCizwz^RFw>Gqd%N1C>A&uwGb z-`Jr47`_2^zGd*rTwUxFI31PN8WLjF!})I9TAO`7NW`%P=t}~%uFYEE4x1x58oL?6 zmh#4O+?$NjH77Ve#8Bo4v+tRg*k?!UOga7We=e-($uIXcdk5Vbx0!7jRMy8o{rb>Q zR5HRmZ?Cp%KmXpdn>fT($(@}~gQw$9hxKw-Zbyu!&yJy~9DLc1Wb*1@H;T)4ja!=l zNW-RCO6R5fo59+~?NyPZu|$}bR1F-+>NHyJsmj__l~puz#{tNP(;JAMS%K4=k4Rs6=)@20TmDW*{NH*l93&je?+byUx3 z&24Yhg!7i-L@}!Q8Bqz^laoF+>l#!=`nXids3qV|y3i3sKQ%4@1~BrC1(lYxB7~Bk zQu&<|-Z+z>uXjf1{T$9)S3mTTLhVH6CS%fg41K5z_eOJn5>55C5kU~8I_KAQI64*= z05aO>V*Y4*2j#P@BEckVU|O|wUJRSu?_M`^VUgLM0_frA6i;| zh?wyhB_rl$n3O|?K|iK-1|Nf#MVO%HEI1QTg7*TCc+Pr5R=?|^}Bo$Ccptg7x?wvJHo z;ex%uA9mU?=WFsl&I53XgcQJ80{LU4$%1Z|EBd0)5Q7nu;e0k7JA%yhevjN7Gy+ zx@Yk!EG8Q1zUrv}HBv~_(Mm6AOQVOZr;&IaU=_s4IJ-GU2NvnM5$MF~yLVLg!c%UXfF-BBTEZk#4SDB55}zS^P+?qF3w1c=OEvk{Wv7&vV8*~jQ=7%x zxs`Tuff2MB+VDD*NRZSe{_&mj<$}e|hb1@}gNB=t1A2G=V6AD%kZ3G0!m``Ysqosc zd`aKV=W@6{*-gipj8m$6=4jL-V??*o6CLVnl>ha6N`xIbhI@01#5AE|AQUoGMl-C!x#51w zWZBo~vgC^f5BA>o@gIRc>+2ydB*zdk%V%t&RggTgos~+fVZJ}pD0XT4AtC6_cBXVu z5`_2$KL@+Q?EbcXPWgPd#+C$-TnWi-cv5COkk>Edc0sa7$YiJzwxPs zR^V30_e}L}!SvU(>v9tnyd*9z?0e}drK`=^1A4x92p%e5dr!pN19h}n^#Y?}MuWH? z6Z!PlEvI_+%O;w<2pN_3_kOFY?UvOc*xYrl{B8?(hfMpGO>VZGWj>46_d2}&xZ%4kPZ@$z2{3mcBQdYBr^M=c-jyCH{i-GJaDam@PbQWm%n)bTHb zHY?`KkcY0v*8yq^s#Z^s z&^0$C^qnT8xd5vJKAJlVw^e4e{svD@ZH7%&>gdmrfZA9o?rroqnL8TJEe`|{@pCZy zUyW8=O)#p#VSSSGkJc3ek57;BQ^F7;!5(Y_lAItlUH^ac1 zRI9U54y%RM2cS8_nk00Y%i#5~5f4$KyG{S9Cvdt(z@8X{ra-0S?}U2^+`gAg6kYvJX?bMEh2egu*L_MPVyFGq(2T- z6X;JM`wP~`T>ZU@}3X^QRH-bdwY3xjYFmNFfg3i)NqWXw(ka1Cl^Q#tjCS?gq3P20J zflxB)z2N0QFu8Sio-F`?oQJx53YWK%Ry70++0;7;m0tV*bp&uB`X%>C{{uu*{L5(V zHX)lcKmAod_366s|KhJeW9TV0JBC#VIrk|ovLUU=aaugkI@S&I2p|aM0+`-_SBX@c z)@khf_%JYK0E#8l+H<(r@YVs8aIn8K_vu)fBjbtP8aBr5yB;8+RE?hZWv03K!M3Q~ z-oqRIc#XJdE!Hla*3l-nn%iQ^=dg*)Q8Q7ocOGmW_EvMhK$@)B^7z4|>bRxKITv*I zY2NxRGNGF*%*w6l)mey%_+AN6$$3FN8=4i%s)6o%lc*@iJAE5OWn@Y5!dG9A^}Ckz z-W1rSAXp$#qeJ@$8pG zVDb9QKaxYLH2@}*0W@`5Kx@G9)5&(Eq26!bFb2`~?ZH!zY3$YuqXqsW^JGP2p7$h={`fhd@G9N3c^Xo5!*ehWv&6(CcM%%~hgZW_*w-|(}V#Y<9KTKDNBLstYzd0o(l@)a<3AIbBFW;$FU)nXIn_`=Az%{}{i z=qZI23sJWw%RO(c4K{&Qz0`3S8|OfxrCar}DKZN3+1^2CBXyjhk&vwHl_Pr;v%*c? z_F#-ctp>lw0X?1hh0NN`gTM?knY$}%piM0b#%qrHZUjSU2OGWnYyXZPsF9rzqqYCU zQhNmbdW2Yz55B+ssSWp>KF;rN_UJ!x=dareVws3HLYEi;G$CMm)%^BTw93*!YSxP- z$AX*l#e2ni^dH???kmu6eElKUGtVgWo8(AfA4`;t8GShjQ~8yaHbV}1VO}TVfw#dr^h($V zDqBMty@9dkfzKOHvlVve>z|6~(`EtKHmn=KG#^IHu0j);Uz|9$_l!Ss>Vf16oYJa| z^c3k2AWQe8H!tpXY)#5#s>-x+ugtgmfKa9Vfr^5^%sXYeCg3>_fVVt08Tp{K2m#{4 zX6$me{-b$Ij|fIj<@Eni%?1AJ3HwE}^^g7JFiebKPR?)^J|_LI&G)yzAE84MaVa~T zVEsBIz1M$`yrBpKYaN?=OcWqa-6Q7xK>Q-C!VQ3rQ|R3R|Miy5GIM7s7iysaUSDZv z+tl%d9a)^m8E)7L%S}_DwQ<2$QF^;9h~3No~! zNzCl&iiWK^+#M^-`fg{TBcrwZ=;-@z9SqUPcHyS*Co$6N&6<|8EO2`aR2;H2aT{W{ zM+=vYNSBU7_7U`bK+6N=4DFUzi&SVJQzO>37JqE7$gGzwe+ha#AP!t+RB9>EaMpp& z)X@e(HM(PZKi&&M4HI=aZInK4k-FYoAV{O=1tI8lINCr5S|ky@jy|X0f?(y}Ym^*} zJ7#D_(EbYN>NFq?RLYj8_{*wa5yDx$v_6O_;fX1}HG_?+sd2h`Qn}f|%D4KZvKT6P zmOfQ3r+s}b8`bVG24W*-vO33R;sVSUM$(ZJpD}e9`?#>|2s#n3#r&^cTf)klzbq9M zR2ma9BQPGDG_pt6)uz%dX_>P5{H5tB-tt18K?b+%(Y>tlivSzzk&=N#`tuKis&3*{ zJf`D6dLv!GnJ^3gU;d>3`I`TK8?Bs51>F+*QUf|rj_{jd{9k^9<63tGIKp)<7}kqz z-pXux6Yppy2#pf`ULj4)R5=r;7rVdu{0xIluQmKx{?*F8i+aS`hvNoe8l7ez*szk; z!z18Y(OzK~!MIoJ!;CDHVPxC|)yn&-2!{+Vz5s*i1lPfmj@tg?o%xwMXIw$mcL?bc zTp&cks8A&n;nqPrKU-x>j8HH6)^PT)XBN#jIXWYZ7$}_}i1&Tl-K>!h?8?p9a9jmX zHdrCOUqP976ChvAyQ3z-&p};i4_WLll%dZ-c4V?qH2}R%=u8co+H9(UE2HJhB@a(> z^zfm(C^X3!Y1%i`d>_8#7kr`SCw472U~5P?kKsKL6OywoNw`?r%OnI8y|Yoj4Hn-Y zSG&)aAI)q`<5*jLSC83OwUD(7@Qz$5NgB0r&JiO|dK|l48#Aduu`u_kifOfcDrFrP zjlS1_w{ZFBwLcG!2}`#lb7dA^YtvI(l|Iud!C@NgmkHfAn%!Gpu)#-0nW9a2`e&hB zQeU*}s~4Cet-^5op1+kdlip)DRH7F znB?>vo5!TJ{L1#KPOHP|hpL&4$Vux_!{z9UgNv;V*%Yjkjy~LXp-Uc=wl%A=P^w-r zXljbE8_QSM&0E&YT)fmSx`;m{1f~cJy6>4a3fudAw7rXV5H|@}+i+t_C-}j|u5Htn zo>72l@i(M-d}r{70r25#r;$b|@OOSEX!8TSA5ffnwh*827zgpTld8F+Vvg4Gu_X^! z;Lp#w?*z|-rq&v*SE6h+POpgwqq{qtP2_Be@b_bvq5S;M)ButY?Be5v#0ftO>_clY zm5X}GROT|bf%}Rml}|%Y^@Ff0_OS7C*NfGwbnoM3&G9Yw^M*sQ-WoywJlD6aH7tv33P#P~%?DrbcB+5<+lJjhmSpV)dy0_V{TKP8 z5~-xS1>KOdq4-DTn#BL}rvEX-{{xRWA#>a9Js(o4TQG=H)?IPZ`0(F3?Se+^?!gxh zows!?XF~xF8tsEweLi0ftA8l`H=h3g#+$Z3v5ta(a{M!%UaosOHyou@on&v6gP>_={}DvNK8p&)e?w*&3K_*53iNo7wRv^<#ca52(%#EUKijV3 zsj@ed4G)s`s4_3L&QB7LE<16Woo9!_h7-pAjR7$!vTybCkv+PO$}R|a!+A}6gZ_|rwlJtfnhADMN2 z3^z*o&yNiIe&GnMCGH!w0RG`T>FI_GpE;_j0fzp4A~n)Yd7{k&Z*f0cdBavOkD>wP zF1#aVduJ2A>Yf%@)sHs6#w;y6?T~n4AkSF_dHF4iE3Wn;IIZfI{ne4y{`XVM zrn(#N2W@7z#18hf-iYO1T3)$2#W9D+om=2HFy;6u0&}jGwwQUgPW}EMgfUYo3^DJn z)K9ju2IZs;15=)EK2oD*JD$Gprt%sq9~cs9TqB4>02XxK4$L?skpLMM*ZPDllOx6ub1s!_PR)-R{GTnia85J+`cRxxapb?f;#h zP#a1bJ6(DPo+_Rl#6dn2r23E55XPGt{&_H763-vX5HmLH5(~R3a9IygOYscbCYvgq znG^=@SHCY@uX@Qf5$!f%x=_o)S-7|>>y8K4+!o&Aw079om35X;KNjBcILol3*ylcU zTIh2SZpf;>YNNXA6jtO|uq1n#;3Q~0g#XGmXO(ZXE9GEN6MxmN#(k$@Sl~3N(@;_I z%w*(#-^MH31f$V!l8nyZ!AB9JN{}*3=_5Y18?~`%IOG!`= zJi0NUY3IJq^Q!dxc(p-@II=g7fvY!Y;hSvI=x4)~Dpa!n?WJnDEA21Y(k7~?tYxVr z@)SPrHA&H+C;M&6E+cu{9Dh`CMt6kS$q#*i*{k}>OJoO(+%AFNe1&5lc*90#2Nli! z3C|X0XrZpQAJN?28@}y-;5eFDrFCxUiVLr%oP1#SdG*`=)RAUBq;QqAvMzLEUSor6 z_kq!S$AZa%^@GsS54h`RLoUTi&tRc-m#>yLf6p0`gqOupj!KiGQ{ zsHV>LU-WDFl(q`dQpEv?)&VU=N&#iaprE3nM5RInOccsI6NUs5Qj3a$5*3v>5fLLI zMCK_F6)+-05Rfp3IY59ALI@$lK0DaTFe~r~gAWft128^ScBQIG5i7Lspux0}->oG?fNM55`NP3PSq$6?%iE zfm+niXDPg@+1}hoegjL4O#-p&Xc;F2&hQ)9ix@v`%nt7>bui>q&EgFP`IYYi=%Z0M zPSqL0hg3I}c$#xTc%R6T!NK0?L$k#51>xz-4@??hoB)n}sVl#X_g`sLU3FK$h(qgl zBG>XuR-Z!i`zYljH59-8CCh@i$o#%~TE_l~%E0P~7vs+5@SuihIHzJZ-1T1#``>89 z+82&2JJrhfG(p<$so6*0bg#6}kOgL=?uhijQFU6KGyRfa&2*jlUx*Vy8>@WhhX;qZrhRoz)W%^!j76I^AsMj!lGMLdE1y1H=&;~g zSG&VFxL0lVJF7ixGjtO;FJ~XFPG$)_qi(M{>;hHl0=g4mcehkr@QTm8&)c@N=GN5? zJ#Sj~etdz@S+$txi>k?5AsE&K(gU|2p4t%sh%B3dAJ8^K-*E%{olK zqK&FntdLlU%Dq#DGqF^`rQ5th@{_;d63*Kyn92Qput`CKG*dJF`ByH0x7;R$+3rfr z3QhC;L=N*sQq`gpzR?T4lFLF}TA56|0$%)p%9L|zhLe{TP%(`Ojy{p|4w(+hAW`1L z_!`&J)(ax=SkQDWB|EiLSI9oG@6C~fbnb7+BDMii)Xtn^yiAtmP^09XIpn!5p6Hc4 z@IU4UJA|km1&^6-w~btHVx+hL{)ZXgL&Zh`?RoNOh6C;e%QaFa+1HdGaT+h#1P6{72B^BTR4JZ+=rYRgEZUe|Qb&bh`6n zR;DPbk<0(Ah7;Q$rV8hU-f19+Q)G@WlG>$Qp0&?HUWId^bAsG2b*^ zORe0z;aA9rQV#icWrsX_c^MygjVLD%$|X{y;4qIVqWk<>Q)aTCF?V&O@Q8WnHOL5` zilG){@)udd+j{C>8@6I)%Mjo}xuiw53!Nv56v{RU?)k5%2Zm7kR66C$W&hXfq5-i| zRP2$Kl(BK~F-OZ;SDIbn#ALJ zHF&9^yQO?Sood*TF?Rd0I+3!9ZQ?2~Z%)y`WwE=u)Q7QMFgCxj%MHc0I;x3j@bk#D zS(F<{AAjGyqaW`m2?CK07W31(7}Dh6$7uye0*y&rx&40cTCkE$&3Fo{!7p4{5UDJp z(iXYY!cxr6Wa+6PRd(5dDjL$l`+xB_Gq4!0!?)sB?$7Y5y zmmcb5b!9V{6V_xZ^DCs-b7a!gSnhCFb+sQrC=oy=P-nC+%;>aN-M@RpQN~r8SakTG zcAXZY-Z)H4{V7k79vD+vCtu0)s@t!4QcfG%FYW~qLh)11Glz-s2bDJ1Q#GJUPZx1y zbPz6b9neO&0E;!KTn>j?t6UV*=t!cXl1WDf$-S6WfO7N#9oM5*sc*rvXcD51gvxL0 z2gRTHtH9FCzY8QPIv9)cY9{Zb3kw^Zb$x*4?cD>I|q zF25TOYS>Gh@E_MwqeZzCo+J&(=u7*6%aI`JB)y_i(U<;$o%63FhJFY_h%XnUCEuy= z&CHv)#8R%jE?(k5OsSJq)1x~eo#7R;95Z3CwIQ(cI$;C z`~&e7mAnaq(}E(&jZhlWk@s=%DUZq;BNvkaLRXE^;tMv3jcIA<^832@ZB4yS!1&{x zw$7N&84!}1YP%HIzdXTW=ro}FvFQbwlaz>@-VS+F^N4sSvoX) z&Do{2?Nprd*Tj(@*{H;*0j5as|Lq2e8wp65ckQU@r?L*rD4~^b>@=$er!0l2AYrOS z%RkG`CP-6eYKCT-95vIL<7l;K{qi{%ipU0Dy!99nF z1(AKw<@n^|#)ibCtkRr;bt4JNy2*gG{g}BMZzPPAmD;EB1;XB*7|^49T2Y zjt>8~#S!+J8;1Ot(EFk^K7TpSwaGF6f`0=_Cc5es^Efpv-#M>4-YKA=WcfxDapCY3 zj1n`wpng0xT{`6eX~wPOR-cqH=QNm4a%Nct{F14dv8!q z92;I!X6oEn$yKm{qLVVxa#kAKwiP1!k}IuVR^!j;62G+<9Yelx5Kn?z4Hq1WmYa^D zb4{K%o=KYti}Pb2P%asd1jz{T>)SMHRlkYkgM2Yp-4mwh3}KQc{VE<+Vjnc@{$H0vD^ZX9$t!#`C}^z8QvqB37?>$ z;6IH{(F$Y|1(}`Em!_LP)}P$6{%h2y>|IVVy(1+A=3+HT#_kVOMgil}A?0?Uhd#&t zR=MX2((0*u;;ZY7W-4C5yHj+##4tAT3uVWSq}d~jT*|03L9Vw0K!Cn*KO}T zyjp)x?=*I^Ed*7ds*W_~&z@|8W9iK(YSiU-9jHqZCPjuKdg95$ufgF+irm`2IzCM^ z2@!c+Ka`E+;41##5C79SGA1*P^-V32FwZz>$$8%)%3e#{geZx?k~Tb33HR+55Fxv=U6r-|`K+SaY)_ z9?c5(K1}w|OWR~$2}6&24`;VH&p+nn93+Mw9$6K8Dap^|0N2c03ywM%;(rAS48wjg zDw*7fyUb?Bb9@1|*8}t(O62V2jmbP>nC`YV_tZGg>kbD*1v-X=l8fu~78#w-P?{%P zYdJeL1;yjWBi5X=heAG#3nKso(0q^GxZmU`+P6beRdmaJU8?P$b-ciEZgQ``UXT?y@}OyS zR!LN*?xH3+c-{ny7leCgu{QrqE-G#NU42QOhs$pzVMr;s%y!Zds3(eJ` zCG#IW4RLLv^XUkS9U)S5oH>K*@^76RBoT<%I?{}cpKZp#j-nf+o0r|iKghqn7WRq< zDk9;^pe{b!NZwALp>kBfOqFH9%08f^FDZ33(e^1F9%~o1*|EL4uV0 zWIiw&yrcJPha}~L(SC3uT%tz!wxftENqU;iIueJZT0i~h_PVPM=*s_|OK+WSca+J5 zaruRzUy$M22a)Gl*nyZ%2}hms38E}wIU(* zWYqRU;&?r`b&cW_*2fKzzno6kR1QDZD-w)jaBb)9Qz3$25B4i*b~jN&Kq)6Cpk5xNVi>Fz3c>{R!DEEOdMR**ws$ z=boqo=?FpCELefQLE?cFZO>Sf8x|qZ?X$ncMOiD3;S2$_Q=OXxyhnu4!s|0r|GuxL zl|0_wnk=WqMpg5nV3#wyXBBUr(LrIE$T{s6o8IOV@8+*NBWye0TUhOjR(fD2lUI_; z%v)jPtWCb~4+^&OH>#$tC6E>(aVjf``mXA5X=@oZbJEKdHX}6PX0?slXfu`ragqZ|MySS(noL86! zOO!T)yUeeElT3wHkmUB3*}YJz@`+kQb^w6lBjGK2_6i$=895dB$H8b@(quAww{x4> zA|XCYtq)kDT6|%)y{I&0Pfu&snoXQgZh%|#J>rzgbD-7cl_TPCFp2A0;<~1r&;ID; zod%uyHP5+L+nLv`9;Oj(Zya`6B9O8&eoxcrac;iv&ujZ5-VZcu7uLM5U3Ct8Jz3li zVhkhJ^$8olaL;{@Oh7+C*vkus+;#cb8MUe#aM~8jS2Bs*-TrXP4Sz$=lVw=%BMqgX zIlUVVzFY+ypoxDjUYwF&V)N<1F9el@f0&8JkJ z$kgw|RPk?E`(pmOKdC0=t`f@m{9r`&Ml{!bC55)~r85H@tp;uAs4ARiQ!5 zMG3uB=kEN;Mc9e^9(QWi!4x&NO)sY(JV?$(mC1hn)2FZZK~pDsG$}+fP&*xqblf#M zS553uG_G-yZjIyyjE}~mO)R9fxY(JDnf;J}cgwFui$|6_2k$6)>ik4^z91VAc_N`_ z=6-u4*dx4)GW!+a2PJhVuS^@gEeGPDf%A;m=6jwEv+f}!?f0t9fpEG7AsH!=x5>-4>d=FbZZh-Q8k&16uc^x`H5INTGFa6Mw=IcobQ+cpV4G`Nc z4!P7WlsrpI)xQMdnnNw|?BlI1e+lM*;PcFX78N)spNgFa$3lBpB>jXfC{2UleBQTI zg2Y4GFX~CMce(T#Ku&4%l7YUu(8H3t=}MM%pLR`-&8MOFvd)O5k)=&Q=Eds!f%a=j^JV*e{GbpbWu2CLoCpyRBe?+c^_V4sSh6! z&LoM48(YjPU5%<7meWD#`3FnAeLLYU`s-q5`||g588i7x8KQk8HrRUFfi5ct($;Mj zU4{H?v}`@GZubqylTGKTQL8ulFSa^B!7K5}`NxRUZv?TMS?V80*X2UKtyBN_O*Jqa zm$>KvUpPj}NV+qKUs(rUZfNRyf)Ec$jwFo6P71Ggs^1!p^}as9WCNz!&6;2_5y}I# z5+}^$t`qe6BPG@=cE}vzv3`r5$xIsJ#Cqs=V2f8eW4Y>viI68O?enVQpKOW%D98@i zpRAhI5ul|&TKD7c%^NEk?6MmO#;EzRZ&z#8qy0ZpcM==$Igp||p$46ii{)OM4^Mp6 zKa%s&tu{p;7kX>8?lY?PeEcRC1@|FrjI=)(#anIo|3wRdK?<;R!{`n46V2%_s7!3_7je)RmVr+#OWr5z5kv?57&#vIRX_2YRi#Sz#1#Kyb^g z6YOHJ%V@38(r*>po|D#Ju-T>pQJBu@0=c`~W2Ain=ZJ^dk2LUC=`V1i|HHrc_N$Yx zySDEoS7vJ0##;!=cqo|8zM8(Vx~ua$7-yTfu&8GI)Uz!aBR`sN(EE;>Q1kwXnh`u> zMr9WLXLUaFD^z|>TP;r4dub=jOwVPiA-W)RzVVBigBgM&bKj*`>`&fS|6XfwOkt+I zsJqCD>0t>JqS`m8=+{>V05^388RQjv4~g)x)Ts69Md~L42v@pn)QmLxpK7z*FdOt1 zErVxlG`HcVhJLHd)@J>oHhhGxmMm_2ZKtlkJ*HPp^TA*;=PEnOdnhM1Ndr0ih59K$ z6aTdr+;qPEP2&yf((h_HlJW4NV!h8YR0X%o7A)s4LNpj*!9~L~_|V}gB3^g-W!ze{ zrFqYMhhFgG#af0YK~)dk4F6>xi22zl3|bw*&$^ndX@5{&q7HVE5%`|4e^ujOz48BU zZrCxDzu%|7)YoWo*W!ir%&Y@xYylOy?^G*>Lyy&YC8HwC7k`@`R$VApAeoTKt>8W; zH(oas202rBD?L2U`7rYLX^KJ;rv1w1($ZEW1bFwS%8! z^i8gUPyPOOTyp>$Q~*lF)L&Ici+&f=6C}ocrXEDrsJR5Cb;Hc2vbDaoJDNmTnbi6C zIzU>l^N_XOh1ZZ{ly9Ej+>_*md7nxM-IR@}T1=8Co^6rYNAGhYNnMOVJA*6-IlD9P z0?@6b&@4gsO1t-)b)}3%7U(JgY%bM&P~LRk=ptgKf1XWPXISI>tnttJd();w{(k7>4(;C| zZ=rSWE4N2uJ6YPkyTA4OvW151lZ(Gj$j}u10S^h)*CDny=|)Z4vRO{dc-^`hdkK;WOdeZ0 zQR_B%$UJLvVQeG~bnb$dZLnk-eJ zqtAl+3#B$x>2e~dnfFIv7$ETD8<~ul6hae=jp`i{j17j9I+kwMDrQSi+O3S_hY+Jh zOUZ)xkfcH4_jdEoqCkCG$|Fu10mKLO*DhkVVH|Y8*=V8X&OFLKV$>QSb#ARB-bCN7 zMwBk^kh12BR@}wgxO3o;F$bu}|EAQcKT%VzZJcFPIJoD6`rovbzDpDLuDvDiRb!T3 zDNHlYvGAlI2-g^?EA32rXR*+DX{Y4|U4*%Cgg{apcHloWA~ouAU1!wlw5MZF%)hy% z&673RLgM?JCl8k_57j5|m2onUY1Go%JG&HXKov^M=z!!f0eUTa=A zS=e}7{mC13eYC4;E%|?ys@>+EbVu`OEX1Bs0d^p{XT#4%k6C-S>)u=o&A^4MlBah5 zhCWO_bI@{whN>|tROD&%p7Fb0x_{V?ZL1Yit>FJyvFfiyl3o_KM35d6R6h~|R$^{1 zECzlfpEX6^nE8s#WHZ?}Cf@LNxy!EJGBQx?XPg?>%VTsel~uP>-$nvUOc6(dDTgGP zxQ7Q;fDUNocj$0}@`0MIl2RdVuqZKr_de8A{vb1HIKhYN~Rj~zUvs#7#yq=dwdX8_?zwsA?$3Dysu!j(T>Ba z@>9a{DsYc{b5N(3han`Kq}!c5khRn6gM{c5qC)!$t>n)8Z1<%6@Jll6nLNJ5OqZe$ zNC3Pdc!H=o&;}X@-LWXI6p5elfCi2`I$>pn!vR$87@p-INg^(%Gou6_B~T)8Pf*@2 zkdZ~2i>ix61#tuk@MYv$U{dulEo0XL&64yfDpOV^;vwaBZwVV>%uw@}F{G81?2DK6 zQVJXO_s+0u~@S4x{JF%&J3)1A2Vj9>l<6-GK6UX0Mmv6Fq zWYv({geV9(5Wv)=f5CiV><&CcGd5Sf#NVV~AmMOU zz=HUW_Cd4x%2xN%_i7IX79*nA`I!*<631*ZZF#3U=e_$c84aZyMJ#JdNnlcS!|M1) zF7Mnf(Ukl;h+Mb@9+b;A8&{4%Y9foA7z?U7$fUm_-Y`Wv*I)aQYFK`2ofBhP1^N+A z-4@^jUtdJH1Zrs);6ACW350@I*qAXbts+8M|K0zSfm)Rep9#p4G+u$-^@CHwnb&o!Z51!@Y874r|*4xY2<*q(dAF!YxG53L`n#w<@YZHM`1 z`7shjlMOku2C()~m9RTKAQdbJat{c^+Ko25l_K)NNgfWRNn=0a9R=!lZS@Jj=@(BS~j>8_M z{q3q8UN1DMW8WDQ%2ql!P)~h~OUg!ScIRi`@YaP?VdY6Zrs|OK)Msz@tmo}BTTsOm zT)9E;Bm~}PspzqBJ&ncJw+byz-I1%vv?)8OTdCV|%HO|*UiqKm=tnmU)1!bS>SXc#GTDcYcIA)0!?6X&)busTSDp2CQA+3G z$-VLCV$Gwjgicj&hUJ^opan-1`eV$8HK)yn4u5*HcKDif z!mhsrmTaD}%lJwyJ|k>KP`IC5H8~!w8DdKxW%PAA`1XB`?^(#*E5({LBE*%A7`un=6+{aG=?VCz#jHA_TsQ2B=Fm$5$S9gT!O*wAiV* zoddpMMxt^FTdF_%#NpN|mt@pYAn8+>;3i*2SM4`+u~DT=Su8;Q{lTP3&MK7IT+)~X zDj&QFd5glTaU@(KFrfNq+4{EfA4@WF4DeBa)Ef{W!!#%Tei%OX5-W?xJW_rtlqbUMalJ5B*x_`IdvY+oGuCQX;u&0&!~PKO4TD zYr2lNd~mdX*N^JkJ{+{%cC|o@b{%1+hXFnEdtDNuigGP{G8{A!(mA^S(mx*ZOY3QZ zjA?s|_R~#%ja?Aa3pU6$*G~&NzHjqG+bm8s9byF9?ZViF=ho;4hSVo%A;%wM7vTes zZmc-uGxVm0t4K0r%Wcoy$)NiQv{Xd_ewxoNj=VgVvF5YxV$yson`#kgr-}z5Ph3vh z)4HwUV(@20s4l~NC3UpSRQ1jKf9hk{9#I_WG(ut8gD6kVTHzO2-^pn3R*@KO)Ba zO$$IexewbLiu&xke62Qc^;0_eQ+mcjH01sk+&B7nm><7)nl$Epted^=g_(yRvM6~m zY>DHzXKxBrc&PdQPGO<9xDQ%94|A={ksl0hL-aWRlk;dw@9W`D!(h#~)Z4+y4 zQW6(TJob}%nwH_+p35<>YxdP{P|K}zMU_6iYHLejJ^%WLM+sY=n;%Tckk4_YUYe%* z&O3_V$bXMMN7%Xcd|V~P_oNy*5k@|z^~;ai#%DwPf3lE*1OT>z`_3;h*lW)zii6W$%6d>jAgNBi)xnL2aBzwP?%mzS_K2q&9aG2_ zgmC3D!JS1NU!hJcEx3JAN?N9n;6atR!6U1pBvO^9>p#{m(f6gfPq!LayIs)#(Pj+& zU@P|p#wrK%o$I@`uceFxg37K5XQd$bAc@B<=OTaDmN z*?N|?Zjar%AM@c9dG@pcXFVfU+ zes}_EJ*Lh4)Nu{ey?5?oucw6W>)^GJ)oWb}2s^9%G<;#O&7=ZULs>t5MMR9D*4@bt zQ$fLThWBy8!IZ5Pt7praEOwmSCoAMZfhtxbL%ZcBd##C@YL|c%?^u-HMH-qWwu@Ea zZlWh!g*}%&hg5jy%I0eb=mPEfyK_}4DWPc7U5M??>sJv|^B-+I8)XStIh7PtYl3IJ zS8VE4z7#89LqX*;UlnvC<06HwUn!*bKFe1M6x6@v16-MmD*L5dSq}6kSTm91t5t7) z$U8A=96Dr40%wF2%fimF`O2@E;yrkY1?hR#cVq|K5Mk}!z<89n&Ew8ujutP0pHeaj zV~OK=V*dsQJ0i}ufxzuVf6}Mp7%KOYQ18OWzfZQixRI5I4m5dCWb{G;O`2@0q_}Guc_9 ze+S!r3vILXU7TIM3~Hp`B$g?S7%yV zAh_U6(M7%orKsj-oy3|4&aX%~1N`RwX1`{f=UdhMbn8m{-@kJMWeBu; zk)Vr*A<+BX4IVj5oude>&6QdOz7>yFdh9|gADy(d|Grm0zVp&2i_MsE_$dO&SltKj zBfk98$Ov&&m$$_C;j`j>r0J%MmnYs+-8!w==@uooI*%A-6u7tJ6M1inlc%Qu(+TRC z@8@oAUhKXQ;9H2Sc^}p8bB3J0jiRRhtqAHJb*AB?{RxEmC3Ukl^OaFI?roH;i0tQ4 zXBAKN&DQokvyEqnrwJANypjIJ}`)HOt+nPF& zE0WynzHRFrZ_?r|VRYs28V!YTiaM!J6OI9;k0 zF7ybvw>|3}$#DmZ>rk+Hl-f2c)lIqOe_jLkQVU44W=hC}Uz*|ghGe+UvnWK|2a(y- zCaQ0OKv@HrsXKSvJ}v5$S|QaTnO1$0* zzk{AlD{vxvqy-4!K$P;&7EB#U1L_`9Ntp$a;_muLFr4r-ipQh_&0m^l>O)4QG*%r! z<~|Z)S(ON54-2)l>_a#wv)8EWbz3pNbsC5kN}&Z*1EHc&D`yig!7?!^QD^GcPLMQ# zFWDjE#avW_S6nc`l6`XVNv}W!?j+k2YWJrCN94cG4fVbLK~m{N zljxOxP3W-VOezl-Fy-X;5S4#nus%ucd|zpN;dhZ5_;`2y`*ynH#bqUvPr;D475xLK z9_orUJ(emPF*p!i%Gk{L;s;iMFD}uq&GLsqMIXsi|D#tU??&oX5sbZ&>}lPQyT>Y- z^MT+q3}a(mW4q1?LETCs4e9%Qm&^D~Ta&Dj&ik2m7S8Rz)#A*1eq?Wq1D^vGCJgo` zLuyV2bZQvJ`TXoEG-m}+_fWTct_^&+ORwcxyn|iDfpPDXNz#$w--=0i3IP^JPnI;RZhA}ro;>C2q(^gr8CO=AC7 zdKvI$BTepN>t=p3N=i)04s%X3@YPGf2F6#eb&+=70EudkH{LhujGMeZ|2AK0%R(pW zCFLIr=;;(I;KBqT zSh_8@dkc@?6dn-Mxc3e8u`J=r$h?uKE{uGI9{fRDzs6C5Y<};t66dI>LuZ+G+()Yz zv1aPYsn8M^e8If~PCPK^*7J$VV9@df@7NQ!%C?&<^p%MbJG#oq?W1|X1nFJA;JWKl zK`Xdfv>aVcXQ0-aaI_&bYa_YlZF561NPp)@J9Jzt4E-_iJoKnc!F;V%d>;k6$rTU3q zHB*tbj5&qc*4p{*pX=S(mhCbGML?MFAj)5=L(XeKe;ITq!^=p4&yW0Q>^Bsf%W`r} zVBJJ~6vr4+;`VO5gIh{B-P%O_nPN6gS7jlfM|Yg~yxb341?OnauZ=sjPxwcWD2IF3JorJegnxOqy+5r~0P8z5c>Dr)=u#kg0 z1M2@t5a|I=P)cU{aP1Xe?Ac7b_jd`T2QoTqki`|DBwr@T{_o4-#)9a-R;*Kq8w4v& zXrLV=E?l$Fa*^4wtq&2Ew%bN}qZcG&BTWzNGEZNGAM`ze^tn${{y;XDWn>O!%G-GHV@+g#%=ugs0uupMz z8ynTXpDcgC%jg@k1n+QzZU}_JWPfN5iW;x(Qvby@p@MTiAM(~N8}inn33~~G+$pAQ z2|r>w_Byt{BaV94+2OY3j%}esM@FXVP$%;DAJ^TWVKKi$xHzo;wClTU&6_uLaCdPl z^IGTq;?Jocwb6u~fDxC!&NXXzuJ&Z7y54RILVm3?&8_o`k9XA|-=d%~*1l(qr+@c; ze^vb!4J@wo*_QCb|0?cZo%#RQRIyeJo{$nYP@D_iJk$IN^?$$w#qBgcTl$O*>?>$M z2PSk_j}{V#x{^EbhGAN^xBcqXP4vU3lJs}4ebF@#CGTa^KY`H%d*?P>Bwa{+q-Vnc z->18Spr7?uq({XK(C72ByyaAHZ#(zyX=yu-BKNsxcl(gvLQft4v90=jQ0jRGR zLo)~p8gqWGY(#myUDVSgdWr!Z3I=QoLe^$u-Y{cruc2-f_!Jfn@X$blOd(b~gX5eh zTTZ7-ZsDE?02cuOzCnvA92|(|iUtQ{i217naR`EtezC%_45J?2!yS#DJ54X|DCeHC zr9Y&K=WCwvZzi}`K3b;mm9s$4-7sa~!2TK_kjjgR2!3~~IyfdnTAvXo+Ex^;%w!@K z<#4O;70|6fEyiZik>&OuYL)#l@XAm3`sjzI1Z=j+eGce|aFtW@rjpyD$BEO@J}{w= z!b|0Bi{_Wi@CCCR2Q(5_bApR;_|8WC3Fvy0kQ7R?w404$0(=R{i)W1?g;`SaZI{Ko zMpTTy@aF_H^6t@RDs7sxF-16sUjELYEo!)R^)gOPujHSHq9B#Wv7P$Quq6_ytmWBI;-|wS)AYFcBx~Y??E@= zD~Lmmrw^J9=`Xr2&AWymL`9`?tXU-_d`MT)SYA87rOezE?GVMouWvnBS6k$i;M&k* zQ^tE%lXGF_+(pm?l)Z3T67pxD11wl~NaHpW|MA$uvo&LH!PFM!mDY6pX{)IQAK|;1 zp%zS!oy^rUOP}gmz`6G;n*G2aU(}hkRyLB&Tu-fBpb^4cC47*wYH}WWO~U`)URVBU z^u6nNji7vx8Qs)3SSk~93G7*Dr|9+!zdtHw=s}%xL!~B1I@-*y&ad`(#d)C&sHr`< ztw0sgwZXO~4RjY8!Hmk^F9B^gTe$?1u~nY(v?$_AYoH?7Cj7vB`8SbD%{#w}C64Ku zbSF5lG<(@|3=Aq!V$7XdThkXSUr?@zcOhHBt25HX~*D6h{wb?lo zBp4kfhp&G-O}K({6snAQ3yd?=tyQnN->cX?4(g7)rhMaz?^NEU%HV=U89hNDT2UtN z8gs)07LwcfY+EEhwOShixGzC(G6x>G9H(SI3CGN`$4&Z1=O34V>_X=b%r)8#E-ku{ zC|qenb4TZgo=UneERjaH15-Uv@0Izv>rhEZH4nxs4?rl(Yi0wgBN?Jg)n16vt<~Dh z1}hUK`+-b)@e#yf)Pz>9R60pKSuNo_L5Pzt2Y0$uR+cSU=lxjwsZkN+cSCyac5UOFa+W#q8g!zKKopEa-5`PohYsGa z0Y*{w${^u%m6v|pY|RWm2t%S_uSz#8DQvlOQHdFM1|~*M4nJgx+;Oo^pCTf zSfqYY1UtI5SkYC}E#(l85~wpJB<;FS-?n3oMx5pAWCsD z_Z&Qe#m?yl(wXCZ5nNFX$wRW0Dx9i@B07BG9~F{gGwY(ahUapb_h|g{`L3-ellbTv zfX!WaG!Q)~q6`#M6*B`Y=KOW-1HP)eRIE`tv2i2qVVYX>kcn5$g~SqqGHzQUI>B=6E;*+CTkHdSJMsH_5Y&3j zOu{3~y8drI{xBE)m^!Uh-PjF+#xbvQa{iy7vDYg2!}hhDicIY!RpuI0#^#_5oQ)b; z^RzBu`r}4*eFVM9%;Dhs>A!LT)F1v7l3q2x{9;(g>`2(H)MTLpajDVoy2LVf=U#Mz z4)dUYgTg5KZXNRDSZU#r@h5Ef;L1{@K=pB|s6%}SIfsW9138Vubpk%-)w%Sj-3Jnd7rbHI>j<2JALcXm~Mh9fZlmOFlgN3d0qDh2__ zooWc6rf&kBSTd5Q+G6Vb*0;x1z5uf4ReX0ED8Uv?&ybhMQjs89n)hQzw@qm7Ub}KT zE959g8ePFdLE|}5;D%WlEeWcPMA{k=lvs8QM3k}4WSm{&4+orSa%CKMJ4l&tPPlt4 z|F*IFq)wxS2WdQu4KH)cQ%H9(LV-o6n1(q68Q)muMyUeaWxPL9oyf0_m=A~dlhAY( zIQO#fv+#F8l_ zo9-+uY?G(mdQ%wmWZK(;bXgNqdGPE@Fnx=gOK)?{{uq$j;6Q0aB>qWlhEqwS30C zsnj)n+SgoaFz$)+4AF5MjQ1IpJfL zWI$Z{)p8}n80Y)7$igLL>?uNWH<9@BT9o0Hu77>~S8x2UnGAZEZ_= zV7Nq15O^#>>Y46LwmxHC+QeKh;2b~03nhUew}`Khm&&Kl)&=;Uu`o!?U>u}J|M;%- zY?EYV%!%-#qnatLL%x!g6f-rccG0In)y{b)iL+Kx*2gd)a11U`m&5!@eZWd(ZLFU~a%v2kp4qbUR9MW=RI7qma)m75|pd*I9}H@_}178qJ4RUS|L}F(XO-ZS5lHV?`s9Je2E}tI5{R9 z!_z{{Nvp}yNv|2?Kk1OyeEI{}_R@=~5hsoWc_cms(Erjdw}I;iq}kHD2eSaKTXmV=6CZ2+Jr1w*rMRnw5Ox!Q-; zv;F8x63APgQqH*1U*A2aDS4o{gw>Tn5ns`{cXVs%_$fO~k?bB*TvzP{7o|qZTYmuQ z$E85^pm=V;getu~;6Pmgp<{Nx@b}oE!u^sd6lrBH8!oIzu~-zl_rs<*%@H=UqCTq^cJwH29#E{8d0Qy zP?#iQ{?QS#WAWatvOS*pi!}!;UA8r}UDe_hzU0-0ITPgyX@{>T;kc>6SU+7ja@C+#6yVNqI3d#ZSX5Wd=av9v+p?; z5x0}!Z@ADg2*!J9?$$h?tN;VpoX4{D?SF(CL^U{-<_t^=-S)ll{DyN5|jbHO%k+y3#5_xq^v> zN?g}D?=eRCX=2wu%2EEY#l=l>uNkiJ4AZxVYuC=tWb9V~t$Vo6?2*_N2M zyq^HhZYxqBPcV%b(Q|D;4XY3!U%`8LUBR_pxB$-{K$g*!#h6mr6C=6220n{~YsQ~If$&%~JWtU>ZW9*K8cMHv!?f)y8Vu}pZ5OUNJNSd@ zAQrB?21C$`s~pFp*Kgn5is-CIu7R^=oA<2slwVOD0&&tFe(9fKJ^dU_SMi-M@Q?Mb z^(i0k)dQm!f)d|{w+1vcQcD_7HXX{Vf6ts&c7CB)d3D66e@Un2-cm)Fvn#dFm5&Zy zx}HWGd%VYKU*e&KR#X{J2Ikz3(%JstT|np{a1}C#2A&-f4508m36W4sS>A4xg^XrF zjaqUV!g|z7YeKFv~$p+ z91gnM1xV^H&t1m6&%q$+8ITq-HqgXnL^ZWP>gc#wfMyaf&3FG8!J7hh*F~*WbTcv3 z{Xx1KCc@jxvk{GD^fjlglB`VCS&pqByNZHD9I}9>>GS$MyXsY(s%(pS@zr8(9btYZ z(gcN&!JAcSiM^Qo?%73IDlJnHcq9Cwh^KPrP+Qf9p-a+eDO3IeTm$z8A_=ncxtO_d z(9t?BkmY$PlOnhhIJZ<*!<29^*~=rm;gOVcBt<&OFDME7ys(d`;3%}KjesH~ZrFKHG$@&?UA6;; zhSeldp_0DIrYmf=g2)V#3~>4uQV~f`QYeZcG$i?f zg3`v=(8Q#(^ zuW)d9ZddKa#IyuvzL`9-#^_s9qNt%~8Av<&1FlZ-<`< zkoxdr=SQRo^U-hSud`lLDghU+a&7?pXjNchfn^?P=Hhj7ztV1g%wA@2Q0V%lFn^o3 zZjQw7f`4z^N_@2;W}W?WvrpaXBMFNjGNY&GN7<4#MW#o*{TkE&tOQr0#-GQi0>IK@ z^P_DG^P}UJ9W|&{{!^>I$2VOw`?{nq)Tu?UytsaBV!H&~=xG++X}S$j;HKZvGhF=o z;+N%SHeivf{!8pDGpqR3Ia^%kb+!1PfncH|%TwIX{9(U{@AxVwr&k>NORul{yVhrg zKd4=GS*yy4ChdDJ<1?{U@ z7&j6YG6VAF>P!JJ8NOgyP#Z1@Qlv9U5n@~v$I6Tl)PB?CZRQy8GGOtLCh=(qgjbma&bE)j8t~n^sjO>PwW+Qy5lB{CLAW#U*7(^l!l9(t`0C({Y z_mym>nE6iteA@&`GKy3tRR<{_)nWXBU8H5mELv2GF={{IJtJ?KR5kyH>0*XP|S=HgKq}T+rHNG zGL+P1??)%jo&JyXLzob`i*61_&MO&BeEmUi^?g{b5uZNyD4r?^8Z`hFtNjJa0(tYXM z4BTRa6KT_nFOX4AnY~)W@HiOH!PUlC-Vw;i=?@nWJ*M_2JgnrUW#dO0V8qjM>w?>NsIC+fWbFQ5lGw^6ZXe}*ai^X&1d;&A^V+z@U#Ir6?$ zQS1H$>TR)#*&%J071Rp!e`k?%o3!AyLb+1=Q=BZy$$c!Mjn{tqR^#}rR@%Dok%ji6 z%GaH&$K2&ZtH!3hy=N5z(07j89yUnWrHu&6+p*2SKWTN``BuM?2Jb@f7I{bpsoVu4 zFk*)Ig?q4vZ|2aRBcYNHsb|A2x5}H6&l<1`jzw;9lTX4iXfts-@-x*8+!wXh?Xr^> z(>V5uUY1l8cr!}9_PJk{psZBvonKDl57U9y5qn804QZvK>5Pfr!^6>{qJSVl$lhkVdjg$O0%D}vKpiYC;4)@Zyr zwh8|n6%G4vz6E+SN$o8^JiZD?T8!bH3sS^AoK`9j-!Kt;+N8|0u#L=&r98fu{D9(- zeQf8UpjZuf(H>WYcuJj)-jhyo4kFxBc^~GfUe&uc=t@nSfGQBWUF6)XfJtCNgrKQ7 zs_jH^B`_tQjCvYr+lX2Fnv?xSc4Hn#xp`(HeWoT!;y?+ETmIR!j7lCkDoHT{DsqIB&D5?nVuc zFSFQMwiA_s+M;wHg!HIxY*oafMgrxTKl`bi7)+>u;QQa?SkR<5( zpXE=9Z_|f0tNhvsxG^)YlI9&g)?suE(ur-xE{mXO1zj4t+34^Y+sPQeu5vNM3=Pct zmB9%TvRuF^b(1_7l?0s|)e0SXi{?lW2Q|y#PHVKTthzU}7XD`DNde^s5!I^`@{YTCioy z3w>;>sUOeP9Q({I0|B3UC3V`9s>HbQFzgpry8g582e*Rw1C;lLIAORL+SgpKriO2Y zWq<-**KF!2oU~5Thg+6qe(Twxt$k#w>VZT0OM>aP(*Rh@ zRMv*G7ZfzN5p+dQW|dw>A_lH7oqjHhuGq}WncrhedygqaAEScd0UbwaE~(}w&=_{gdX+ z){N>%^qPEX>^oZhE7A{G2_xr$oRcnOfcfGEw#-N<8ysUUI2#$gG@Rjc>Zf0#mXjQw z58tV-(UdO9+Nb@n7c=-dP~ej<=s%{2SOl`mDjW5@l9gR8;8YD~cV3(z52LO8vabWv zomWuk9?fM`=#{j>aNey zbi$IHJqMGX6d-SV0!$pP>j@yUL)qHl+o@bxWsK6MuG_%@Thp{7ino7#&t#|~`@Co? zIRA-di=kalB&5i(NDQTsok^sK4$vmul*qaY*rC;>dwQ4ocn4HKxQohPMd_^mx#-6w zR>p#dAD%8vlx4o?Oaq0sJWQ9B{EyH+Al4m$F1n^0Ak=azY^1%vyY!m5=uJD)7|*F8 z6jK6w@@n0wQ01c8s>xP8gPlrcQ?_BVSn75)R0Tt58j;i#TL^*hDWTwIC$3f3(oo0F zJlaN>-&)CdbU}01VI)7aptPaatILd~>G(0ta75o{^iPLF?wMOjXyWnf>RNY9pJi`q z!6XEY8aB@OH{M%zIkFg($xhg56RV+;%eo&xm~;v2IAT8HzuQtRUZTmW@`&kS#3;;G zx{t6-XMN)L|NW@{ZovOn zVxcLHhQJ54AEiRiT{*aoclOXL&*K5d^-~K#$c>veb?wpW( z)}odq^@82T!S?Mf0!m~m2X15rT<7Pmuu@@JE5J8%vP=eTyWXkF_!F4eMAhE}Bb{T@ zHqpv^|MD@8{Dqi3s1LdV0Tj63*li7Q>!rJN5g=$I{!`gfU zPy>B_##mc&VijIl@$y6=)aOyb)qfAYY(bE$AF;fpH!4>9|I>l`W`5#_yzn&mX#;iB zH@XS&0Yk3Vw08`%t-~R*t!fKjX3OfT@a9d=R7i@K&t`SNCWjC=Tg@(huz}k%gpt$Hc^!ijS8j5x%>U|s6wTv12(|VOS=;4G4L5H8AJ*^PQLn1P3Djrz=O_FVMW@!h0l`mah1H78Gop@4Y<8@cAc zv#-#rWh%@(gv`k)uq}!C9;v;5Fx}$nB5wTNQ3t;JU%%L&xkb3n3A&|v5a!?-5)v`0 zVSHOhEi>%w3%^=X`xUBFQ5jX*tJZE#jRSFb|YkNB&PA&gDU-J4Pql0)|6Y-#pM0`;>dQs$DLRp#_h6EaEK= zKduSBs>F}}EaD7@Gud2+D zx682%7^rQ1IVZdH{;h4@-#Ok}WpGxGn5!`y(I=bH+G?~=}_o(gbdzsKU_c>>I} zq}+E-QYH=w@7hhqa+^aMoF?%>KS~@2gl3H#IL*UC`@s#j=;_X?-OWYN;)7v=5A7?o zKjcf{4|-nRq0ZHq<2pG3bN(1ELytbGtpGXOEAqkO8<|r7a3ZDkyagLS8s%7R#ZF=) z#MH7L`!Q6(Yu(1lsRPD0vs(WsT|6mb1Y(8B?0~?3wTp*<oZRjNI!B~K(b;hXekCHa!K zKH(N?r=wgG*HDH^AP}}2gKPf#exbwn+ANas9C?13AEemC%Yjhk0+3%spuU-Wag)l7 z!ZN~!`e`PCD=R|FCrfT(2^-6^JezQhM=^X~%Rc}i8X!ir^=u7QUb4Z_XB=^RlTci} zSbjq^5C+B#oBx1o-IrFb#2}-^A29O+oRMT#z;K5 zMzaE-v<;6{MX&|YTs0IoU#`lik)O&Z_FoIxf?e~!XzT~ntUik@`Aq5buKZL#tStZb zND-L5WHY2KOqasl0J7gf2iekGU^hhwi!-P?IV4v&Zmcw$W`K}0hGcZkFTgQ9G*L66 z&#+VTvd=U(#ZWo{M!eU^6OMBRaC@aawwMbHL+i=_w{_#T6Uo8YwHB(N-aeP*Ha(*s zlK7;K2~wuRU|**SC@WAOlbf!AGFuKLvQ-ji^xQKX>^#N7W0fZdw0=RjbaAkOYmMcE zn_j?;u?%K8$2c%f*z?AwEIgX})j}#&4!^pL>ta6B*hjmE_|}6Y?XTYpWsAG=#k-PR zX-F6TZ+Y9dAeqKXwz1KKEwyvB`W{xdzh5ezX|51QHQ;r8ia&kRra<{03k{eU2% z=VV?b{`UQ-n^rUE~ z#a3)*qTYjNlj^IZd1&%=nxzd9ZFEPp(|`9U%#w|-^}7N=>Cr6rBXnqB2yb!;-Vr*m zWfnVY2>MZ*;tscd0pYSd*v!VXre{qvujk1lz|2NFjw)CV=S!Ldxriu1_Ni^Yc2a0h z1?3wW`FpRVDFok2U6Xdgpx(00{7$;h%pRc`eC|%NJ_-b$Q7iPBJ)1{W;f@-*hKsLB z0FO_|ezpt`81-0AK(B#k8j=&3`rR?X98~1W7F*$kITMm6ylJ;<8d9Tv9fR0@`+Z6_!OJUjw611O1E|(d|MM z2)EsQz24asR`42Jvgk1v0f@}JhK1Ny&Do)It<3dQoLhc!5T{0Hgd`#=Empg&y^?8n z951_uq!`0H#02p}SgHcV{1a)aS>yR1YEHQ_ox84KLnXezm@r->wPaxL1efd#cqch`l*>$-7J_9W_QT)mRNC`gB~CB9Xvm?D7JKur!_|m&#Qo55^u@ zPS)5!Dyd87!!le%r9k#KS)O+HBaAK-vn}c;a4|}VTYw^57n>__)iW4khQ&|s&WvH- zIY!;_D{o?i7W^ofd2$b>BDnzJH4%IhabpBbQ98 zuDN2VP{qkry(`pcE^_(!GeOY$!0*TPs_o;AB>RE=n^OmOXQbQnR`))5+QdP5dK(6K z*D1H$KJuh~lfgG$6mkCQk!bHn*4FAWqYo;zL9VMG&L_L;8hNW1(XF4>23i_05`%G1 zHR5w^lqUt2W0Lxr8OX1X1?e`G#fjs(M#~SLDUQnT_Itp=V5_c8m35wBHPD3Fs@ehs z$Xq*FdlG>e+`oLzlma|?yDU#0VS1Cw#h5JqGS%EB#+6!kIV>X9;NWBlM4}k=bLegA50l%ZyPYQ6)}(t) zIIq&(k@dudo$fC8p}IaUm4HR41Ohq4i{{F)o2%twwYwRp(WLG9X6=-kf!djcv!T6vU}8C1UVmJC=Aq4|fe)P!zvqmg;u)H{$5I*zJM>I^AySXj*IwahaPubouK|Vyi10*Ncb1H>nrOl}v^<%MWMBsiX{Oh|j zK9H}3VZ&H!71;?h5x+)n4jQ$r*WT_XE`TNJUh)ik7x=`*i604DF~}_jkow zNZlYLrqlu3Y!`fp7mgUU*M?a`5E-$A>@Et8I|-`j=-^&ql-cTKe=U;&l@a ze=zjL$ZYS{W_mx<&k}Rn^6t4NTb|&gpoNKFnHMv+@^iG-WU{2$`D>2fpm@AWR{$el zzoPJ>aK>)&FQ#?+(g4%~;qHo=E0qatM4O2Uc9;p*KnrxdhYq9J=DMr|;eRdbjiz7C zHBF6;P~)RC1PxVlWK6vXwRQqwd=w-akpBmq;C=Xpi{vI^^*>G@q`q7fS{9_V69wKl z`d1Pz-vpUTyHc7xp6Gv7r(TtlG|BJ)D-Sb1I>0gC09{ftiT#Y+X+|oH=Az)TK0_L= zr223R_;)0*y%2-Cq++>RS30W!Zzob#y6!B?GZ2F<@SwBhleCc7La?+0B^1%>oF zHZ1VYKSGN9@^c@&Z&c4U8m0!fqcD^>X5$u$@KgPaO>ZVeZ=x@0le)GBX|fEdkE{4e z)~Lgh3N9HI<9}4>@x;(G7_yBkPU4B5zG7C_vWy^lcU3 zcF+zrO+#*J>xZjaws$(}_tOJbOR*fxwcBXdAepTYg;L|_4=B(2Oq)u_OMys4ahn9j zK`d@0qq_I`2v3oq55JDXbBgL1oY!-iQ&B{;6HXkJ65?f?6q#=s`m((=m*sxkHEEHO)n68wbu9o4N@mK|tO<*B@>XafE| z86>*%YF{B|HbuT2xR_v@R&bT^=ZI2BDNqXZsV#5BSQYg}wk^#h866@$k-Rz!nimA| zefVo|dzdRoGi}$>n1>Dl(7slAyZjt$W1;U0EJN_Dl#y_#@Af(yX z-THdx0LpY>IdHM4OF#0UWMXHh5m4$v?}&YEHvg$n>6>2vWOA7A=Jivl_kf58FZ- zj_Hl*nO?()d>ko>`gWM`dWMqoq%J*(nSYh>5Mrmr+6|4@h5I*U(ke$_zMm{n=A`lQ zb!h@Vi?FO{xn~w0`EEczAx>2K5y;l0lb7LQ2@=^PgvO6WUE|GKJk^a>FSqBqBK?HB z+8SdBm0Oc@$6lrT?`rH*Qz1H{43)h@mjvf>iaO{)J^P#>b6bNDa64tw7n_2J&EFm* zW?rQt>J`ZIRz?<=+f$G_ya275nVI}ao0?-qYgFBsaQ}_ef3qG(q8;43V^?0Oe_a<{ zAX*ZsDRGl_X)7RR;vi;IjkX5HOOK$QJ_c>k){f!ndWO>I)w+gFgkkQ%7Ilue-tK|| zbma(NCswpOrCsgYqk5YQux}-`6KN*M4L#M1$OlEPh)i&v3V!95gbVy?Ru!nDR?Z4k zc4ZfZt05{~1LihT%yW|CI>a@&7kh{=2J(C`PSwQlsfI2gZdqwv!PPaJ_%=WV*=Pb$ zmQs_W--Xq8!O8`-H%fODXdPYi4ugQK8X1vdSugxVl&yDhun-b@5aLdmi4e7@1g6D8 zTcy%g&>jFkpeq{|HqNWM5{epb1~==d8Ix+}$9(={$3#5{UBB{}D0?W&(bujlX6!>9 z6%r`dfmzU0DVtVx!fB(Mtiq`@36sOPD3H-Y04B^NC4!al`OwO8o{*L%xLT95;FFi$ zQYj9HSijfx};NSuK@&d<2TKYsYw%S!@Cs$U#1f04zm| zh_T@c1Z@#_Wa^|in<2MyykAeBGCgt(%j5IC>4z`fDRE)_MWPRTW;b`aqR+`T;Vs-CuG{X+!kZe?P9N!7sN#7>M=Lq^8a7o2E_>o*O2A330A3Wh+^k_R zv}@3asQ*&0fvsPHJ<{g#O*d+m7grQ2>x*7D%+5LfnLTwXRP+%G;tD8E8hS)$#9>LE zc;@uBEQ@B>w9Xk`oDOI*5ZrjRZ^Cr@5xa~sg{kb@4PWfj5TK4GuUV)c7W4yvXReK+ zv5CkIhdY{o+q0nKR)%R7{=A?M-or5?Qlx9w3bJ1j)@b9r@}lsfHM18cSx+!r-dsLG zq{zR^*HZQ~V)hp*fx_amO7n~SE|Y}1VPmna1vrk-68|gcoPmF_)atX!tp>J(-u~WAdd_zR!k=FN*(U>!5_-AhDkZ-aO&tcqk*cD>7<$3a-zgMOcW%`L*WLD zh^RFvrDo6LMev16a1Zd-wNfbt4aX6D*+^JVP|_|JJ*>zDZ(Op(OVPW-Ej22(gt`WB z0*1hTQemOSw`V&Rz;I0)RbxG>Rx2Q_dDqplclk%(NNDnEe2Uksu%xi75!2i91M2lR zs*DlSsnE@pU?GgXd1{5^q>FJXPx`sXde;clsk!}O|2XWYYt~I-cJu3t;vPZeQ&m6N zYMUk0>lH|RD!B)C#Ioi#rPZGFenhbFHcPG0MZtAUj4itC;hEYx$ll4B^JhO{rzqFY z4!S0Zn?`MHf7Z#eQnq_~WpjZ&_2O9x(#!Y<%dOaMK{AUU;3ogj(Vu59jAg6{=>bGr z;?3iit0p!RH8NaU2XXFibd_+QmUOn?@RYgHOSB8JK&NL=hBtbWWYh?~Ej`%`n$%#} z*L~d@ofm8Vj!1M|Ec*(&?UgsUe`p{s1lX)WO}(3vmJ>YJzI~5eS$Vi>wg(z*nbY2a z?^shNZ`H1$dhA7_WH4|*rt|N3fY;O~t7i4ASY5)-)i+O}A53+4N%}7&)Q)Ok8B5Q> zRfdUd!8`%f4|mMpGtVQZB#1~;zsrjNDpS(V`$!;M#Bv*S-e9Pf zQkWCbDna1nf51*o%Ydm0U=pK6_n3dwE=16Af}iMsu8+dzZSlS`lQ_@YCoI&tZq$4d>8h^*%>%Z>+4$@chw~lj?$j8 zw!NQRxF^9X+B7_HFCl0PUz7+HX-;ZJ2UB0CPUb*JN22E_7OWF$F)$IgVOAaOzgJ5~ zOH;m-OA9LqUsgC{r}cpL?5wLDI%%-*7>SF#!$W%6pr=w66c(*29xb7148@|s-EPhG zDh+?c^SpJ~NC0{?df-nVb=tsEK|s?u{=0WzX+=IM#3m86&!1M9-|f=0d^8XDP=mB3 zI()m*@eOK*ZTU|N2(qHuSEyXboIXXRp1E)#qN`qh@>maQ<%K`YbMx&uYV|2DMZf+JhrN;em4r{UeU8C?7F zG)a^D1Tyib=2|7zF2k;n#rLs^@b^WU$2TZJ)tUV^f7@c%#Iw|j=Y$^zCV4&f4S#H| zJfEcCwq!+m60P+hZtb!Gs@wt94eM1Bqx0-%1U|^+ECF5y&YPuo?#=@}{2vPbs5X3{Hqc=hE&i*#JRw#aer+|ULp(|V3bcHh zzBQK>GWN!d9B2&zb`vPwi4(sG#a)&sg_^VecHR}pa!P9oO5ymm;;8rMReW0?Eu{xf zGa?IEhO+^tvDc>T3BDMnVJ3tgo}@WE*_kuMC)89~lfHbfj4c6v|q9kTAk##miL z^c{QJ7to5%K|E_*-k5%X@k}>{dV6-)cSWNVm=6C@b)RNZ6SwlPs}4AC;N`!meARP* z^T0Yu9sc>PkLOx!D)h@$@lU7!^H30RL5LO1P804v-lk%AZFMA-xF)E!xZSa{sz=T@ z?6KQS816M%lfW6s)Dg_O{r{;PC|c0}p>i3uK#a&z*=OfIpTn~jrS)&@r^QIJysN#Gr@jIYp-pT<)|<0WGb-!wO%Oe-Z?jGjMf)~p?n6Sb z@>0~(fuW@d)7f3zXZH!}W~LYYy#HAXU{P=XP8qL1hKf15bfwEc31% zuFx*D*&`?~ASCA-OwU&Ny4-aK{3e62DQoCEmJ{jqs=ygN<;nw`G?cWOUK@?Ppz0Yl@dml z)iqDz?ZR*cITh$0R6exQ)29wUA7P6cg|auKfB^y?C{N%F4q&@2McI`r!ab?S#dGqC zuHTt!xZ8Xj2=>Y~My+g&==R(R>j~Q`#L9~!YZi&BWbYcFHi&Q$yWNLEw;_6ECDaG4 z-$7)xwF?^Mj9%e3=^8%1h{xEl=5I)+QM_3Y|T$|%|qXSRg7Y{3CYfCp+yh7d_m!sie&Y|?xaSjjP5 z2@C`dB~;-vtEv?>1X_6@s!d`St{9~{c;I1NWx{z)m0UD@iY8SD!N~D21U#@Rm%1Ka z+~nEFf>Z2phvqIS8L(2c*_$Zeo<+x38fnnM1u_V>nfxOHA9dj^w|E1@62*DiYLybz z{L$?{z4#I20@`{|A49e-*+-+d#vyyj{$CQJ!>8U4wF$(74OJ?7@a%*1`!Cbf9k7Mk zWx9+y)~X}H-2KsHSgT9jw)~+d5t!nW6~)c3VUL!^5t`$=FLVrjoA>&G1jKt0^D83d z!)s`K3)@?v(|)=7u!?KrcG_6??i(6k6bQPSN$@5$CG~4lKigu0!LT%<@)7^&*NWx3 z;HiVD0wp@`O3pWTWhdt&1&04V$tJK7HvtEHpQ7anzk9DK#rx{F=s1-8jikng=zkL4 z&!t04!FgM_Lw(WE*X~Hump>12K#+B*h#>}FItj86tl>e^P>Dqa+Qy#FK zeT%Ej8(SeY3ZXU6IJgzu7!U#Y0PaY4zpKa(cA%(Qz*(^6vA%(C7?K_dHFW|JV0UD- zc%M;KAXCUUnW}!UNrm+W5p_T2d4e`>u78<}Xs3I1?qv5q z57|xHk*SOJ##Q#%SFNhQdz+$@3J5`8@}R|SdC++8lSRtYZpq^w296*k)g)fDPK#F{ zzWgaeK&cHYNEy(hAtXr5Ga-E$@Pzs_h)=fE*IY_=4+4i;tlP6(hwgo|M!AyieUo?d z_t#%vQ}v)}`E!*nCN>ya<*Di4uRFM=>?5s`+^o`aInVx6{+hano$S@0Nm|zG#mpoH z)cu+4{&7v=ueXa_@0wI=T?KJHs#4*-&38@Z*X*_{dS%}mc2fA-zU4=UGxs~!W+XM; z+3mr0VwG7qn1?a`mtFBc#_xRtJNll{STK;$#b{$FXpcxF+0$Xie10Y8U0wY0xR2}t zIp>X4YrGf9G)IyYD|~1R5!}0Jx=OA|n*Vv7mdfZqU8)pA68}?aK@jlzf9;vZ5>9Im zPOi4e51`O8AbOq1~32@5SPmDVLiM@AjZyC0pX zVUV^}^>#52@luSH_1ztgc)UkP5%+x36$oGYB0Mn&#@0V}6coJp2jciza2MxTt*!Sy z9Rie&M{H%8^|m&Gdw_#mQ`S~&wLSYhjr1~@XeD_3xG~5()3r9879#80t^JlSAz461 zOU}otAzXM!PZ=Y?>@-H^e#yve1HuY(SpvxGRT(17PEQ0bD>f=M zDtvEXso>A(DzN+3X9)O`SaGZsjLAY`$ zOP*D&u7btNZo68K@s}GQB5wUXvi*Hw1zHS*OOPE`Yfs{7%ilCx7yg2gj|7Z|?KCvf zurMsbYricHP~lcQ_`1(!F=%uWj4id9191Zz;+;U&KY0yz^pDKPzW3iH{(-Y=GQdn} z@IbS&!N8Af)Lu>dOcL0`hoOho$l4g+BPT__j=yI-wvWtaeW5Mn9y3o2L%(^D1F0Bc z>IdqT+W8XxiVWS%_$?{Ai9D$-NsU^p$vf?O^sPQrzvcSazPPG0EriWbm(I8ebs%lPM1+fF=WHaxglT_i`8 zaByo@Q{>&KUXF9HuJBOlG$r;Y_65T=-vB(oQ2mXyey?B*xT7G!YYLev&8M1rA!1%JT=c4F~=E1E=_6_b*lre}Dl2 zCw@5u=XoA`fA8>P$gCl4JIx=Cw};|&CZJEVz?9L>Ij=kaO!9|n)}hDu`lwxuVpYB_I1KXD4fE5O6IrHbqVW!#{**q!bku?hTCo?- zH3jewe@?9t9{1Hf>MZIwuLRnx2EtoRzM=t>REL`R=EANx?yh5h!TGPlf1&dSqpN-Rsx6A8E)69p#T8F`@}ULw-6Wkb5tAVPwX^ zCCYje^M~VDUh73n(RvtM>pe|}!B~<$dK&bv9j6O{OV$GNa{tFRJT~x-vu&P`UfZ0} zrtemDFYT{6|6{Z(vfFfWl(Wa|+ca;L2F?KQZcc4x7o&Vx_^al^RqB+&r^lFaV*AjcY2=OJx%Y0|xa01dC#s=2gp?!;uLg`wSu& zepu(Trc0X^YC5gL{VU~-9H92yOk)8eododZd1^d48*_FV6cuVQojere8i-I9OBIsW zBC5c_Pp$;KsUT?i9V72i@}K#bVwps}1dyIaHkB5ctDMHlr&+L?-w0D+0@wE-tU3OD zs>aF&U*OQ=wM$G@I^yNhCA~^;I=0NVeWKEUH>ofln0rjZb{er+9Nf$wk!jjlJ;LuQ zZEg1Bd(v^inE4O$B?i?KMr29u$JkTXK6fRgeUmBbi8{7BUz%#o;{Z*%^%s*sA|!l7 zZHX&lLBfk{wZnajPfEw9!cc_?9=XT-n95PIsRM+1O%mK(Jy$yJ1LxJbM?c+`DRpOB zD?z$s2|-zzJNJO{67SeQ^$&=Z{)>Y<2|_2@(csTVS>~Fnx;s~kbNbTOUH!taUl*~O z5(^7-kIJ+V_WLhdbCUe)7fgu?sHSbU_ZvgcqTQeQw}o1^Wch448_~zpM4un1c&}Bm znv$az^{ZbH@Iy8KnUXe$aizIh7G`mF=^r-XiXZBsE?9@CyTV0?iecOKnlSpNPr~ID z{7UwiO5S~|jP-C4#@xys37$&MXn~ITt;QamTOhmLAKnX7{qptr@&0USp`1%|5;4Xi}M zg|m(AnL!;rb0q)H(jJ+3IGU?xq{PVBH^jT`LD;}CrN$l?k3(VF5sXC5VOc8v;KN#8MV(oCq=Pbyk+?wrg(o$rOlqVOzHtMyrXUyNTw2p% zP=&KYM97mHvU!9+39}EAo~>R^ZlZ8Cm#=#Nf|&RctFf78=xf1i(AntFX>y1hxHCIC zTfmoJ#BH84YW5&>Bc2$Ju*9#zr?!Wx(~G5jF==dFT>6!%Q!QwbI@tP{a~5nC7`^hU zC>QcowQ(hFRXsZze)=%eR#W63%`lI7YU>uR%lJza!*noc4C?-f9mCDxt!mzd&#Q`U z{}(ErHzs4P%}x>yG8J!`CUZdC+|Hrfew6zwSCJow)F7OI4`DzQY)V1^SB!~D0|)lt0tJ?4v8DY(yQEAZQj^ihb*!n!Z} z_LIn3pO88nK#K8X0ugwcP)tw|xug(WFvNKGL=P zjLuo2FPeKiZ!$MJg&5EbE&9b)lRwl#?|54x2yx|gDCDqGU2bW zgV6G5uKU%H0@fzVO|zFmTP22&9z~PCycerFY}R}aW$rk;D`9=?{GOKDNVq9_FliUh zXFN^ia#f5+{hp{tB}Lf~6=*p)9tZ*&j-Y@r1mh#-JEPhq?Gnow8F?m&o}&6Pn(N(M z2!0lYw<^lef(sETh~G2jIFTZ}WF|?}y@h$JlrhzAdz~Do1 zbh>1AU!<)ew22w51~r!Yk2|RR5AEcHkf1QXp}JL>Hlx*y-jW7!Ll1Hr@1e}x-};#D&@7Er zm&}RAxmWGIJ!SJa!H*a#WH+>F1f-yjKN0OVfRa{=8LtCcvv#G zIK&&RkIlW)BsM>fGTY49*#kSyqlm%$*4WDMLvxe;YGV;~7j;^6fcQ^q^O4KwRM)>r zmGln|uU76AY~;UQNZQ0`%uvVjMnXr)@#BQkn82I|h`_3+XNkj8`uh>aNtrj>VT2go zcrf8wnMJXoHd|k%yhYV#BS!&__P=Ux4g=lgsqn+W%|>-6^>9^(-F_AELG&+v<-KO$#q_gzzn6`rowOO=%Hgp~XDwwuZ3i-Z zm;Wjq0`45ux=|>14I6d6BW4#e6OJz)vOnWQI_(tr_$cK|a(>7o+3Q&7oTV3*Fm`eJ zg#tf?1cueMpE03dSghc89rb$LyeoU3vr2QhWIP*8TL4oz_ma?44b6fh1_5b zudd2zU+d4{+MG=BEtAysJJQJo76K=X8}B{->2s7d>GO#AGfy}y|8x396tPrV+Ws6Q z@v9Mjlw%GoDYzQK`8(Az6q8uZ`6W(9Yw&8_m#TiukMg9w2pJOYN0fa1bo~tUK>MMZ zOHMcI*Hik`D0Cnos#26nPFX*l@%hDMuZ*pxr`cf1tyAzAK*FEyB1^#(`*K{l>>CgE z`SjNtAMsB1WEC11He1W1Is4rn7^ioVSbjv@QYrWMO;4({5>l+g1#-=g)brEXl3TZ* z#%7qg0eKp2xR-Jl@787s&bJ{_A8FCn4}GTTKiI?M;Hu9o26pJJz&@G;{po^v3;3^W zW$V9-=28gZ-kXMbg*nIvWcFfpyAkb-ew;BM6*+rjnd9(d7A2%Pm^>n-Ik(YY3U4k> z*Iv1{l=oTqdq;BNQnxjvtmY7AZAAyNlF8UIZ6!N$`dI2gs-t)7>;M3I+d$t%g4brL>j!qm52OyGr#%|hcu`AEmJIp zHeWR=mCry@K4};1xd2!G?CNlchMR}tyw9iy6pp|w*Qk{2XSt|nO(UVMOKZ^SMPyU$ zIln&i>b^Dj%@$Ml3iS*eh%IZ5v#sS;$(sTFF=X#|Xb!wF_U{|l*!}bWk2ikcARItU ztOq%?3=_7zHtJKKmh<)6_4UA*@Z+ZnAA*im4e-gh#9``d`bpqko`cSf?{59xl|SY+ zhT`YR3?Z)YMNuL92~#HQs{zo+%dkF0+4;CurP$y_dufm2q?r*`Xm(k&i`$YeEDt)M z<+j<5cr51k?;)a<1M4G4Li}qA z#`FWO4RYHqZzcz8xqCg+JAQ3RIc`0xY%eS40}MO8sff6qeu7QoO?k+p z((J`_K<`k-lON#qD*4gpd5#7A=*RPmWlntL9mpkfQd2E7O?&qU9s ze~?%~hd+{VEc|l`xE6E=xem=7L~3^T8s4a2^j-Z%#VzG8Sc$J+K2st5<4SC+iHuH5 zdFm$U<&9~r>@sVr^)!2BwgEzOOep$QlkCT1q!*f5SU5*~)j&BIqu-j^KpHq?;hjl0 z009!{^P-+;=UU*3Cxh2&4L2XH`CSG`aCvE{EFfd07hIz2a~jJsYyk_E=9!J-zNL}f z460T?-$pNcm6x6iG7bAmBbSqwo7uvm$EzYW*9&JYn9D7hg}xt|*D>+lVgWQ@{xOrM zFCToD<-feA^PViPY)M0o_Ando_P7H0p!mGfiI1uJdD%;4fj9WJ>vr)?VvF&X(z~<7 z56UrW-q^`uIt;PeYR0@b{X?YRGWM6;#^$2%p|~~;M=9C2=# zfmobH@H3bV^ll~b$E_g}9j~_^^q+_Qiu;WhKMgUN7+j1N>pZhzt`yv1Fsp`Fn32jVDEoGkv2|e{DleGL`Zz4EHu3QnZFfk4%A{(CV|T;C z$G&-0hu!Ia^QW-uIB>D{eVKM8ODcBo9?J{!6}{tiynXj6d`zd6XjOEgK+-B;D;7CU z30(rmAC+!mV*|$nKR6CRC35k%lFJ2GpSwB=lNHkJn*(J*ve~V69IeRaNiery6rydA z_T>B0L^?@h$CsZ_TsaL@Pzx7>?d@yey{&Fd*vxi(_1MD@M0kbE!PGTM0Dm%))2*)^ zRs%kV>tA{4(3%?5PFx;)?O?GOsWu=RXMhi1V+1Vs&em8Lg^vgf3MBQ7%nNJu6C&~i zn568sRKW)86dl|C_S(nafhCEyoasj1PrS!Y8V!+wWYD$zLQ`2XlQOBh*vsC05#^)Y z6%@+|-dq(_m!B@t$^hxi+2kwurm-LGA%DzjCWIi&np_VW>(3)xte<1ET+)s#>cv*r zH}k}NKH>jfxxtGifMasFtUyC<<->Gjcm>I0EJQn3pYf^}O_!H)c}CJ8S7R*xeEYrt zJ7r$oGD$#tB+$Ip=6RR@epcm)0>0Cfgm@nsTq z?hD?GMc+MT?90^*EoS8ZgMU+kptHyO1flyJ> z=InU}LVdlG;tv382^aj7@eS0_~d;?#V=Io)m>-v5XU)k+3o%Mo? zNWj~SmB#w~A!>mqUwKuuIdJ$l{u|(x`D!wy9C?7E#%zCSJwf*Mo*Dm`J>6v$J|{Fy z&|s)X`PCb>EuC}{@3c>q35T9`2LnI$Kz?l-Q_>u>2oU=fvhVI^y`ZM|k`?Fd?~uR7 zX6VJSKa74%nM{y|N2$BoMeY^IW3`>`)Ci!5V+Yn zbDNy3=fIS>?GA4Jo*kDHfIchPx370rgUqA^l|w_q^w*G>fhBL68y* zm%(3yZ*_{F?bbbTuW|A(a?6~48iaI%z0%5jVo2%n`G6u917$OBI-{!D>3NOqMbUGh zPC32-;H8m@rBW9`|FX~f=!p#Sk}~s=2R#4N+?MN{QTV0lm-)hA+&vYuiLcCWY@**1 zxGA<+NBT97-OA7KJ1D0b)z7lo;xwnoQ$tb0^H?HKKmAxHfks}c6kQiwIQ;n&4bxp4 zl@MW0y50jm&fh7pwYlx<%Lw{vn)Cm~-kS$Cm9FoiyW6ce0j=1eOi@wM7C}U2NSa1L zq#?Gpf=m%m5dt!VnG7l-A_4*e$`lne4nPE%r!WsPL_i2*LYP7bkdP2ECwJlg?R~56 zANP-2b?!NJ&ab|zRjZP)*2=e5KHm3vp7(jpOm3f?Xn-fOA8<$hauM(d%;R;)YM$-! z;sVrxI%yqqy|L`;#gmnmmQP-K3@K5eccjB|Nq`K_!&KH{zpS{miC1k%Oj)Twh~(tl zAdiQ#A#MDz8bU|WSIY=A=6clIUpv|wMH1|DJcbcd%#31K`qV;xi$c#1fnL2QFS=07 z9u@ekFTBT?-mgeT6dg$lZWpB-MEiGlS?=!Oz|GXtsmDlBS}C1=X6pKi1lph5W(9_M zX8--_|GM+Aua)U5=2`Y@=b6P;HFs!k82#{@ro+?^EC=7_8}}!61(FQQ+-eas&jEEOd!{u zVX!Ea8%`;`(<$iPZVoKXkFPrP|9feE8NGEz?ru4Xt-k=suK4Ea=ahWV*H3 zswoKc2+=Mv{^5NYSxn-sH{AACNX-X;*&hkgBz;h{Nz2VADawpf zcKa;$9haYf)tVB3VgS@CJ3pXunF!Jy7JG3D*Rhf}5vQ*~?_sDy;wd=3;x^(1_da`q z0=uSL{XQzXlDtZOo@2dy66Rg60%+?hNE1jHbBf86>d!G& zyN2BitEB_6thc7yF+tpe$eeHWI(6#?U&o0a1re_Z*S~W|P~mDzwP-&S6gmgChtBEV zO1_IymXt3arO?r&Cu8J@>Nm46R-&zC6UwSmw)<$NSGM^b` z6XnaXRqD~J3Khf**}^0p5jp7EhCVEiJ)qV-MO6&kB-}=Uj}031ou_EQi&~jqjCt7| zWqr|%9fG3sjB*u$;rh8qg%$*4DaC%o&AVb@TKe<+5?uOsIXd7B_QL&t<*RbKkP9`*w&R)#jFCTN&84N^>#7L`I_&Du!Ar;S;UP4suog;xO)=~ zMZX6Ych;d-4f|3IgRx@O)XtR`D&FqM8Lso(U%Av=c8vzds~2v94NW<(`8losz{(4n zGf4}(Tm-58sXgo|<5$v7^94b6ou(i}@&%v^xqmDdjOy)4)Zo<9+)|Kb)MuJebKd7P ztGqp)M#|F7zEERgXav z4p}e7joOyN4lA24L=kH=1vgs{S+cu2A6wk*tl!U3?bn|2PL(SvMA-Z|RhPEt!)1cy zc8Z&xEF{p)p!j^BHenlM)cg|cBlMxqspe=8doR5T7eEC7Ec)L0-tO^r0F;^n!3U>U= z>aggnRZtP61%(RcNWg{JlkT1DU{$?J&3d5~(Px{Cco1g{%tKgpSJuOV9&LinT(E2H zHaGo8gsn;SZR#~}a7eC_PW|Q-w4QG~q7D#JhKB*!mErUgeetQy$?J0Gq|WBvpe98n zhiT}tk5U)e-S|CGfco?&r>E3FD=j^J+tut8b=p671@U=Gl%8{F|sH~KHrjp+pUpJYK9C9f5i7ird$J`-_r zTjK3$S&qA>X1|B-Hd#0xQ*YH+_=vVQ9a|j=N=^mU?SkHI^ZW#swe1C0Pi`<&e9F72 z(T{gAN}Us{Y4@2QK{egKI;m@>Ycfd4nHvg&+?y$5*)qCZ0Y3vrMfht_+ZnK^DQu6H zB4kkZ)2COO@h2~>!@Cqa7?f}kq$gjbKKag)alNM;+WN?w)}U}af*j=&6C^aS1x=z~ z9&>u}14v+fTS%Din}2^aLh>#A7Wb;t;f4&t-FyoPM#g-ak8ig4f;x<3`mSa}vEE1E zDznqV&{yI@qOrQ4=v`+1QpuLjZ5dxHlD@6CZ&pW#PogLz9--8@9&H7(Pp#B^eiuxa zAmp~6x?Q#KZ0+Mo{OFRIx}+k|D84hN7D42chkVZ8m{K3V^kk3r| z0rB*_6*4D0@jyOx>|M-qYaRM?uYRYxM5qYnUmEkpqX3rJm_d39q!!-S!^hr=#Kp{4 z(9E1bKYk8eyj&Vz5#RR#U%k|oyhYXySdct~td)CI-*50vG_HtW)4+@YSO!!h;xc3K zkR)Zb;tD3&Si;RLW8Swd=Y}Q4Fl(%@dkO|D9m4wgOU6A=Ot?o5VP#I40{}Gns6AwrWchHh}v8l57_I!Hz)KbRhto zU*bEF9U@7yk9~d0WJ*33A`ldSyEk=kNus!kM#3B|a8rc$n@!sjCc0?dJTYMSP1d^EcbX= zn~uo z^I*a&4Kim)AzrU>JX;uD@TDW@LdOe#{x|nDS59+hz52=e36BE#r}a%Im8Oo?&djh{ z$!z_HLpO9s+u&R+n{X%DiEYNYy3~-}rpeZFDW^_QK67g{+0=r@i5qTvJ8eClKZy|R zQ&J4>yvRHB)TW5>qVpnPCy_FI@fh?!N~x-!k;h5kNO-9y(&e9E*6f3Gn7PDA|4vgm zv%U2_RyD}fz3Re{nXFY4I(l;d#_E0IH+h=}uKy)xxB7Wd@XcG7`{DjR1r58q zg?wYM&L~iUdtQfV~9C_tZHBO$esV2R7;yynCVj(7ia=}rz6hkl={JU&F0dr79>JKKGV317-G<`!)ke<7t&EIo zi$j;4E<@8vN)?8cD4Oyu39Wc&i3SVxHKji%|HLe|9mzD^b4mU*K zaL2nT7LC2u8V#Mt`ZU(`05MTyA}Ns+j|1)gszMao2*yll6{c#9?6rmOhKL6~=LpU6!Siafmx* zgtF@Hnn5#$PLWSmg=KntSCZkia$j*oDHcQy%o<}rwwS9 z2P(9{`NFE?h4W(OL@-{Otr5(!?fGNZsZqRMZvJv3#Y!Gya4moz0hb<(6^b-M&Ix*w z`mS{4RKW=i5x&x^v2TU?rz6M$Y#AqCR3-RSTp2P}@e&lP-LfmDOB$2Wt`ZZiHY2=+ zTq$B^F%2<)zb;)dr;MzBtM&Fn76H|nB!aZKK}yt&aDLf9*O=RQzKg6Cdp5=jA#1qx#rU}sVWwTOpL9q; z@>k!AOziaXi@^Jd5;*Jv#s=_1!=(N`N-OztEegiL>ps)|gYh_XfVjT`4!- zr(EB&Nk)T2WN7e4@$GFD$|Xk6T_Z>I-{vW;fjzB6>5%`De`)A$$bon7qpXSs}@`^S?(QLk*GOOi2IZ2v+m1xhU* z(Y2%dVs?Wh^8f@3xCABYnx14r+|nuL>yXmyPzw?zOll+81C_U%DcIM}7It!6s zEfU;DxeDkeBrc)%-2W8+ZqR>)@3{m-KS}bdRDgL7L80f*ygxA!Hlr)U#L#~7@%^hv z^603}!M$4(4qH8V@#~gX+n-$j>z{4QSARM`_;Z=P^O>F&-Ca7@t=6+o!?+#@h5f%< zb(=nU-xhb3O@d+t*r<0KD+XW84=#~bh*#XNLM?{_^P(vRy&Lpk8CV3$U#(}X)_(|n z9Xc3nTiC7wD*|D9uz?P8^nHuStqkCq%C_(&{b4PBi52^s@r+gW$-YaXan8H$5$rny1WDsLlIkH_unPbfTsL1ardwAJp2mC?%Oy?A z9g)V+CS4nsU+L{@6VCt!aK*t5VFnoKOl*g>;AIu(t|$ZUa(Ij}a( z`EVb2;G<$_3LsQijC=v3c9j%|t5;c6s}dlK&_QXMMD+ubwj`1wbS_@yzxV#1b1)2Z3?ULe zO$Ox|ZqKs~4?dq`^qmp$ULCnlBCa|!S0HG%3-_$^FKF*g!JPH)p`c~f&=|dM#&QLHn4}(xCIw(zZzS!nl&*9hrQ6 zBRHha!NjD%y>>~32MeR#`(^<3ty8svDQ1G;IYEk&Ug*S zkf!?&aUE2PAgwh!GeuLqnkAtZY1@XNkH~%OYN6`?kSA=i@2{@Zt*b7nSrvwUFV)US z!of>Q{j#4!n{}O|kIOncpIg7f#c+1?sv2;PE!=wKICxFEMG;~&70rMpGQZEJ{HFcV zK0!d6<2Pu7t{DkR8vV#^=qA>&jV%f}9**v$ZF8#eCd}Fx$MKkpfrwJ07LdiQ!N)w; z7bT4EF6G?MLb2uglju7&yyt0lq8~CfSIf&)(7!{~;)s1Pr~? zBNq~Es>>U%Yhz#P)~vI=-fSzQ1kF0oS#)h>(TMAO`XuKVD(MBxc{y#cK{>_|e73K_ z=WdNM-~OH_TpgFkrH7IQD$0Bhz;!t~8oW}UN$`KuSW=ac-8VR6d3(f{{_%AZ3)GWS zeLd&Gb;hb_K8aNW%|y*8_1q3$KDzeHYAkTC3u0CVF|w5Y3KhbVtQ6@|yUldkQ%ij2 zB}}{DfW?qZ`rc_3nK3L310Pb}22G+nHRuI`=ZKCsf z@=%<$HPgkfI|NlXr}y_AE%Bz2j!`Oado+HIFrAY1F5S`m(`4dAfT(u;YZjp#Bmc)z z&u&cZ{K($IB$hC28&NL!P8&Q^UP?`3omi=!${D=o>VGVPqS(1S1bHcCbtWbTZai%j z!ETRg5g-M!X;$pwUI;z~$gP1gn16}sB1R4N1>f&0CeHdt`MxI8?6raSwg28~;^Ps}#3&o}L zAgSQSrIq)1KN`X>A>bD|=bN7}3_`f9LKKJwW4O!H*IN;70V{$bFlJbBg_g34dC!S? zhmuwjDI0TgT2B!~w}-a`4@SwBF|EBBW}W?<C^j99UX8k~VKHpCgh#h4qrS!S z#vy@H)ag^;;L=cbuU2A1xTU|>hz&oMhy9Mfh1J5I!>2bL@X$7)I-d&slcz-O43S`^ zlUajM^jgjz#A5-H&gHuz_2;W~Olkd3Qad_p3-2hZcI_SHZEG^PagjJ0Sx_$0Wv<|B z6jie)qiDJ1%=_^wD<9y+sT&lZss|e6?byMX(8FSTBYJ!Y>P+uf+&nB~?AP{)uSHrt zDm&*0t3tKGJ)Mi^_|GeKal?dCie9_P`k8SqT&vP?!XtFx{NRZ%(K)r2ggbu1-Z7AU zmOI(JSjvjsaDaLU?DA`SyD=Tw;B+)1U;4-I7t?uU!{WLPms<@(oCHxr($a}M{g~+P zUpc;C8jV~SiOL4aX734_K}$ac{7rI>SgHSO%JCt|MsyP2_>P|o+ZK)MfIWERv4{`j z*zmj4!ERn*JB6=@Zky_)-wxxanisC;oBspnIA%~>BlL@l6%L77g50O|>KDt(_IL=x z1EWyU!`JjqmQE{-B(SX2s8I8@MX}RP`b?iNCQIj`=|PDgrLBwg)sV6eB|{VfE6zKMI$>YFiTJ< zj(d{CR{702=jm5WYmzaKS@UyI!$a{E+;RrI!OyDo03I+;X&8mqPM{of3zwf}5+XpC zl$NYe&q^~5r&hr6wwJ@V2|mA`wY4e9UJb{#YMn5TMUH-3)6#Nsh^Qk{Q=jKqf&CZx zsi~jd)lJb@7ryo%1Q~^jySlCpO|2p#kKFQkrl@9yzB%`iewSxzkyd8lI^NFy$3F~r z*qyWbyDMH^19B#0IJ8Qnu4y-RV6OF1GUz~!A82-?d0@bKAoQwR#Z{L0l zR^Zk9{MV{nD-|Wojo+PiL;k_KzXGN`6?}t7rVOiFAb35URa*`3y_6&PhfR_z4pUEqI)jsDVPF3n;!O&x8B`k@E70XUwVKmM5NatpgFC?C-bf(bK%~$Kqu^*=< z<{Gmd9}?B52MU5O@g|tnCs!ga1%5%z-Ge{-er%!@hrM(VA2r`IMXPhY3?*5F=*edVqafJJ`Tdcmi{Pf!Xu-P%qmj+ zy}2QNJfEXV?9y)GrWUu|7I^06^PT{nOeAiI3 z5iJO1H#}O+s2Fk6vh;1APb^WQ63oi?cwnyIQ=%p!7UwJ77~LPa-;P)G!@qV%VcH`fhO|a z$Rh$?aI~wD$y)cjpKQ)7MBH7_Z|pq7u6T5abRjgbVOKV$JgiiC4gN}`I?ZCp8INZ- zn_`J;ieD8qb%X~`=)kQ!6sG2xWwqENkZ|Ax7c|zBGkF+_u|_9e+^9c z&Nju1BsjwTG2lKSxY5bHjFvy0-2+o=GfN&_L}=819zA+swAs$agPM>5&9Tv-VLI$K zbkQ>wAoc0g0R}kp+}-L)w;HMz8IxufSD8$BtlD1dtKhBi>GlwGQ!w=UEoKpL6UY;7johc`4VsdCvs9 z3M*TT8xTGQ?{oAn3Qi7?kO1Ft$>|ZW-{#(ofNEJ7RnDn-l&8rUTubUOAWMe`L&pgU z;KSOCzwgSA;7kQ?z7y1Z`HzHF!7b5^?`P9dzbG<(`Ej&3tTVkgt`)i)o%OG8$Wq!B zc&Jy8pU}+QtD`e;@#Xa4$go%d#2=MKOdIT^aqMHIU2{4pynP@Vd?(}Zd5 zZ3O$zK^H*#Y0mtFxT;?kRK#u5 z5j&I>=&>`?A?l@o2e|%YTm1n_qN=*~Q(B@KcuWhzK%B_!aGq7)1u11Iw}=7%F0pQN z%OTVyb+=?KWe>sGZEMGm31j|h4R){3Z>6;e(fin~KN&}*-t}3nPqhV8j_VSt*JW0M zZW>^-^2}*(&XEnTUcH%bLFR(%uIpL%#Al8}_S4#x5oypf!pFgyp5Fi?`~px;;jAG| z)eukha{Fl^tHh$;DtE$5TI!}}al<_VJO)Y}FlvD)Uk%7D{`p@c#doV8=$WB;Vk)c^ zGPiapua(ADQXDy&TPB#)ZzG9Yq8%67n&-{jE_fozzsB|*tr5rn?LXa>*rxj-S6|sY z@+L_*z1IdUztHA+vOigcm8|ysQ|ue8QgTTo`ydsFz~l+a^sJ?a&yZ2*v=}D^{|oJB z+EDaiW95wW`8Yt+px-t!*E{A@*rP`%CZAzI_J6@jpF)NLbR`+-AHbIuT5T)zgE;jCOv1$yz4SP(H? zZ;;lW8edH*ZOx%|>jk8H_X<=xH=il$!ojGG@Sx8~57ti6RrLxWex-{;FBqNRWQ zB$zu!F#1TzF>cYT>s7M#e<-K*L4N+@ycwe?;}}#JB$Ghiv0x{=EX3_EU>mvqYSa`} z@*SwOK5e}jdANS0(viHKK>219*$yglSgcw%dJU?kX!`aUUOQPGDWl`JTd<}jV^u-2 zBcbh#UB4=V#g|-A71}9_C_xv7dOmAjFERIOBqiIp1##Hog*7|t0I(3MTr^+{v%15c z2;z#9IB1Ngbf(8AF{V>v6Uz9SE53XdLVt=pECv?x*%Z?wr)uW*woBviNCWyV&_{wa zy9Hi}j`ESlARs`~S|(nxsCs6no!W*W@aLoukl%8}gxugG2Ht5wUv_BdHrH4TT=VMN zk>nFrg7X!PviqVCJ;w_tEq&w4Y>*tSJXZukMh)IwIVx+>=?YrSBn%U{lzIq;Z`oow zvPxK!h;20*D*qEXL82A-ET4aFY={Gh6oG)LC_HwNS^Oh-8mB=;Vqc(8heLvBiw%B` z5eW(;;x#;{NA9|%*>c>{wF#!O(O0|YhtiB}kA@NSD@nX628aswhUy+&_XkucYi zo5?}~7sHUn_YhQ;Jib|$9c*P5hsz{D@whRwh9Cn2j$Jiv#`!#oH&yLO21LOG88G^0 z-MP~si-I@~9)IJgU}Dku&v+H_G&NScDRF1!=;LPa)KS_RvT{f>;`$egLZYS#V!PBiC-<9g?;+{$7mv)+iD}4PDGPX;PLkDpPJ>uRLCfB zL=iv+o?Zl6L?m`fTE5$;B&$F1)C|7M{0jAV@~D0syGsXc^3c0lF-I&+ao$!xs6|Hg zHdmLJuJyWW+ZfH9Wvbqwrmk}KD@ks|4edDO@>1sYT&$_B6JxV13z{8#4&tOgo>&c(&63Za2t3k%@~gHX^um`E%X|1SzB5HbyK1PfdZZ84JmM7O=A{xhi+)%^y2O?#568Nv3R9oyNqnKSs zl4&;;9+!6%|3Y)JCn}Llz4#aG)nl?M5v*YZqAD%q#SP zAha%T%%_yw$r35YO|Rlw6-Kp-KLb=SfsLfYWm%0J2`G!_jC01GmVA>YtV)zh5D$(1 zNz6uHFg@xX?2;wLY~YA?1YzISqw$ZSU?4I9Jbz+B)=Kwy=lF=vbM_vpaYFSqhd zm!u|=I3DO~Zy6%q*;P}-pbRaFr&mre|HQ{|GwC0o2c$zSJHLsF zcpODRqFfH)X8L-IiUkgoD2svceTB0}vs7e>Mp1WV!nqWtnh{}J-+}MPI^+ivGJuYw z`?de<{i(7?TP-&mey+i1rC}!a%RK4I?P{a!3L!6Fe|=lS;8nG;#S?voZ^&OZJl2}+ z6lRXNN`O#KiJ4eJZUTwA!aVo0Dgg(acTsuu$G+QzwU$KH8?V`FlHxp@cHw3xw>bun z8U_k%!D{cJMFotO)9+6jhUHE)olvM-D3%sW$!hDi%UZyvf3b&}_6JPwi+AUYV2Wk+ z#)~PVwk5k-4`$|v{QL7V7gs{2>e!~0xi6;clQ6~qfvqq{rS zIr?kx-7Pqi@@^%y*V`YA*CBS1VqEx6`hk(!k4SF(qjU%`8#*nds7A%Y28V;+E0 z73&t($Zh{M2`%)0UyU@CxNS*^u~>V0_tHfN4VqEV6JS<8$sUj>jQ1{Q-uq^0qHeee zV~)mrZ|)FJdnu3%Nm*lZ@{vt|vrVD+MswE!ny$*16GV4;I+9-2JycSIr>=rSW+FhP z7X(c;4(j0PAphK0Z9*tEHsZwY+Jyr7&uOYfN*ZdlG`*j8vDkNlz7JL=u`0fvjrq@l z`Zo}MJp#fnP$H-Z({(a=0AM`bNFUC(r4bVKyb8?~2q9|$KH@o*4qDs6I(CnoqRGuH z)Fr7Zfipa!f-vnz<-_Xrah<6u2TIG#!3)6}3aksuX{oUNpaF^~Zs8BXIX+c>dAg22 zpY9G63ST$f1Ll#pz0_WK2u%~;<H`ni|OUiJZ7T6!07hgwY+B2+wbbsLtSl-cFh+2@Sn99ILj z9lN8M>=TfiU)mgf%bO!m@px`p@=uo|YENwU$3Gt^!9;_%N|zYy6B29hf%7VM+WlQ> zB1B&eS<0nl5W%y{_4@q52P50oXb|23-&pR5eFW_a9)oFar+LreK5(2iTVyUhgS2YQ?h~Y?|ShQM-02PR}^U>k8{F$C7Ftu_RD|b>(ZI1Bf z*{p7Z_~SYUfpJAfkF6C4{;7uqH-kTTHIIx2 z0-rMjL37V^;ESZ_H8+OOJ9$KYJm_*UzPAF52MGbB>s0XwR=BJpLKnNk7FjMp+jf=0 z6x0kt!qIaV6pZvf=^*K$y2hgc`ZatW|{JmdrJ9nfjk(G*!TS%7si0B?jhiit2_-(DW9(I zb;Pm+C9NP5HpIRF1SUQp?If1C@Hy!<5Zlt4=PRVwt>$1W1mOI49H%(wLzLm zs9MvnKZUxwDWPcq;=T$ZCO=jB=USBPU0JuYctr{aOlQwCA}Qi8Xz3Nr`QgQ{OcA7M zCoOygNQ^7r+nS^8$oqx;S15QUl|8#f4ipaL|Ge@jSf{G7 zDzU!T`%8He%lp$6>x!M_KZZEx?;GFb-ieU?Z#}KOY`=oX7oQ{rugUEcw;j&d1VN#*E{s1pA1c!s5BxT)K^*>JSz~vhfZ*zbCzLJiYRN$k}3xT}lQ_{#M`rpgl zG0IgQk4(XT8TIirvUj)JH*Q>Yeal2LZLQT?bLbM6nr&4HHsSsRQznnz3(eC)P$|nt z^)4{SgC~|Yb?w%6IW)YY+6XEDh$tHaJJ%o3z;(88Vp5K5ofbN$ul+8beawSu^AzcB zuVboToNGMp5i#4>{x*pPbho^7oft-6Q;AjPjUO3TOZym+^3NxDy4GZ?=*S_rm&}?FtTD>8# zge3NPTwa*dzLxV%Fca@s0c$ZDJ2}mP39;)1;;qS}vN-?6^QhCem??mI(Y46Atr!5p zH~IichFS^XE(OO;gV_uV-@p7a47w#?P~0K3P+0x1^(6pLI@nq!lbxUAjEg5dCxwUpMzWM#Ak1gv=UDVWFd_9^;l(G58+SFEc{!Ts*q*D9~cu3@78l>8zGbWMzY&xN6b zCRkd4;LWG~ftUoN|B@m}(YS|Ply zVY0i0q9$3|XO?Qv!I+VJcMhNmA!2^C*rctd=M8KwO=|z@|43Ag?g~gwL68zMih*^G zAZwFyG*eKTDE?8@Jx{wx+nJrE^W7x3dM%TJ|#-Lf-3% zm)@LLz=ha-;2AA)aBrOn%W}Ld;31~`*VemJWbwI?^bn zFTU82w6Ylbdcn!54kq$ra!I=2_NN=tla83*473+ze;9<=;T{#;4|}7Me-f|v8(}|J z?5^H+u|+$oGcV%pi7U8Uj6Zz4wRK+JcE~Yn0l+~&DGwvoS}7$qTaLR`Ub(F?M$ti0 zBs`oRTA)VHBADr6RrBE=%5Liicef8I1V(vBuN@;g)cF|%t$rCMIUfzoBp0wAJ6SXK zb^>tb5J5aFIKaD9xpCpxLeflNkGkFx!XnKy-_68JLXB0L!o$=iA}FF)WHW;x87jg7 z(3}Y^BzWhU)jB==3Y_vMfj=L(A`H#j^)|!FLqC8cbi&c*hDy6tIIyt3FggMzb|Ifb z*(Y?j%?tB4$}5KIILE}qg}(MIH{|(B%QVOSu)Ncre^6@=Varf+Pt(t3c~lJd3~us% zXGn=7l?O?(S!`a8ls5tS6R|KkH2d2v^pR79$ui%btiVhU0lQB`>f9`LsAigu9Wrh^6!`JfA#Qs+iS7*bdyk=8`Gk%j}7f2EPuBj zwF4x|fYXxksjB5N-*M0ML`4NQvn+<$7C!8N>qkfoe0-!QOO(CMR@%r3oVMSH+oj07 zA%#30&SUXl#F+a=+~ClNtNBWU$WdbU46{m+3GLxcGu8p4Jw~APxv--ULCap3CGV(B%d8dai(vjzshJDN3LI$2>NiBNf(kd`-ZL zFIWaPiYsJ{VHeDY-@`OjPZk%s5R5F1W%JvMZLKetX2F85n*JpP^7B7k2{X5RCy>PH z^m-aOW@yh4^iQB>F2V&~JUldKSC<*d4i^mjxM6-$oG*~w)QT6i)4`|HRAa?OwwGmr zf_P}x>vzS{UBn^T+fw$f2Akq~tP7wbgQ8yNzvssD>|>V!}Q0oDUr$5>MH`onBh1eOR`~+GViAeE? zD5qdJ*IOHAAjHcC==`5t(^qH|TP6*IxgEy+NPzL9m~WCbu5-MzLW>quD|qhU&I_b_ zTd@8?cffbroI}07<*B*0U9#uJ+jQx@1y4t_L6ZfCCT#e&1y7O+uD<#)rh`BIzS0_UwoZ9>2W_Au zZtV~#j1^nx<+sKk*526&b^^Otum_O4mDzy~IaSr!=cND*AsbeHzv1%8ncrdj=|URcRih$ZvF3*{Qodr&;1`# z?o8oArC=f=wKag10T8l;zk^buP%;qN}N<*YWLoTUZ-*-VG6 zCjUsCc-pOn<~Ap%4KndhYpYH%ZPZ1UwncNA4-^z#lfdbBfC0>88rFI6-tgEjkBG|E6>jQBq z`>7f#vu~AVP5KVVmc>5kB@W$~=6-3AfD#1^_87(dMKcU7&VqZcq$MYY)7wl|*Xq&5 zALcP)As(PtV4DFnj&`pDBH}>K@rxl8dPgqO&F?rq>jy zP&o{cGN0lnm17dfsiAJ=_Pl(K>F6dtSqsFx#Xv(^y&zc-^Fnmq8+SAPd61IKBDa|Bc~_LQ+x}7d&XzgL5-5^ zOKKb2{Io10tD_&2;Z8*s`W?aihe|ln&7mMQekNv)klJpR(IVf_mK`>H<=Zs?#RX8} zXA@p}_kY|2T7KC7oLRgZft{({z&4TEsYVMe20*VC(cig5(%`SaO&&e0x1p=UNZ23R zUD|eE%dy@~#c1u`tZ+a>3H5{{aZA3X8Rdlzej2V(r05v@eWvMPo!2b4;7f9eGcRnO zH-21G?94VFiJI)sX0`UXOu{%jW+SmmgjYaH1N-*1qWa&mqfVfW!@wm%drsSYeRp5CI~Sw?%<3nksy?B!n`yFemozDe3r6XnJ&}BFx1 zf>=&In$NJ6Q{%jD-Jz**H`H#SEj&0K8)-*CSBcyvchj=uEf)yuOJdv^#D49T*f=@h zpLE!hFA;~uT}ppEK1zGVy??$X$)iV6noi|>|QOa)1Fd8 zzH6(Y822_;0->a!z~dEYG3EF???i#sk%Yd;Bc=rdGR}rE1R8=ZSoYgDm9=QNWpW8tw=zSh-Lp*YuEnG zcACa(O`B6RVeCo~u~XFaj9c85#_njP))~E@VxovKw6R7uLP}88R_i+GmR%z;OOJN7 zBE7{W#3g)Dq6A6n(n?5ZIzkc>a@#L^X8wRZ`~C5I-uFGvdCqg5_qlwY_e0Yv*zaKs z70G^TCl*b8Txlk$u@5GkRv%b~6gYzKxT~*8i3maRg%rL9mg=3W!mk|AEez$=sLpK0gz_Hf!C6ZtS|4ByG#KTMpOCY{1cgw>=&hnUG{3^x@60aqJ6B9a$Iz@x?8P3R9Q|3%S_I_L@L0 z>Yw-r8rO5ruWZQ4Ea=@?y3w6T<(GEFu-e90INr#4cpn_4#-rq& zh?y_L;MW|SuE${mICE9mXoD^hJb=+o)_0Zcv|6$VFh(s@&{KRTTwaz*~!Y`z6>k zj6I6QM2gVN{%N_2%D8f2%;(V|(%+vpQ&WvYN;J{BSw(gCI{My;BQ&s@x}`Y7_j1^} zqbjHD?g@0mE%B0d{2h9867C1GVMv176DAa0`rk|8=>NR5-Y^oXexj((o`)Q$ZIlpF z?I?|Tbn|i%q92>v=47p&m-sG%PIOO*q2++qQ!a~xb)BPn7QCecIy&T?<(uhYi){~9 z00_J-**RE~c+2vru(b+v93{Y(S-C&)u!+GU$FYJ5+!30{r_i)h8Vw1UyHXP3&Kbxu zcV6X-$bgzPIGuPI(a&EIOvqo;7xBf!w%%!6YeCihQnG|pA~pkFhnnd(hCtY5&@`U$ zTO{)Berdxxi+A{5KULSqLBGVh06<{#bO3{bm%Z=Qv3% zU-@PtJ$Cjti?uhW`MlfCP+I!O@-Zj!#x-Q&Y;R~uI_0#-q&*!3X=DpQ#0X|`GTgmx zw^cqFYWoD3Ec&i17i4q@j8u66h3Og*Nf%u(!;|yb6b{I5?Io*gobBzP1)v*ecN?nH z@R)+!4K(qm-+u1KCb8*!`dHRDLXy$CmQ)5?-P$$BssI0>hHc$tsLM#8z%QDaxju2T+pxa!zSKL~aLCYk zk;Qz(a6~(zSAWrK#@Pn)`w9?Na;bhZuhTHAkZt7XSG`!iLWK!pi$T$wHXB(TygbDS zuY}lU!JB+pnVaX~R<@M&D$5f)_xfdch%KrGQzGT=57xJkhQ+8jr<}Hh+XUzUX>pNA zOF_;KxQ<5RZh@~~EysZCi=JM~pqG=rbDBTeQgZcI>Emum;{n5 zGWK=wSL_)XxTqbyYE!)$_a;9GWvTAi`tXlCZ)mWs)n=(f=A0vi0Sjp;;=6#mbWKx7G{z(TcJO1oYE6 z5VomnqJnNsbN$@Q4^+ajz*+QP%osZsnGV)gLYVh_@3t|-Y|^f|3jf-CzBVh<`SWqVdbr`E@MHzi?5xa(IE z%F;{3X>sMx_U!$nl)as;Z~lJnTM&#o#mjjHSm=rp4B?!AX}Y-Jnc{?sq9l$_Jmw~8 zu3oFYTL6G!#soW0Cdm3mjQYGaYzWjnatcw11P8m zGNwX8a^gZlByzUa#-T zi6CUbmXPO#a}%%Ox&jJ8PbN{?^@X$?jN~`LB`K--fgAcNGw+~kHtA;LcjCEku2#DF zR}#4Gb_k(XwcK;qaE9SXWFr^@A0C@X%gE@3yF$UgftA*M$aV%(4S7sRh^C8cb)&7V z1N*VGelun0+~mfSUj0?LCKMFr!?!hZ4DG(BqENfa-ip!aQ1U1)2Wn>^N^QiFkOy2y zZ-X&Sa&7W3v+{OZ1mjOt8Ja_(Nb(NON~NIK&Pn=cF$`Z3as^>?h`-ZEZuY>qSX}DC zcvXj_kN*aCC7teQd`)M~@LHM`EF9U_NAN9I6#o+Y+UYDMqY5*R-;4Zp zwbC$Eqi!baP+Q}7G$mZk)|6}E@)+i@Ui11tQ|d8U%bt)T)4C0bkgR-9fPU5-xmf+S zj+9Ti9UaMc3}gO*xRv2)n6&%{4lz^M13%@D2QGUU8-kkT5s0l1#|dJ3J`yIKsDB9R z!ktgvH}d79KH{gd#pIv`OL0p!;}5hEpq9-?l4QOjkRy=^dZ!9IFXkenITj{uriaaL zJ0_C(QK?lQxf?KxItNVqHs~V}m*a$-DLc`~;=_0fv%!kZmYCC7GBg43p zz{U!|5q$Xc^O3kOFQ#)a?E5*VkHP|cEWMt^DsU(yH>OYW7MsZv*4Lp}KAb~nIp>(0 zT}U1*es^n0i7;&T72ZW+nfG*V_EUl~Xn(IXezS9?cZ}NPEp>{K_B1K{HE{(60w(N7 z(q@uWdMmGsIc;Z-y7O=AL1k}4e{wX8Jtlh0g^Z4*f~@r}ez4_rsl2v`7eh8N8zlL# z=)hM)P+Db*_~gU79InYc!dGY~?*khf9gZ?O!)hT9y0Y(h8CR7^4X&?*9G_63NZ<+PoVr}ekz+3Q_6J-sScBHItv2x2){ z>5^N#bsjtlqGP^7?RCIZ$-vFqGNu1M_wYv67j*13{%*6HaTy5a=Y6Tp<#K#A*AwGZD|$|-w&VnRu$PEC;DTjwydIjfhiDiN(){^ zKIC!5+Ip3ocJ=h@5Ysc`08U+tpupF9G^z;Cny_^A~sK=?3YN=~e?8rFi4-Hl@(Wvcf1A$!s~RL);j1RR8p_suQu^piuMa9L%#m4!m2gRvchO57y;43dPI~ktI zG0&UKu^Uzx##*Nva>^lAY0S9}5J^+(zkEaU#eD)roqmgAi)gC=L8wXIH;Z&gHCsNT zKl`ot0(|1D%5=Nc!PWPNw&#N9@UBJAUJpt{=5>r}rl@7Os26SFXV;PmanV2WmPWRhX6arnqJ!o`zlk;szB%SFfq zu9>J|sPV5UbI7v~U3Fw;&cupr>ozJhB+}J3XdpIV`@kx~W<37s19MDkT|a)_N*2`s z-SA|dlhRH-pCrE=zo=)qN945@?9|s@l2+a5Z`i(r2=xXIt**gXLG*-fx|{SuLqB?% zMr;ayIyN~rd5{~V>5}L4i=;)0)VE3*@Gn>_IJAleyGcEL#`jD=(&1T~R8on(8i6uH zo{vMou!-3lJpEjzY2yjK^Kab~&%fojTNI6J+2wLh@-(?8y2QIAzxRLru7xMiHBck^ ziDZRQ^b$fBYqWKKenCWjwE7D*-tv*MWi=f2QH|M>&*P^P-CuGls`Vwm7)>sZSyj}( z@tdfgFdA1H*D4m>n%-vS!Ic3U#mOMI)x{iQ~Xh>UdpFKv>xPq;I}F6pvT=PUd|C~e4eqmMvxpwc9h=WL`;-(I!#AouCW*(kgj9>-VWG}ig5@&|P$AyoNKi0iA7nWtI zV;wQ%={fAxj319&idRY+rdtl6Yhf%CDy=kb?C|d#k75mrht#r%Cd5QLsix9*geS!5 z@faKj$>h|35uSK4Xt#VG3od0Em*FH!rnT-J_2xg{Lcwog54eEmgI9vLHlgh0;?OX! zFm${isO9Wh=An5&d#T63r@~ZenY32da zYM18W3jJ}ZjnNeJrZg{UwG}USm>8H=>uM18mLyR8BSZW?uE2^GL z#ww_}jJGZMEH%2b|EvwzK~Of#UsYYJupyUg@0*oAsBmDzk;Q&iu|irE4WOc3dmDTcdRW?XGT;lm9!cXtLk+x76|>hdoRCUe|F6J)?1pG>?rQ&Fx)(nkn2!X zP{5k6p%ywqx9;YcwcSufr7LUaE7Vj`-(7EX@s+Qo!Q^7_Sce;3a-ux%)zN5nBi zJG|j8^Lmqq6_Zp{$#u50%ZhDl#e*iEXy<2(xV^CkCiQ(4Ne4Vmr#Q!q(ks)w>=m^Y z#OK3Dqe6sMguysdjBGlGRbvC3IR~qhvyxN{h?>`DC_9~JdEqMDMU4Kv+#|(|9E~e9a>aw4l&R>RH5m4PpQKmVfo@XT)}Vz%64_VD|1OQ`4RRsRfobX zE}gZ;iKCb<9m}i!s}t>&`s>}y!-DI{IHye~lNqXd&wafS(Sf!A*@|R!K0j~+YZyBd z$x_#;ZzIf8IU-9Is8c%}s4s?4Pj{!p5>vT|(XfRbG6{v}`1OWyZHud*e?Z-=8MywO z?QwuQ-h%pyvOcH){V@|7V+Z<&|6|ehT{WxSshsD+&M!|r&a)2d?02mL)}_v)X45oM z6dj<`COx4m3w z3>1`~DHPnVdt`v)?Oz!1x$X1k5jNrj6e92o6Zkl%!2G`Zp+pMo?>0gla1BaOK}cL2 zI4T&}8W~yHy|K3E4R`3Ipv6HHyTJHkri!Zes?t(i2G*7gdWP2e zMhwoDHn-zI@i=nD?%)F2x2GA&NdCCR-kg_ARa%Zj$lBJ3 z8{MkW?U z78ZKo4thHmD|dMi8fKL`0ejr`df%0mjfv&2UH@|F|Lv+|XJjj6Z3#Top6^~@zdHZ*;;)W8jJI?D3org;^dDyd zPV=GgFy2Xx52ed&QXUW_fvK>pB5(xM?Dh{<8u+6A^9ZzI`Uanays@C51fax)1r?p4 zH>VMupA6%*?xDkwQhcU3oZ$vFfA9$i2!LnMN3sb41qs8Mg3`qm;7|iT zq71l&Ye$BKy`z}or0Qj=9e?uugT>1m7a~^Hl>|<^Wp;1;qKoA&+n%v^524{O1fc%> zVtmYDf0*u(O7?!=Z9H<11Bt{73g%yYc{w2`@PK9b80QSDtpC3yOy`y`Pd^^$`@!%~ zFdz~_nwj^){qOYLU7ex^*dP2!>ir!Nb6`M{bvuds!v8|m{!og<+l~kI`-=A+tuRTb zm!64b3jU9Lx}$4A6__L*EGzH*99yvghLy|`=ZXHi!MwZ41>|TC>dcd#+=-0@3!oL` z-AZ{+>i^af;6eg8jc9LmFSW3@a7YqH2~I5%kM79-i|GIU9ElyEPMuRb$yxU^5)%RD z%_3m`ukDD43=pJ{HeA>J^LC{GYe+iNR>itw;+>TLyP3{C2L@AnXZIInwg4Dr@sxn< zzjyr4t3x4iy?y9o(o-rBdR4lnw@wr?qb z(+F;Q&vEXwVr~R5qnh%WslR&RzwqGir7Bhc*!NK9Tf)B(c5B}>+N|jJn)09A`M;D- zn*`t(sy12Z{q13n0QU!aqhtT2hCx9)0W-4AdS*v@H<)uGlpsx!TLurt9j*Tx_nmWq zj*Q?Ep8Ig&q8#$H;&g z2{XovzP@t|Rc`@BVYh0<`o9?b-_8lR-r6_1!}h(@!t`$K`;$t`tA9&aAAbgqvn+Yp=u-?`EW1F?OS!d!YYRPpN!{jC=3CT*FaBD zb2G_ZNAGT0zc!yAX!~~ABXRV{tGYu zk8@#W0DE#W7he5^Fu+b}S?i3F9{o$g>I4CfvDdv$g1+lW(gWyN1}^>Mf9?0JfEg_* z%RIYxue&b+QfH1uT!`*yy?f4oa_KLsv#WgD@!Z&j;l5)fU=FsZn)S8?j|z#0q_>nOt?7zny@t>P||9R zRny#Qg@DK-fH3))CmX{5l|TPMtatbAB+EPzcQ5aKiui3R7U3sz&%?iKc|V}Q0KX7n z@Dpv1YJcM&QcYV@@)q~a(epp2>D^5^w-Bl1US<@N;L(8jbdIrMyv&FJDhb_H$&gu=_QbffrK0wR))i%-WSLHLkGt8TpUe@ zc&}sKpxo}asU=Zj;N#HA_m^OLP=a1DimC?EVGvNb*q&*u5RiCzr2+9qG*r0Hacey5 z0{CL=W3z&+a&Z0D3uw4HAV-Hd)%SgIkE=2#yGinoK8P`({K}a3q$^nq#@2AS@^USi zDYkUKE&)9-YGu2Ryoe`4r_@-yYnX}sPX6~)|NmPgH338K!&&%;UrU zp@7~Gs8e1k;0TdXEZSCSgKd&DYibk`+@C}T3B&;|UjlkBHG|HmKOzc^JlS;M<5!3; z64h^xdEG|j{oa%H-mLTGQAw7K3sl?9DJ4JC_~{tY8h#bkS4{>PdOKHW;ItWP+LO`t zr5~-x|48bb473|%j?dY?7{g71>$%0aJB9g3{(16qwbO^8+iWJkEP9H@W^3Ywe_-I? zO>=YedQKBrVSDTW>G`Kl(DslqgEf$?M|DryV@0qql0G+*D84h)Iu(G?Ck+}-JmXa@ zSw`C*BpI?;d_i;`=;7$b$+k}qU7)*r~3>ICqACR^kxUQZI@8FoyJO%v(27u zKF>^B2?zpC>>KT)(~852C?jRwSG<8eW>`DWCv`bbszb9(_-E&`jmhC~?4$XTR%Q|T zLD;2WHAU?nv>8@}{x5&*9Ymk<9i$R0a9ACvhiS{H2@@O;{vH+1WYbUo2&w8ecB1~! z65%lW_$Z>(`gmQwFjwm6<#&)$(2pnhcVY)Bc_1-;v2*Y<>0;5=Fu|d0`Jqp`L8Q+} zH5Jii9j{c46z|XCFKz3K5i2|B(YcgY_HHB3_ofd#>PR9Rc^Oqw7t-~%GzV6e$Reg! z8fE-&IsUmKvy%C$EuCv$0y7^&j!2plPrhXKrzhi)5O|+KmX^A>yI1C_r5LrJJ=&zt zBLGZgP|^m=r_?XK`zH;dNr>zQ*x9*Ce#@+AbE;Z$HrH&J7F2CZiM)FvyP?6#|9f_G z!+=+@>e=Y|D3}u(lNip}*tiOHIG7zELSsG(on2+VC)BeGX?p)ZNW%)V%7#Xf;oUp& zZ2?aj5J`LXv#4olR@R5Vp5+|ybvRFDX>mLd)0Oj;&%t5@;O=;hO!HyA?Quw`jojc1 zzsrg%1xOU`LoCAm@At<2qeb@K@QT2+j<38vebL9#_JJ#vheoTExnq#n4=`9aF4+ z9I!JjJIUwJa4pB>k7JU>pNO<~95CDdv<+H(*#4|#-z1W2>1t_@d%vn~`#jNVQ!Ex- z`GSw2+}7?B0WZF#G{%!T@;?5sQ7+Hj3Qf?6Gn$gP*|O6L#c^W*$n5Lm)*2K+l2Z_kkY{dw4LXqL+zojs(s`}0Z%itY~*qC*#PauwAm{z*zHhs6R#NRf(cmjY)$Tl88y1oDa%$eq;@nN<#I<$8J9qq;XPuU(;`< z$bOlbz@S=f7W~C+MJumd&=+y>%p_j?Y~P5o_mpD!`s!jSkkO=Lxkn?PS+Y@$q*AHK zIPor2B9VmEd9fp86yo_OhDIkb!$^%b$rMuTuXH|3c6d(Hf@Dl>8`lzmUm4yH&&8N|~*Y6uOAKO?S4x$%{Xf>j?lxInEF#I9t*Ovj_8kKH@vg13hg+A|9a^f zuiE}&?%78n+ufm~uTgRCt9h>%e@=uf6`v%3j94Z|=boxnj!oDfucaiM_(zxH-A!M7 z)Bg6MT!tp{g6|+-EWYnAWaMigyuN{f(WtkhaQA+jW^0jY|eu^d19jS2CIh9hI2{i=1whM(kZ?L(%6M^)_Fp0Cf3 zx1)x>9Nwo;9LUZZauQ0;u6pgHoCb>TB|0*5g{3W2UYSMpm_9stmoijbO|=w1J5R^0 zhAKHfx5rg&sN0n)9wep~Q|k*_FZ7cU=?WB4?QB@;SR6_HH5~Gs+%O1f?==Wt z&dJ6Yn{QZ3mkTrS>c;zepBVp*cQBNIW~Ejtc$EW}OxFz|;bHsJ;b_y|#_;iyosS+| zJ|OM$2$y_CWW$LRNS>U+&NgAXLyK5gN4dhUU$1#S0)hQ4b18HzEP@PyxO{n881plj z*+(^8rEs|;=clqn4;-NwhW9|ajX%GVEVZ6#a1JOJ+L zXwAfs&Xn1-nZk+EiOS>|Dvsf3Aun(%DUDJM#)Pe#^5%4IZ z>!{!7a9_+$;Bt_VfKW3~Nm0XZWHrsh!H22k)kA zGj#HM`s4}~!Ff-B5?b&-hB1Lm9424TFkN)-AvV;Xv14uvX7Yq+BveeUb8?og&xsr+ z_9gWT=SAW1dXwaeT#I*U=TAv#oUBYa8&S}eid_7(t@OyBA8zz*L~5QV)n8Dsu_^j_ ztWRlA;&EFvQ^)cw*@3S&7&kI(KEF~yM zg7ZtB1OK(K!SqO)hDr6sc5Qm{TmG_$==Js(MpY3sVp{YU&H{+(eDN$wt+JP7eflxXv$C!%LJ5g$p1se9OsuzQF11>wd%}e;&{9j+f?x#_JhcoP$W*X#!7=mqiBmz}bJM6MyJ`tOkH@ zE_>YY`~jg0EZ^q#)idm}@hsj__z<)H+%!ry^9i_(u|z~umrxPe7va;j3;Y&mOE?M{ z9`zSi&TB#+f+0DBnHw7*Oh~RmR>kvA?2|Qp-D+hSvtEMdb>PX$#RXZuvhz(u+HN-M zV=2?Ub$m2Z+KZ=H#vnT!I^`j=nKF|io@>L$;F-%M9K+%Da;{>dhsA{5iB}m?7ni7N z7t>rCc?+NEvaF+@N6Ki`aHH9CDVWEnljWJ z`mAFNohtVk?hE~Fp2_?iKF8~UayLe`s#W~yI>WtD&_cy}uMo$t%Ej9rb2l>ug0kse zdKr}c6*8gdogIz766s#OAOx!EKgBeIv&B1}u})gQno|i7Ynrs%SU7wjQJuT0SfuV1 zHOG_>i5}wv}?ND{VE25Mg zO|KmR5;@AQzc}s)`%Ip{tB~cyE!Kv%mA`k)6fvINI#lGDTo`uL5)@nN(IjgeGm|Gb zSW7t3WhwGwbtp$+zr#cUhysoF$#$>v=~T#;dfnF5ptDre8=jwx7G=?CByxe({o1?F zh}{k!o!Nuj=ftI#%`DORuI#i1r9f*v7IAJ3CLz@%+UvN`Pm2wzHpg(^D4Y;8hG0ZIz!KQk znDfQhN@|_jf_YbiXG>V*bFKG)<=>bPQq}G9mBjm(e^OVc00_#dK_zK@8K^CBJ#d++ z9rpA;*6mN=mx9}5{C1$Q5OUr6ra_ZPbkp)Hj>r!l*soS5SFuuKVPKHVK98eX>WV6} zdV79|TfNeNm~U5JftxHFmB4wGUqQ(2)S(cZ+`rN*NVdW>@D9DLeNdKHPYeI?V{*oq zo-HV^a#BXV7p;ipL{Jtm>*?w3ze&p~kWjujwurIdsb|i~-cVxLo*%-&|mr!=0psmp*2ssJq5oKRXhzTdWd)+LNy(jzWZQZF#TX9Gh zT!uEDN4u)^m!m{^4(j;lYo1poa(N4lz9xvo5QCLM^&ZN7lNfk#=lVf`U!hq1o@hZs zI$Ba4Ys{g!NBwyYUOcuS4c3g=z)i|KO@D{#gaLT4kxUUwgi;i&!9B3~ANbp^;zjco zqx63D_;-E;7?7mS=0l#5Bru|3$NsPgX*}$Rk}Ut(CpMRT`3;^jLJSDYjI+LA&TC1P z&Qwae$jc<>OmT9T%(vt*2I^nd-;`51qMHlQqA9B8PID73RZ2HfxjK}YTvJKkjH+%I zxP=;a%%7S{I3Z+c!Uqqf&M?6U@IMtKDVM%JBsV>6Z#o~lUIJq3pi+k7Q`3tP{(J_v zr#Y_Wxuq5zuk!kMKCxKVJRp1tO0ICdQd%3V%^Ek_;&$314Qy1)?Q7IID3IUF2^x-> z`c}2~*yww+xa(M)Qdh>Svs3h+6Qjzd!f(fSnCHCij5=^4WF4n6h-ndvp|1c#wThE%%}#7?FKL1^YT zN<&3{Y02M72SY#rc`6`fsF3&K#qswWK7(qvK0dHEfZODNs_XcyqT5aA9H?}Dfkj{G zVF0tr@ZqgJAqwMlTkuSh!=PPMD0#)1qsV-{^05;u*kB=n-qf3tZlt{wD^Cjzlr+s< zZa2QN6%3aV#Gat~v4hB7VjZ89a+K=jvm6(NiOL(Va45|h&E@XIB3RMFLi7I1*SKSi zMTR|b!|RSFa_eQX&pOi{5|=E$1a`d>Q0Vj~6v(n)U-~|w(Hz?>J?_zHc!uveO(71S zfr%;zuC|BW?^l1$>Iv5kcsfiAxtsOnF99eBk_E0F?U}cM3ze%F~MSIl%DP$fg&C|>cHv) z@~=m?3u50!mcX@sQG%f$f-Dg``)RaLE0SC^w3Gj4GwdP%HAS9sQPl3PO_@eqK<4(IkP_AMRZX`M6&1jCqubhx6$5#<1O|A^X}VQ zI!%bX$kF+4?~&z8Bmyi(``x;Q2d(+RrB>(p!uO%AhL>626)GeYEccncVQ^N2D1 zR}SYQJ9a{21xrt>YiS>!{nc#;k@5m%kzxAsV0>24_2{DMeFnKpa2uHtxjdEP9&mkv zs>OzD>!nl-iYNpAcB?U#qh!8lwxyf}i z1@pTtH1VEvEVXZZOZf%h9h07)u6ZCXO1%jh)-Fg-cAu$+IBdJ`j-00LpM0iMwv+P; z&Bw#M(GXQMDiAD_RR@a`vgJogCKarlmTmr-QOa$UqP(_k&nL<#usa|D@RV$?7%={k zh#j*ehn9{Vz;)8MZ5A(Vh#rzgoox)D5y4IqP~E3v%P zSU5OTky%oakJ9Fa6tO{Mi0qFc=W8oEi&3NM8?kLccnp1PmZ_n&N6tFf&ozO3V&_>n z<&U6WvsfdUlNQWz;nxl;3GC&fmH{gG?cAp~2cO6FE||_Qp0;CA1*guY-#@rvdWE0iv{*X%$9v z${Y1plEwughU~?}ld$rcmFp_YX+gSi6G#$R{b)H?5iQxq_VndAS#1t9$jp>4p5^cz zxPVw%3F4LAun^2<_8qk7#itF9^BL2(d7U;UGuWt=>X!*?YCwrxawc z2oDBUa7$8$hWZ-ANQFdI`wlG1GC>)(uW3|^!xlTSsH^Q(qQ~u))dVx+ktJH1TRVAA zW*Arcbixpcvl$1oq@%O5sdMZ`!4t~UZbKhGkEnC}ju?H9fE-=1*1yfL%Nq%my~Iq) z6VG4*&tNN_m)c=|O39RKtb|1=83(mv@MN)ECpFZskC!IUvsT+d{6!`O^^sVC0Pene z3ETm=>hw$ZxJ}3BtF&Xv#a@&UJ;oPbhwXS&7>n$|6=6+BIlOP3-Kq&qp}W{?SQ7D}0?6q*)15)Q6@yxIVs`3x8~Mk_1-1 zgkx#9s*dXA zxs}5s(f~5zdYyq?>T-a6Wl`2}skvQSS$NfdwD(coU~#Z7bA162V-E7B-BtA&`*@9I zUfs+(J@1JMYgSji!%lyj`|kyWzS|0Gh?|%=vge`@F+)>H1ts`nWX0nUu6GVY_$5FB zepa7INGFrC+nbPN)XXocJY2x8B1k#>la_i(^Q>c*J@1l8mMU%0h+N8Wc4+MN zRi}hpO6V{FCJ|lV11{siA?wU^Dr>l(Z7{N{W(x@_3z+H6z)400Eb+6dD0fzO`DR zg(;0jP5NT@np0l%+}jg>Hjh2G$rIN#hvE9XkS{Lv)f%H!N6Cr%Khsb9^c(u;;d4B7 zE*USKHl#m208*~@beJRph^+@@4l?I>9(n8mr=L%^qxEDhNx>1rJ>(Y>N* z6x2rwf*FrC9_-3BiP*0G(3^HI6U^atJMLVNec5jpjQc|K<<)g<_l;}#Vi(Ty_?Oqz ze0-z)V;gm7y}xQ$k}J468tNBLP4Kvtr}?nP1uUEy{aV!-V#P(9{BK_y7u8a%eAg#v z&3?R+0?Bp%(uYJpUYP<#>@x6)E?yR!)0tM?W6EQE#*(lWMWUYh_~RxAo@uZ=qKoZ$ zJ9#CdM8nj0`jXHYu-fMipE24-5T4jZ(1vhs6ov2tHp3Q~f85}(VA7RSP%QUVH~~wZ zsHvaZRJT93FDR>5C(7gl0Yp`!FmD{}l73cxw+ux-ieb}@6<9g>z5M!jsxkK- z1}Nyqg=BIg6qC0JL6pWt&mC9~($|-7w9;v*+z;36tDj)5+s3(n6CB6^)6rPI%Ve3% z{BB^$b@(GxE6>$F_ZiOg2FRnb%}`RZZ_CkriuAk9J#r(a5UOswJ$nw$oRrgeyTkG7Ia%|VhY zCJ|8~=fhjD|9>;%RZFkmN?ncXYEW^w4m9QGA5M zA!U28ZpnOqbXLQmmK{E02zI#OBn5z0)e)m!=4+5cLqnf)_An-Uyllyac`?Eh75`8o z$Nv|0PIEE^j=3xY`+4{g&Y03A&ThP!_JjRh5k9|7h6s5H>H8jXLuhqZQig)cR}2RN z4mx_544!-{_46Ep2U4&F?BwWbd;S zw|aHI?t9zX)rJO2GtjMHNPmeH4T?V~?>%sdrvl`dL3Ir41md5g8qR8r3}U3gH3+Hi zznUX$GMYHV9Wj#S+pFR^{dxt;lrrr|qL1I4uGDF`(W}o_$GQnFK&Sv;NUzP5mvIeg zsH%P%$_B@{(1!9~XV>j&vUWt##T2Hptqpc4zo2~4KIsDjAX5l{*ADmT)T=AGWSSb} zDCI>HaM~(1TzDkrpE^bvrJ+qdg2XW&(BU#MH3Y?=I|q;>F$2}(Qa2_pCZ;K^e(zG?~5xV^~9l*SV; zYYnzS8PQhCbvg+FWq{7&;}|n=tWVF_V_CozFY2Un<_C6vBBmSq>yy2GYkRudSn7Us z(#ojKZyvlhdBD5i+h$Z?XUz{lK3mqGRsh_=8oQkuNWIXkUOqP}3R~<^%Gm64DKi!a z?}QStceE!G{16)0=Myg=EjS18pi-c41$Hfvr?q)dCrc+qr|vw&3UdXZ=!llJ(Vimh zgT%UNhCNV~G#)=C|4pfUZ*t?DgJ6y?BD(m2iuR~dh9>DU;CN}hj+eSmJq2w=<<>XH z?QB)1&qub*hO!wXvB8%23xY{hKw(HjdAZTswf~ddWNA0{4dO1~YuARpc=0%j61h~Z z@cChYc5qO{7iwGxTk&r^OCH2KfnHjm#)Q$8pn-288EE=IiDG7x!EcC!=gY^`~!9y13Ob!qz@qfb7GtWkocY+bEwoYeMP zSEWQf?YuncddgqRt(iJ2n#OlQ2xE~88aaQ2)xusosM{vV`A^V~w-m)3fY9@fErz5( z>T?yBB(TzDPn!8_S9?c?OU(JJZT2FPO8e6uE-tOy#LSgJ-nM!|C9_=7otZ8e^KQ?$ zt&7#N>o}~i@VGJnZya2?suBRzO<1~s2vGGPR*!eSK9-PV=T?a*UK?{s6QohHGzZ`$ z<<*XSf#Jmt(MWd`wJg8TGx-Ouk)W4qe)ELooKPF`zA(^a=!^3-@XqK z_FSS1o3Zxk>tCQ7l5@cA>a>&w*1)(MnS`H10D_S$hA>N8hM^||g~%~sDCD__O_rt} zKTX&u^AIG0bJJkH&|etuXVz14PpnU@d4|$ti0LpIRK22I$mE(1lq|}Zwx?o`7Oq=@ zzsf5c@cv%Z`#Y$X#7n>!Q!Fq=Q9rq??$PKWl0?hXoBp%%N-njo1TIs^-YmpUb>$cq z3)}3%ig5u)tJObrk2s$H#y;ruwCXw|L8VZ*(_bNL@$x-JM9&^6KZ6k(Gr;-UdJ%4tYk-bbHzQz74 zb0FF`3pB~l^SyXXSUBQ#-mO;cOe4IHm&{48TI>Z^eY$C?|2Tma1vY`tP9`-qEkiZ@ z6{uLiqwD#6Cl*)@ZI)nyuY_~xDiqkPun)h&Y!N7NygB|DUxNEqwYOoCuptuW7;ab; z)(<7dlbNfbBEk7(U|Xfs&CzGfp;Ik4UGIH^APB6;`*c{uc~6Z-6xb*_mE#W{)JptS#GJOmmzS66&U^Z;=-%un zKb4($CksA_UIjLXlZ!6p?g`bq9z%KFu+>^dnk1FIGWE1LSKD9eiZMRX*Z&pTt6ZS1 z0h6dhW4%w>|F-2E=T++i>Hy2Iyy#2BwtH5N<_1OWms})u!Eqf^Nafh`xk+@<_7saS zKS#S9mIIR0^)#{fK30q6iQQEOc zWOr_b*pRAvUSj<+7DuYjaHG@O?Egg0mv?eNDq>1ww{X(zl zSdDwwVPyOYUOJ?F*U6BK39Bg3Sl5lIAKX?1GDVkyvoe`!@%< z5E`w{&%D)<5V#D;6^;%)6CbJ@#W(v5N$g0Kcz)~4)JqoR^KJx(D+sO~WY{ZBhAHlv# zU|)JuoHTV*St@K{YyO!Z46j^tCNpD&qr-5Tfs-8b*%vW$$EWI6$8HUj0(hy5R*ekd zZ$);0!8r`r<=GaqC$qw5OGBZ^?r1S8v7M_h_*MX=PthS`hByea&?Zai`dY@5AD1`X z+Gia^`*A4m#wHDBxJ9@~iSXfe5uJ72eyY$?RtA6xyOZl?RMhU|(#Lz;ND{~E0HCV^ z;Nyy*n+hM1D_s2F*Idqe=AA3*$)ibNeVvk&Fz8 zRR?@VNHPy4_<;uN<&v3p`8N8wV-$!okZP6B?l|Fam7T3)10Psac&=5Vt~0sXk1%lg zenT*)Y^Ap=<>$ghXOmTm->tRTJCqZ+;jZ@X>lM|(NP=Syg#h3 zb%M}Fb#hOywaPdFyKMo`ub9F1S~GL_=d!PCW^b5|$FzRWxni2=;!nh#NSLz&HOme{ z&00rk5!r<28kO5WgnWJXC43RLBLP68v)`c2=t{P+v&2(&olgn;X3Q%og4fN_e4+@) zlTI?aXkdj_DuY&3>AB9x(*<(Rn=;901`SIk=Jgv7svNm(*Ch%!xy}SRC2+2Ch7a6Q z5BZso@8PN3LRrA{Ia<_iu?=s&Sh$bv4cJNYeW{V*%FI=%=|M6YX;I%Zv=}H)Pu>wC z2+NLqxy0)nl@l3q3?)@Sn)IwI^cBkt#$SR@=fHMvN-xrVKsdn|8Yg_dOA^BANm+&+ zsUlC;WENs2y1ogCE{#fiwts7Cvro=N$=r_T*C}?8EkE7{e^AxqII>#D?*qrnKiX{0 z87w{Pv)u1H)*5kkm?MCkf6xAUG>y)`!Au!|6F#b*-KqP2@#CJA=e0xRaNUI#uRJEKo9786pSNW4qCx-60A45+8Gg#9BCPW&1)9#KWKvE^RIr2#xk3GZ+9kg z#{>~fX~4kF3yI`1KYONpj)rIYJ|pgGN8I%M&|pN-Scy(4*wm(RTLdYqcsL8qiUvydfC0ePA*qya&&2%+l*VYNajPxwV31fMBb#*n2F5|QED#Jc%^q!Ye&PCpZ(xL zPf~QNEZb09uV7B?b~Yv9jK?%7#uG69^kV*s_t0T!uzc|1{HBW49NLongZylSnp1;jRlQ24rX**~A;KfU=RAobI zdT|`TFA=zmYp|D(`K#uAm-1v5?w?Q%!0F(M|4|{qQy&E`T-fi}C_Grhf`9^*^>PHs zTe~_UO)$Y9CUaGoX{CoFxyp07;T(qGchCtKL_?E)jG|qnWXT)lyzz1Q{+&T`@?<{m zxl)m`nyaH#jINaaLm&YfsL^7d*2=JjUz^^lDd(lmfRQ4cYzWuIbNeCw9{eomT3{d8 zv86lp9+i~E?2^LYKX_Ovd!ml$Mor}6Xg#$rxHzn+$l7OPYq%QWWQPHWcCt*ZN<#XA zG#6l{j~4UUf|yUAhUo^}ZazpX1XIwmsGop&tUfL>zmhk@=djP;HBBAPup{-Ky;e>cAEi}`q|Vtm{CR1LJo9! z`*h-_BU2PRXK@5`AR5e=-ipZYM1|Wk0%Hyv?cop*GR@E!e!UgtefSvIMdNWih5Jhi z|HW$NTHy!Cs56xNYu!Y1KslGejO|t5i6TIcyW?JxDjiI1zP`;AHb!oq^Ix6U0$Idc z&gn$jAj7=OC-^v2MGa?3M0q}JgZ#J~`&`3gz1dYy@bI2-UR@b>Hi!F%*lm9d2#8kR zs##wy&5&$I1TP3FtEEP1=0aOf~hR_sKCFYxtmVFjb7pKRa~Uy3s^g(JA> z(&*TP1rfjavDDturQG%1T6wi0WJ|q6%fYOMU}fr)s$6G-)r!}r1A;Uvsp!T^tI?k0 zT=k;Lt^8RdG=81fVNM9pVECpYZ>m(Di@)tdhUg_ww{_ zhx~#F@(&I<YA&)59ktsILYEn&vFplLS-OJ@p#+xJ< z5GU~-qe>Lqtf|DFw-E5|?}-LGG=`nl;5PT&P(B9?#QpGy0JM)f(SLOax`1M9=#0;$?}t-<4yAJhu=W@Y0U;JIO*7&s@zMpTo5o(iZCV^Y@apg(nx_SGu<}Mb63+HEJjz{IO;3NoGVJ8x zdHxP5H;t{KU1icqz_Ku7yW2G4Y6Z$tm)MPV;$cZ1Be&yLlwKPDfYvSORW*64oyhB1 zLcrtRJ-pOf`}|oVcOCt=Z!Vh0gC7Cjq{nzsA!b1Q>rn=R@i_gw6N*ljNuul574-W| zi)fcD43fAJ^Sd2a%P?i~W}{WcqS0*6=2&8c_V&hHO;$iaBg`t2l$PrxB6Cd_ zD2$B*K0s;ckp&$(cLr6&r2uL>wh!~VGbV|?A4MU@J zcXzkM(DCi@sN;91b5|$R7H^=v<1B0Wj^0*OVloI}FxYkEHfEjlZL6$|2)?R`GibeLo0yLA63OYs0@ zRmEX$e^hlnIrDN0$S{7y1rSXZbM1Th;GKt~EY_^*Vcjkw{Ja)N>om>mmcCR=D{qe- zAk4*bhAc(i6xA=gPjLzt*B=JFxc0sb3prON@t3$#w*8|a(2SzKmmGi7ii908% z^`}s``o)XXk^!nR*(cVcFPIMlvP4_PhjwRsh?uRb^btkC{$g%8H=qL&Iy9B~re7KW z@569VSWH*Tu!HVg@{XZsOE1?D^~Zos;B(q$(ORM?rA2D4IP;83>Uh+6uGG5F0%uXz z;cA{Kz0FNcmON}iHmR*;zIb0u?Fed#@ffH_LcD5M05G88BtYbu1)5ZBDN zn6~4sKH-1+A*k}`#6{K#Gf_A_ew)UQf*VnVP5Lt(H*Cuy}V$+?yRn< z8fBzT5n4VqblkB!atZXaw>DtxPEinhc?i;f60C|%5Qc%$J*mu&GK^|j4R$ebA6C5y zeRHIt7fxa6$T|iOMX8%C&~EI94X+puPYUBe$6~l!8iO)38#}Tb_)-v^Nt2%V-Gvh= zoXskQ5QhS`XHk|)k(n2_anOWg6g=V90?dtGo2-cw@|-&&2q=%=A2t^~FuX2bM~EmwYnVaQ*1`dG}w25SvCTLTqf)=K~PK2j=Tt2Ke_5 z9-hB+p;rs89IIb5R9Mz86`;5huFxx*TSwkYbXBR%~g_w?B5^fletF=oq@0Y)Bkvg!625P_GMb7jq=ykXrL z`7NN`Ay6UbF8e+f09$Zw!_JFS;`+-5IB3f>?1n<#)~hp#@e~-!sRZ01>E8cRmRuBc zSHl<*E<35(JZtPD4;5SDangBBtABTP-g^Dve-bcJ@y3_DXQxBAZg6e^*S`*#!A#$%^IU=37H{P8lV3=8Zfy zqYQy`3V!aMJ`FQ`$NnhU!wV@jSbt1T`PgTN6k>li0i-pHw3a=bmK8}>S1B7KEt*Z| zlueX3%6UMgJFrRHmXl-VsrRXkQy$tvjTlm!_4cje)(1K^7 zCIy%YAo(XjvX;0f54Rz>9a2sUr6bo2{_MRPIjTGS#=FN+?qUH7MQecqUAS)E_&OB> zwHSM>oRDk&5n;VOayt7(wIQaKRcUGS;3?jj&jTqup~mT3qIL*v9L4RXbH-;9n5$H@ zywMzNo(edyhUbd@5h4YI7ws7n5Cp>&?+hMV$V6RsM_!BcgPo1U>-4+tY67r(UToA`;mMafgz8r5pcVCK_IbC^sG6TVp2iWAA z5c!L)($;51SFgZk&5@MFrflzBuO45T}g@-A9A8Qu?F5;bjsgQF7OmY zSD3Yj_{4fo4AlA_9Dm@UCcL?OhW4&5ySbHjKLD`gFp{sH7t@_8s2xDV;5$@7VnWz8 ziyy({A26U_F3wN0iXF?=UPg>(Aq__y=5X{v{%jbJ%e&Wa-oQG~sJxU>x0~OklI=~_ zgZAAZzCvbEt{(ePG0V&h=Q9=_11Tf#K3`GXmNCxqOnre-Prbisub8b;VLt`!L1$`x z!1|Q-CDF#7)rBo{UGe&g5w%pNMn&g47I=5NXIE_&$1cCjKHHBY70FAkfp2uMnBnpw zEp`>?UCIX9ubtLsFIdlg@bdESSuUQ>=Pc;;lUi2c+6=z~?70`OTV@tJw@2()+Vxp$ z!}s})hNwuKxgpVI?dx`SbCvdb!Fhbn>(C~)o>HqtoRntnWQ9Ecl7oX|l#^CZF4h5} zyx0_KarM?Mav>}{Hd>qMHtlT^?z^Yocy-s(Uff>akR+OsF^R_4q=7g!9+^OpY1z7B z3s~CK?{8QazPX07yRMz#=AqUlNVs39?NhJ{5rI1-S#MF%-JNa*%6~Ri*!O* zxNZA744uQ8HLOW35rIQRr%+ngd0NE;9UrI{N_S<^Z*2^u`{-ZJ&wIWhfQvFXVgIJZ z@D;^$5ve#VdDnwopyWFC)pt(_deJ`p6wJ7oerne%>lh<0qBi}n$p0r3=h@}nsVFW} z8)B`<6f8wNMvu2xtjQj8meQDqkbwUU(P>wr`CozTtC+5|t{HXozy9YFJDuN0P;6&ERn_+uoFCa!%#(H* z3O@lU)?=m7NY7RMP)vo1L-dh^?FF>2@PRkvlOz88Xv(#hV8Bs>5;htR4 zPaOxBm%SwbyDhp#F&I1@mPN_+{#Uh9=N=HK_)8vHs(;5T{@?#df7IRy{krBdnhoJS zvP9s%5%Y=>_J*&o44R7g5SG2e{zL@Gx$Y(KE`ar0EV?a25+d%jq>|B0^9#|;uAcn? zd_sXrGC>M{wkC=<;y)ehNVr@N9GU<^&iarK^TRJD-Tz0TeB*5*PXWX0;N7%O;ujG~ zc4TsJT^h=fuKHY+=Gn9oE-4mF&0sV9M1_E7pu%wW8-C>mSm($(mE19*xI*ej6x8IJ ze5#C87&kIRo_Q<9;Y9e;j6abG@Lul_dhQj4!4%MZ(h5?weH4-PCEc0Q8C z?{JB~~N}dXx zU@bqDn=h>*`?or4Ms<3Y zJgyp@dEB+t|K;Pvyg&+6h{Mg3IPDjbmikmX5>bO&fr!gPU@Ki!UrWYQyFQ;b*V)kfXtcb z__TZr;Dm{Tg2EA_Q}a25RU%QBp#xbC84^3s-{W4qD+ez;Gt)r7k3;NN_m)~RcEH}s zK{1=)B!3<~Gp(7Ek^Gd{ys!*yf~?nXPSiCQm4c{{-Spub*}j@ZO|itL>U{^Qpy6Km zX$7=13Aky`ByjIc+s<_WSzxkgb>(euQ&7l*goCG$r?Tq{;AVKO@m{?- z7wtc8FJWn!tBT|6_iR#~J=?=m<6~i=5`cW&s}2?t5)!rjf{XbX3ZQK{Ek?cR_0{5J zSI45|%%|!;_1b+WXtRAj%-d)Px+1(WdAOU~1^5U}*d@ce9<}+PU?955K#OpKlP7hm?N7T|Q_yPW5 zR@fw|c0`N>mjDreRFXKc`o_E5B&b!trQ0YSc=$@X7jWk{DqkKR26e57@8ymIhvtQ5f^ap_6$mobmv3Lt zcfJP}DG!%nsN4xxwsEU}{yQb{j6bS=)5Xif-;d)XI1WbpQ117GAh=cJ?f>qlI(eq# z2`UN7s7!+qC~CK7z5Lz|$SWm28$Z=C;h}e)cP!%+&Mc!UR8KOhqx)IJ=vs+~xXz*U ztpvd$^Nym2w`NUtK}axLXZNXEW%bfs*<8ONhwaFYWX;n#^_ni@6O_C5?aaFy8O8`K z?*;j8jfP+vaz?&Ud8`|CtM^8@ca?{Y`U%KIEJfL2MkXn;LMtf_!65Ido^U2pQ8No< zqt#vFJG!mt4QDsJfP>{9Rsi|vNs@ZU;{u-fP1C4@t^FIj_G{)uN3zBZnR?9&>k~PW zIt`o3*`~*2z#^n4r|9S?`7)5C7fJT%6Z7^WWwu@8Xb-oYmtPC;b+#EvNXpQfMyYnJ zEh}&;ApyHH0eL=}OP6=GX1Mzi+DVTd3olv)QL8^9;M-Ml{PpBw_`@3CgNl&^U&!;c#r@NpX1z9pqD9OJn`;7- zym2+s9kmZJOBi!z8G}HY+q{i>=~G?1jKgpQV-e#*DdVnnUIdTD2tnBBxy0_$St*0{ zsW^r3Lw=0@9^)P5{N=v1&w}s_rL3xT`$oL!ic0eAv2a~Z{s*}(jQhHKJHk#^C7v0{ z812}vG1nWtj(&rGc81M*LZ{zm>=n1nsya-r_)U|>S8tS`pr0PsyG=SpQC;H6328N* zvk533-7|`>cMsqg8=Q{~YaF7>ZDA4IW=}^o4)INPCvwcUi3%Elw_SH8!b73@4pKi| zX|1YbChx|(BaS@ciE?-2L7Vzq#@V?XO_CDEBjJ0kc8eWGF-PjD3VGR<_IGqg@p$%e z+Fha%aH>4*ZFS|$=dg%-QKFD;SrwU)lJ%W9?9EZ8mhD->-)Yt_JpoPIP&0oo{)hh% zaLgK8>D2ZAu2@Oxj;do}h*ATL_oWu;lQUg+$DO+SCKB}4fqNj$nPseP3RXKke}~}d z@=PUO&P+q=Eq3vGf1w18RtCMCk(6gOOLiZK2A1xCC@$^tV2Po8Z;`ry2*hfE-8LRK zEX!B1*07Yw`>SAH11=FlzHL6G=mQ~meYLc+MptPeuD)}6wY*@^^J#6Fa~X4?sqdLU z*wMj0N$CCE3q}!#q@?#M&TXGri1AE`kvHstRjHo#^v$Xw-`a;w4&<2xlzXwl=o4@g z*o=?lFpIJYq6~}sL85b<;4U*KX`)YLfzR>?l%SiA>;52}CWqyLjcm*_Ui`L{4-gJ< z)|9c~&Cm_xTUbEN0sHzM!;<-H%Y_j|c1ydAG~Ck4ZpDWhf89C^lGv=1mlzz z;s!Dc$DpjU<8{dX)>`-t`&Ed&;&q|RA8%7*(Vn6m7Q7O_J>q}|jeUztN9%j;ZL-qk zD;R-PVymTha$i4+vXBbVw9Y<1-!JLB;XK4+G1Jr(x+<~mQlHa1VhVfJj6mK>8@^ql z2qBA0kBr)w;{$e(SLnUZlln~@V>n2b7#Cus#Uze%bxp0^3mb=Oc=vjFf<|cDc*Oj* z+)%sWxX;=woCyvJ4FBF3fYw&E88_B zJ6`A!NNj{fyZdj2tU4=cpYl=g+}}P&?XgOJhx~~Sh@N^V=Sb9K{xX+NPXvY@36Yz@Kz!D9 zdeyp!MJI@>uwqEZNsso=d!m(Y7)=ij*F)jwkjMhQ5ahrm`C$7?LyXEfIvmNKf+Z9ClFlmRcxkvMglabVWVM5bY6T$r5KO?FSlo7jLfZzY8aG0>gZ zJUEqR!^iA{5bNza|9QDH>VcgYz}T3Kwkj+rrvgEHMy-VU7MsaOTn!hanuUFiLYGVX z>a~w$#7GL=um*UPk`k{o$hVzAnol=rwl=Fg9NqpnnHVa`d3j4B=T5v#mj?xl%j#Ofmn|Z<7nCG^B#(>#K^yJ zYI=9ut_UI{wwf3{FV&hCWIo-HEF+ugnB|yoUU0#<9UEspYo&UR=PK%*xae@a!Ggev z>i2clkazQW>?1keo7RV1pSBLWt>hG(44vOCroXD{nAk9td3v2VXXxF46n(Gjs#aWiP zELvc7b9!e@PK)+3_A=wV>Fcn~eBkW}UG5bWOo+nr%rod$wGn?Q^44;Z=S!|pPoc^H z-{3%Q8Xl33o$iQW&OqA6hPh(kpyVJtCPDEH&KwP;_>kUS8lIXmpz-z(T-3PXc4slv zpuFzFp_74%(1w`DPyzB&zaz}Wd=iKT&ZWXb>_u(4ZCd;joxC31=1=H0;L0qr3#R8`&nUR%GUBzF-crf z3x@Fg7Xcg@rN)RM0a)MO1K(8>m2E*x>{1A{W3tv~E<=zDgU7mCAO^tfL}`$zm!7mb zzPPctT?e8Way6HNhlOZ?D4|sw#$1uZU5&~vn6G*DX;MN2CKf2x@et$V zMD_zZ+j;$X{a%~`?xDTWV?U2o$LBTJ3%zO$Py~MJK&ef%L=&@V7%OI^c#+IA47VZVh&(q*F{|8>iLx1Lc)QoG_=t zu<)l^!)xXB+drk?6<@mdu*BoZ=9_}iVWA!2bi zXTmm1@TM=fymXYTx;;LVGn|miUJyQFO_PiBn<y5hcXPiS#m`Hfr1YHWf)a#pFwe|gWU_Ur-?W(#spiQG)U0$ zFsHgNi)`Vdc)azt@&w)3<$xF{%g{Z?m!=?g^gP#O2^wml)4Sp%n`n6+zpKC@cZ5; zlenHM{{+g5>}b(~ipl(?c#S zI_NHh278^#2VS| zyka#ybZ(SwxTYQ-wo(g^3_;Z}>hvTCC^;!u#)RjRKET~9ec^AWj6?5D>jLX4^MSi; zR^;q0!>DqY-AqHMT%y#?m1WCpaSWTprV#CXN)Q#xChhg8;GbuIP=fUl7y3Ru4nj|@ zam3AR)^*v!spVxumEHJBf;(bB}_0MHD7S;2mqbZ0g%N@MI1;vFhsDMNt#W5n5_1muACt!NH&_ zIGP^F2n(-!b2o^3RzD^KXJJcpn*fd4Nh!}zOV)d(O09GB438~63VGchHv+hygis0D zX?y!&N+l=pNiLGRL43zUTJSgy%=u@tIcd@FWGEXs-mv{RhQHzcq?c7({fs`r5ns;d zI28E{QK}8x&*Hhh|J^V7$`?!Z0(JnA!==!EP(xJXR7V1`J(QXL%vF>FY$DtO9XvS6 zcU8nhfab%|Qqe}X-QsBW;iy`+ZsX{d=Z@7)t%X8OfAv7^o)5S$ZDY+I+QqJY?pfK| z4_|erE0f!#BAG{Sxa_Ekv4y zIIM5u{XwB8(F82rZU;kUIDG1^9K?LHBSq!T#)>m=wNV*#CI(30vw*@w)smF)Qp0=U zn$qkJsswDwizC&kDz1#*Zv27PF0JxWH~SL>sQH`B12XFA)R60elOgXqr`yK0W9gic zb3-G2;B=-h>Q#f9D=ZMFB_oK31S5yVlph>^Tc-YMH6^f}sLnmG>QR-0HRD6Wwrh&l zXEjv5;BPWxzATY|jHgA?ewn^V29DZn?wlkjhfnbd_&Ln>UZBKDf7@gNj_mZ-`zZI@ z%KHY(oXVTyvwA5wK!-I`p&z^h zYWZ_Y)fQ)DlN6fI?a%vFRf`8xN`~e7vx-yY3}eOV)Irr}B%*I>k_=zDZuw+$EeC)a z1SJfn!uF_ggYB`IwKbO76jn<288Y^ZxC)F%$#Q1y-Wz6=Cem-MTNX#PenfSvk6*Uz-l=Ds5r}be>Rj;^Y!CA#>FWT`lIO7BeOJR zBmkg9%_X1joT~dcDXRuiS-38mmjFy9=J7Z^X|^EYOMK7MO%GRH>Z)vb?=KhSP>Jrn zI?A&$7!uaZIU5z_n0op>qJozy9=jwZE6`h@gW|C;SL@Jo2$|4?!qCN z0XQN;3X%<_(Ru07!N_vsT)Oe@42SYu`^3<^P@5kD^KZg5{G=miA|MG^Q_b{t=6c=* z+7>^$5Qsc=?VNR++JUj?U6l_I8*r%XcS6+yGabRx`=xOX`8U4&`93Q4p(S)d9BByQ zc^w&O8)bY5?MkY@r(igNefvFsk)7_q;T~8<+j@Mx=6y)pMGwJ5tP*;aWZtX8}~VcYM*rC*jjI1 zAkkf$ZEV`BC>xG{G4G$~v{Uy4pGWhJ!S-^u5N!~h_G;WMsbfGDRx7*}H%Ao@P3FI< z8u?lr0r*{02+!yyVDLnIX0mU5L$&r!GoPt@6~!pJHdysp7V{DblR%_69%Jb8H}jO6 zYUU{!wbivL;MAyp!^akiGrQ^^PJY{Tel(6VB02aIJWRe`1TH%7H`$_AOsPF1(V>KK zD4QTQZ6=elKBQ(d!Z57vIHG-v?z^=8J2wA{-|@WaImF$huHzXPUX|=xLW9{~ikQgL zllmgGrs)xId!3zN`8&(054cX1QG!5C!3t*1kCGW6cE#j6bscKISn8`Yyyq=k=u{kJ zo`EGk50|KfV|8$?y4=3S_68GEPl}knKik74yw9Ieu|ZM8MSsg+WRuB2soW=CfsBku zDYW3Lnw*-bsU>DYM4*e(!8xtDyFDSz-op#uQacnZ_+Nm z0(&adktChrCgm{YZG#s|%fM;5zL2|SO={RYpfO!~iLbL*q#A2FemB57Ug=}iRrBRA ztjPoTDRzx8V^(=&tdGxv(0BFNFrQ~2v#hJE95K}#pkjGqX(NGds(1A0-XJIUpzKP@ zljhv*Wj^X_i7jl)FZ#3VHjGt7tJf2z1)upHl&sjGj2WB3b_o5ccNBhe z0#~}FL3~&o#GrO$ z?78iQ8;%H_(A2$^$FI#ySwg}K7*QmJBAdsPmaoR zNb!t)aT&2ztHCWIW}}u?i5BhT1`uLRXsi_zzsEeCQ9_}%xAORwb#G_ zcY9jm-ZnCeVo~FNDjIj>8J`mDPJSj~b0@kdJph-ke7n%a{#akW`#w=gd+Y1sSm+32 zKCiuo=>-9rCz=IL@UUS*Q*QPB8!Zd3p0iW0Fp7`tbS&V|k-- zHCE%>N++JJPCAs~7nF8TBmXRmO%Es^h6=@xA))pi@=4iyY1`NCNY@v`Wjmx;AJ4`< z#2|_?rUW_AR@Q^YM9_hqT8-@03e93^OxcAlIGHcs*7MgD-o%!>3lKSv=Zk#D1A(cQ zrLDX^W%)Cbx5Skp!_N<d27h;EH2*~&%m%qv@6v! z_1v=?wzu0pl$JzB_I5k3P~M{!0jX%l2+VK!M?qh2X6p{}C06yd3>PocigT7I!UUD= zO4>gsRt4AMuq*IBfmAu~O)L7&E?DCl8nrJm;qzL<@mXV}6={-QY-U?*1AntcxvmO9jbxNTm0O;nf*fwgQ{g_V zBCLdYF}0B@)hlNoceERgiA*Na#pXR9`Jyc^KMUdwAzp;+!(P$%>@%lYgL}0{ZAOlj zdeR3k&##jBRvJ!6M;7bWr0%Y)i1bDZz!rNm#0#7s#z7<-IAd_C=#O5&4oZ$}_IG;* zH&=r(S5g?H*`2rt_>6j(?9g%UzlqOk(OovDm$$Tn0loRl*5xJ7?-s1+T|Y`WVL2h% zSB$!n6I_+^PHLp>?y_Y#RR=#Q_{`;(SQG`IWZZz|j<6zK$GCg0fvH;iBbT}3op%8X zD%H#)>8cz&@VgwoizbV&TydVOmDq}!!?EO7umo1tdJs|JQE@nDrsXUo(njKPUN10r z-jNakC4}Wq+;zyEwVUt2Me<#Y2qf%}V}Gp1RCDvq8S$tgyQYFPA${kbkiI6l*0rHs zi$p9%0C0(OGVbAULaOnRbCL!N@ThRe%g`bkLvv=kSR5e$x~l8;xP6$BVbV=rC}sn% zWq6&Vz4f5-;dpn0Wr(E9t!M{3yOvw)+gmQ6<_dF1xgy@So&!%l@7BrZ6h-_pZ9>Q) zRT+UAReuhP8j9#k*ZyWJf9#7ZlrHA993G>ZLN|>d1;u9AYmdWVbEZP`zj(T%Dyd{x z*Hu5Z$h{%aS330(qUS_*cd)hXc|ixn27MEZYrkk9LatMU&~+ixnb-X0S030cC5Tl#uF7!gJn2io*Ep;<3sK1I^ps=8_?MbAPYX#; z!p>yQ_@Te^Pfvc?C@{&=erq?{tu!PEV0rYf`hJu;{HHN~PJQPs%4Jw|MdGgpMkv1_ zIDm6vyE9Ij3V%Gx_rTvPP+1(R^4jW`Ga&?fPk8hG?_8ZfeU-nOUL>eQ_0Q~{Ff??F zq7+a~d^u**ukZcxQ$Oy)e~)2s=pbM4$7{! zqtTlb>0tOT$eDkxMDpVTeBUqr7htEW6?_iSza#q^jCI{Cyh!AaR)%H?B4<(3zEX1P zDtfT1Qh@rZ`ptFzZs58m#Dq5|ppWtk3WDC$+=spme@JX1 zC|sfFrxp%0r5}Kft)#DXQ3x2|nLD76{?AX3P@RIKSBWkDC9igcqS`}J z4jZSV+e4DpfF0A4sKG%G|7TD9{$NF7SFNScw|R^ajCC+A#q^^S*KhC6FAPp(9(+-D z?^*>CeQuAyrKA-jX8V^^apKh9s&IfAmDwtW{3ycpb1VO8t(TvGqkMGD{+G#S0^0At z{#)e4-_m|GN`iM5c-H?S!Sle`#+RLrKy&ZkbFZJDzKsIkoa11^`uS6&9w@Q$i9M?E z-?la{M@ev-0xq8}>`wIvgRp}-HqMPA;xh5Q&69=a zr09Os0`RAWz;7H=H5{U#KB3V%opQhia4eIbCm$fChm=!4imu7T>~iyEjS80LHN>=H~S-Aff4C+PEIpL$T?i&A5R z8ia$S^MvduJ~DaM6lwOSOt7O5l^fh(qf{q$YrWP0ZAe z{OjwY1)%-rqNB9bNZ>`PUo>?Z8)&r1tJl8N*eOY}F=u%;LF_ssEhx~x-e;1gQy-?B zD1DPmnOQ#@p;Hc5Wi=eKJP)k3;@0815E}OqW~)@4rjW|gcW8^_KR{gnd0&4c+1{b3 z(1Qp9It%2x=qxmDEb~}Cfb}?aQ#kS4D{l`9b2x03lr9(LOV>X{8V)Xk9y{Hz({j5T z(}B%{`3~|)v~q~oUIZPr1XLGBcV#Sb=}PA;tR2$>L75x5qa*#`yGM?pJn5h!a0ftH z{u)^1A3`yaVL$ReyqV}-e}~HSNFW4revc(;4AG=1jTaAA(38Ta=g84#StKJP>#(U) z;&9<@OPvLdc=agehFm1Tdv#~begKHD2EBb3xNO%~Gr8%p%DAna-5e^ho>b)~`(ED9 z6!-4c<*F3e$`SxrH;)u%E}QA$U}`Y!6IJ4LjoK06U#MZ_l!6TH%HxT1=!yO5(>pSm zSf$DbkyPDQd0*zw@n%~9{EZaP*@3|5f!yb`zPzff+CAa<2lSo#}qoL3Tnxh)*8R{v{_igc{_RIOA5TL3w=)g-ad z4UoTMv)-(avIUQQZbZzX^_SY(F?(Knm(!{JJ3R{ho;2-F8-S#&@6 z`5$lZTd=*VeJL)gbf%0?Uu>5h&(I}0t##ANiROxEJ>+wTa>>J)xn6cDX z^!U_ickX<}@@U0@Xn#H|dd^hcdT_iG%?*M*fcH}N|f z!!45RZi$in+;{%FDCTn2XBfS_KYMb0bVI~U0M^!a&#ugZB)rHv4WyM^FmAGcCgTS? zrwg${BrOni6stUTSlU#)Q3tTeP4)l|rn~TZ+@DW_ILCQ+9e1AfIaIaCu{(9PG5WEd z6rsWDA_%gz0O;?>fo(Zg`b|)Q5nH6*xhCRAzc8R4Kr>pBwLa+U{bx`7u~_kiG0+|&4rj#D zuua<=2Bmw^AaxUBy)vY|Juh6W&>XMnj<6o}QFG88no{1fM_3vpG%Vu^l^+-Fnw~Es zcwy=Q)8;BVE;HOha8W0)Pc;NS1dd-C2hgM_^Y|#Imh<$}-VBXk6DLpcXV3il(^G_b zwnUt0K2^zj`Va=T2Bl_hr)g|E+X-O6l+fogAq~{<17&lHe{1fOl3~c|$Sm}&1u#GB z+o4u#45FuexNjAlYtU)ywJ#2VhJB(H7s|^u?w268OHQx73@!~wpIC1#FXqlyp#!Zt zpazqkqM!k2H)Z(&LSTkbRVTy22()^G2xHG{D;BP^z2W^iDc2ByL3>|>@9?h$mnfOv zZFgA$djQ>M-uB*}ntIoHWoQ+#Xg%MZ;Pl!L%KFt;Qf2zkH>eKQsh`TQ5k+aFX9us3 z>xfY|_H`}=uu^Q2F7d4uKGJB%y#zIEG2iy_)L_osKv0BY5#o!|)0mH&h0J%pGWj~I zHNyvnA@PSfU@!JIi3gR?8tU6(z^4jtnVK2(KIf`A z{Uh_)CK`Mey~61-`{;nA77^6dA`YZC>QPzweB$SnT>DVlTuS8dQ=R=Y@;-6u9gSbC zyX?F3%|+(D|!ReE9UI6Z-KU|5cRrCg>=I zp2XNbxW$gg#7qx=F8mMAf*)+p#J2?f@2ez5Ipdy;WuE`T0e)HlU&+81c`aE&uzvE- zSF5OO)KZ5kFZ34v;gcy?PzkK+|u#{VeC2Y5o} z%rxQXdB0;s4f>_r`WJ-iABxeP0F(<3q0)r5kxAPz94V<2j6~;Gm7k&S?@t@@_fvIz zQh~?22rNvz=!-M2e{TC23yS1s6z*O5xfg#%F<&)NbmC|cLhzGLPMmsLq39+%^o#BN zhl2v&ZZb7)-X#_E#>=9dVAqdMyxc|a=N~#Lc4qUu~3=XZYFHOXfN2GovAZ$gf?WDk_4c2M3LuU6VKB(t)bWb`&0b! zsp}3f@p1Wl2+=V2zY8*;7xJ4YSDiaR*XtjaWk7el|C{K?Z$|iaSN}WU;;I7zo3!{t zj~Gt=`r?nJ{N{gu6A5zd16=ccTEyx~fp#ykGQu&Cr!vw|jZwD|6f_29YS~sn3yBva zcK`N%90LlOCfRu)^3n^}-9QR1!r^_MleXeW&!2*7H zMZ_K)wW!Ob5OiF?@Ci}Zt;(Z{3pfj0Iwj;Xz`sU^x0K- z0*SQS*8H|wKCz5j8WaP*_JN0d1#}~)!S72l`hvarFLjky(&s2ZIx1+l@>=ympwU8k zW~NM$^C3c-l;!-7J61vO2efq20ID>U`F*O(1*GwqUD6zWjG9ZG_uKLN*2yosI;K-5 z=qaYT4^Q9*(IyGX-35z%b+rHR*|V6bmdFbie~k818vReA`+Cw<=L)prB#chc4I)DO z1ydyS)DK2<*~0aeC+cN&q%DAb9?i5-0qw}DCi}uo$q`=NSQ;F+$KN^SWh$gxO}?I2G%T^)xlNd%N&zD^6(2;`;_{-wIvFV z4=fH>XhmR7f+^6s)BOS(<36e_<`W#wAHNuVUePKRp;Xn%8vO$FLr zz_zWQsMeXQP%nGen)>4ae#61Oy>g&Fwi%kktTlAK=*Cro`Mmpvef(jpcK17$X>{F3 z&n6^O1j-UWN&ibYxk739b!kS6R%rk-9*2-ph?<&B+LdJtH~df=(OPB4qJ_4a^-grZW}-3Mw(4*G7gCRdJCqS1%LFA1v%)>Ny$=?CAMBX>_709u=oxzwFOP9Mg1g~U_RtX3o)C(>>z9>{qjXkCFhNR@v2$ndSp`M_ zAGs)iYsJ5;bOYI>Uko`aD=V)O>_s~sO*1QRSlBAfS8o|7$a;KzFN@E7<`KaTx6P3o zM_GE6TeV7*m)&)UMcwB7X?daeuP@WfGK#mBg6|uazH~kCm&CA%%b9aYA>lF; zJk^aCK4-Vs7KSVlVyDI>d1IfR<+Q?h<}B@JmKg%9lv&bp#Mwi_NBOD<%jy^Adv>=R z)Ol9#bhUO>e+UMJY$9Z3h}fO>Hb>qF&vmwb0d3o-fW0W2%wML`LQ{2e?tjXK|M=jm zk?UvBRCjKTrHOUudtw}MuLa2J#pf_V%%?ofdvCeClUi$C8p}r)pCY|Hl=E0BirXlE zw&|m!37x{D6+~S2o`iV%x&0j@&rytoWUA>5UxJPA0XNsab0F)5&2X()TI`aTfuxO0 z9oFUQ49g*(nJDccAF7G6nq)zx)1YmJIUH)Hb@vHI!grCZow!yrG{1gS8(ohWWgDOj zOxtC(Te@O!AmZd}dOa&2BPFvN1$B2hDiW%afAG0z3b>CBPw#Qi$QJqy51ZEAEP6@S zrf$PdY^6|ty z9mtAvBMPKU*ySTxKR|Y3PwYju6r3}`mv-5!UwV7z`E+Z=(L=tdKtGdVjvWo2R@mqTR{3fYVwT*PkG{O7>mWBV1@0 zvxN@|<0}#8JiM=q^eunHnedb~r5>ac*uUU&nn{%iWm0U3 zd{+dYwh$$H-MTRzu5FnIYNlu)3C?71rETzIEAKb1emt8n(Rh!owG_ZuxxJ$v@H80p zy24D>tjeYKO>wt8y>Gc;JA!1&2T5NU zEh;z3WahZba4NlTwuxg2UwvlblkkUxKsq*hio8ZTk1y^%&?EN6U68XjdaVv}s3I0c z6(Cho40g4^PBnGI0)TR^LqWMWvHYce;MkR0ul*~*t|^jLv;*^u?b2)po#<8Qfb-b< zlKhF+sr>{;u84sG{f`XZ8{*Z2%p)C4_vZXxzfmuqet>QoR4+hkJYWgf`OMTy8fOP= z5Re(+w0TxXM2+BJUQ=ml;5OqZ-VYns=)w&nR;Y|xr>#pUf2oD^HPT(o zU*4kh!XMGV0`0sqT)wW&u73`*(o<7BG#Ff`q4DD@pL;L92+@UGEwYCh?-k}GdNEbv zsCmr7Iky^2o&D(9eGaFt`HFEMFBz0h)d#4U&(xU=Xx!a@8kBL+lWX;IuHAD%JXDsF zLTw;7dwSJnnl7&yt~GsBYcF^vLGa;c`3i$Khqdmg>=&o)|0C@zprY)$zJD8t2^Jy^ z64EFLNQ+1f>CibUB{|ZPBO;=7cM3x{!qA{dmvnPfbM3v)-v9mE6ZGM1YRVs?LyNu!cHkj>Jg~(Es<57!3|KTSVmXFKnusJSlMbBo zR7rdSTDukp+JfnuI#w!f3`QlN5U#Or*daW>&Juri}_`z3eYTW!bsb z`1AHv3DqlBq*RcrNU9~BJuPz+Oh}ncFIlu4#A7&IR?FKaCad!&lCmM=$W9O!QV*k6 zlh_X@BON}^IoOZs>j@3*EjD_i%XbbNZByzWVEfB-vc~jf4=P3E4@asL+$X2&hw_r;_`)vW5GMv6sp29iebh?zABjME;H-xI1cH2Am zZE&(r`f-xUMe)$dnwpA`66zHb)r)OUo_v84^7J+jU5dha0?aQW~69hmAN_>{% z#-fucVwJsBrUqNlg%#AKh-K}Jsy_3egn)FEF-c)W zvLy*s(s`8_Xyz6%VPu7oCM0sdhbHyO@3#Yec@IP+OF`n0GJRN;6zIcnY@njFA@miN zzB!@odJ;J{_^oE3Ij@n_6Ds&7RYCwHj4@H$Ipy$XJ`zZe9B>nI^dvO}5i1AYG4NB} zxJLt%L?P+~QE_c!1gp)k*wUE1$0uB5-+pUY>YzcC;)i_tYpwR@hK}hk823||E7#H) z$UkKO@gEryB1Xy_G&y28F?diRKB0{%G_B00VSJ#B&W6_j#X`=lxnA3tJI)!?WtY%J zOxpu4G+8f(Xa=&%JfULy7+=@5mdnL@gt?7GG^roiqnFkN+3#4j$_ug2)Nnc~XDRp^ zL-RWiwht=`;W+W=TlS^ngZYC1prC77oQIR?w*B(VcG6C|Ej?Do)4621uws31HkTW_ zK+5xwtmaG@gP#r6%F_k6> zvr$K6=eScl+b_>Q!glp>`mAOEpocJjI5kOC8?G3gSM$Ti)@N*w(&iTLnZ1IW_}BB^_>1dzAxVKP;UKD3BdD>1UTBMXk z4%(Gj3%FZOLhWjWm;p!MPE7zEHEgu><`A0krQ@t#no9M_Qv|Y;;wc?w_{KtbAZe}w zHFk+eEjKegTRW8ZGa1_XXB)e5uLH?ibmmJ@=S%Do;YepvC=pplG&Oan^{SZFhP=SDwKoe10!Ht%?4tVL+lLngUc+_0rw~=uNU+*1 zbI~Y11bcg!qy;c#zTW|s3$Xx-mi>sb=oF%0h3H%NZ4A4veI>_C(gF8YFHDStT#F>^4ZKmLcmIsaF~$6)iAp6Vr9^|> zwlGdQ&s_(8OZ`7RH$(GY9Dbb1ES`-)Kh&l(37`z&qxFvyaZ~)*N6NW~)L)M2(RR}+ zJ`8c6>nY#+i$@9}HM=ky2ng(IZ`Y7Lo$H4aikPIncrHs{Wrf!JeW)kB`>g|6gN`qu z`dafp*RF<8j(s&(M2Re*77B{Qt(-ya`M;c=N;PvH?QHF!q> z8u%Z_%?akSVzy}n2kd~o?W}`LxRxV_r~Kswj2w+v=GJo@s(|4CYsh63Nib;F(DaUc zbsDloB@6*zE1$!r+G0qETMvc5k|>OIGLI-fVg}TM%@nz^#AxRwz1wEHVr&co-@pgT z#~w47l;x`I@j(v~vMxEQi*%lqs1SSzS1Xa@@{M$50j59H6y>@rtEPZH*6RRjmF}KZ zJNnN{^5#ZFIf(mm9=`?@ij}z!PJRI3lf)`S}tHrCCL3kwYg#z&alstSUO~gb#LD$79>adB%nOaFv)R8ZJq4iu-a zhg5w#+TM@m%5c|n(^`$N9PMz9jp$H)gmx!>OOmbLhKq2Y>5deTGtSh^^Anz!r`sav zCJ-*&*@L_zL}q0qsj9XagYwsU!7pVsGzsK#Q>zyMIQ3wh?gs3%>;jphn^VHlSV@y@ zQ2;MiICt?)&Sc>t2SC2_=6YmhsXt-wz5wBf;!%KR{}to{gcKlqKtf%YpEuUROKBvg37`Xi}c8`VoL9S%c<0&SWr-ruA5^B`}=wsjCN2W8W zPh!hl&UaQ}$v}XDhC}~GHe$cQ&YwDs6-q*sFT?v2cCdh04rkHNW?JwkJ$?c?`DIN2 z=yujEulLR!&e<={jAO-kH{olKEKPU9u#5aBs;VAb3)`I;oSa^k^j#LYQOV`}!XA#{ z-MTqbRJT@-Kt7*2XbZDst26xM?13~@wsiAZ>$FB1+Pq5N;qwC+&XZvza1MVp6N}| z7iR-(TV9S!QJEk`f zR#hu9f7<``0(R!ESS^oxA1ydKz+UYLiTPd~o_mT>6c;C*nT)TT0kO7-s7v3*0uzuh zET3-LY9e|GJ$ybG#q%>?n96+2k(N$UuZmcgT|9Fw_6yOaatUc!x9oRUz-{%jm{?i# zZP|L@&>lreUty4<^N_U&F%PgQMjcat+psYhVYz`7tLzi=aU4;DV;uBsoc7;%gz%bQ zGcOxsWk$-0VdsU~RUh;SEo8~4Wsl_n;7TZaG`DP8do;kS$UD%oBzr=gMfF^WZ*O)* ziUrD3l1!9xHFs-b7f(5vcQ>Uqk%if4NTA1(;|;jFjOR=U0VE3)zg_64~u~>U0Uymkw_RTPkpfd z#Q2`BC$zpv@$V?ypd*%7_W1GRHxe|r;)D?5r_NKEmjdv~-;d&i6MV5hkHu1(4=adk z7NkE10q_SBkY0(o*|38ojp)1lTyaC#QJh=l=L` z8}$76s{7j8{r$Gp*;fYA=amzASTBB^-fc_D?TQ#f%f;r#J1$SKsTg&upbwIP+*~0p7beqUA0&u`^ zYW7fHRep@flDc^2ul&oD{Vh{KAiKn>zW`@l;9(a2D-1A=L@s6D56zq29LZLpZ*Qt7 z1M!cNI<9OWNANdJrZo5OVDzt)53UeC~DlGpY-}whHS)lNp;tQ+-h9k8TuQl zPR0B^|H~Zy+Yi@VJZRi^B}Km*zAgWInBUavEs?y1o*A|t$1gIQk^`MwPqwd58!_I! ztLeN=VYSVsQTqP;{6R?M!^1&I<<>Tlz3+p8GgDjqC!qSTAIzI+rFelVXWj2Ur%!k_ zk2YWNsg$JVDwS`&CnJ$oM`1{(L_T23H$z zXsI@2^sx$0>(}gkHQD*z%O&p!a!WJA16@NBz@7zkm4uM@uNI;fUKe9FHk*dD#_qWiLKYpZd;D^kJX;iZH^DP)^{4F7@FC%_7+E3J6Ih7JVnyn0&bC;>9Zn%z zBiZPIeJw-@Q2a<=oF{I)0r<`K0jHnpUJeHLCVWK79(LjC$X( zy*5(DGE`RZJ9qC95_(6iB^)}*1OBK%;My#^{Uj=LFF%r#Dlyz8Oa07b-JdE>?-gCS z>%RZ$JdIGlZscK{oHwata7zQ^JP-K3bhqUg64 zMQA~ENe9rc+*24*9RvW#?H3rwYoSC{Ow{-Hy^_MFPBnsp_R#p9GxFt;3aisthB|Awby=Yl3e;Oy5 z7}51NB`sBpbuA2Af$R7@%+{yV5VySp@|_22Ej_-6Qy*U4mS9j?(KQ}-jxPRUIsBfb zs;FSamJnI9_t^hzpcf6jL>h+MVW{qOxEq*~@HeZ)yFKSVS8;kk2~mEldRMSsC3bjG zDGTo6)V^iOg>+rVhqfB#kmz`nkP?^JVU;32))hHk3t&saFoq z3i&uMe);0XM*~~F+Z6mpX_Ugx`!FTXv)&409$EGVumc@tjqu!s9T%79L79f@Pq@nv zfv3e(I!y{j#vLI5soHK$`V;=xn+HQh%$q%40rJkjVUfKeMq{^HmYp5Or?ifwD*x8> z{QZeFxCfVptHDc!0k!!qVWZ}-?!)%*mBB=%m@H8_;2gh`vMh?undo9F4!=$ME$|G% zsB`x)CW%Y|lsATpjQX?PYcGaXSfBGo2;MKHs@XYKRyH==D#_aRJ-w-6=Hk+iu?Rr5 zGz7vp3m*p4)V`!JA-k|4kEEfAr6|>KpJjo#Jw`B9#ckphsJwZ3$p#*)U#?SF2x-7| zO}M>~eu%NA#CiXf{CdLoYUaBaGAbX4AeSz(JV6kgyHPE-S}ptVfuwR4AXq0cOIEx% z-s4N#<8toU@zwT^oG!6)GAo?ravS5=+x$Y{&)#r{2>kgrie8O6lh*6pwV-p6oNY1o z2+a={3@GV$9)fY-hNYP#9tX1jcW28*X^JkwC^9+GwonM zUow}k_;Ma1AR)6$bDlLr6mpG?sR9Q8@-05$i#eipUnB`(?RZ$vma}n&jnQi4RhE#p z(&bA3=%^d2FlVP62&2;$@|4u@K1fd;BfDkiSI!F3)Db<{$S_@jl08{1%qG;Fk42>B zq314SWBP-X3)~Z0amD3@Vtyox3WT~Wzf2GpTlCPY9c0oovva}k+CV1X1kTn)rOUm3 z8W^p{>sx(OB-E_*l4{o#CMhUz`*H!)B9AS4ab zKpeg+gqJR9O(z!_O;6k>nEBYSS>(JdHJo|>*(`wW3AMxup$gIGlNM0>!pdmTme`w- zltS)3rx{n&2K4|e&p6awM3BPB{pe`E8|?VwldrCa+xM>F&T3)MJ8=#`HG)s7-+&q-hF(?qtS)Yl4*74 z$EZEd+*5n%!+7f?hO{0`c%LqE{>-DdYlkKkh&@R_(6VDYXV{nnpU$u4Yb)+jj~pp6 zJK6YxjNnJ~ivsnY9jv-c78pYGg*o|EDklEh`R5HY<5R?uWoIF#>AdK`d9JkfXM0Tr zwTVcd*!H6KY-M>q8^Hr5*encTSmhtG_moTQ-EH(@GDROPlHsHl` zwC|CsMa|TUdg@MOJqc*qcTKiZ*m0P+Z_p1mN;oL@YVb(;`&`Z9#`25Bi<9fYi6Q+e z#j+joI8RVKOXpkYNmfhMdTUp&fUpd+BdyI6NGW7ep(yR5i5I!#E>U) z{w1`+pXl`VJpU)glTTO7aTeRNZP3Df&u!LdK7)X>7bju{#*92IC2Nmm;M1$L2vHpl z$v;#Kd3<6|N*ojbB(_$mxF9w#5i4Lx%Rea8?HY^YTs~~yb$b5xR?m#>29ql@9)v5X0hS*d@CzISyoc4tWqFy6P6g~=I<>LwLrf+x5e3ea?(arn zT_e4P3Y+sWpIS-_uha&K^|OmO1F2TFp$Z}+rtR~?D>-Cl@6P(o-)pVp5F)Fb=l6K; zUK?Frx$d;|{mjmO+rmbd2SV?-iZfIEaqM&`zrHDRxrI2zUMFDShtrVaSQiK2McXBW zNW0Mnxro+}lC>2lL+$>W*@@|DEmjia?3jSV?X)uPxi5?qhW8&dilLf;B*1@5BepNH ziP~*wWV;^&H(c=PZ`^$`q4(s#_=;O6tmC`-L3KeJT7?uNF9NUG4M|ny5!#}@^>Il3 zAi4<&2-cj-m{z3RV{Vhf<67lr)%GjKDN6!H5YLJd0J>{N7z}|#NldOvv8F;SZ&!mf zR;B!jfT?P29#D`qF7frG6UlW%*M5#UXtqeo*Xx&Nlg%aXl6n1Exp?He@b^_jkI9!@ zTU&iLmwK|v_DGn!o|Wp>!o&wg3nZJpvehhV(`W`VxH-t9$(spNqR!LRpD^wIP|y;| zABkCVUKAV+{F<4}wc;W|VaLuY_5>uNV~v*g@!Tx zvNNLP@`hRHDNdfs=+T`&?q;NAImfV&q>36IpE6`(*}EXA*_w9=H=_`H7?uXFo!hRf zWj_?>Hfm+auQ?POF14)ePn#iJZClK*ev+&5!%Npq;w7*y1R-Y1t+nV08j-wuXlNrb z);;Lhgs1yrGZ@*ysM2`;a_nmIn>VM+DX($z!=+?)?H2sBbmP;V4?hy{j&_=sh~!?{ z?%}E4bt%u8$g|P)GQxy#$f?W5^4aHjr*`R8>Xkj0y$?;hbzg@?CC#sh2C!%CKCcLt zr8EVJn5)0dBQ}{M?PJ#hfas7|ek;i9^T~x34xKemD)T&_3kqL#U=$v}`a9(ek;m-q zpf}O|*;2Py*q3}R`L$-OAYklYvnEy26*E!mud?e#bN57}&lWZqcEdM8D^{8AnOd`U zzonLHve2o|M@*Cdq8s%1L>px23FT4m3L-N%ZfxVV(9I0KlQ(u(KY%2Wu$wM5B6n2F z+hPEN*ci#?(pmISzJB-V8&@8ayf$T4Fvh@XV>)6WS2PGfVcT+sC!4>}Ia1rji#e7~ zcf<}F(fn+mT1&rYas2EdrO6W!KJm;~PDYL^6KoRw?a&&1j6BrdOfB*WI+WlToB&9K zoJP6+Wm5Bhc31e%%Bs<}oI;`la!B(BUG*sj3lyuUg#(C69mbta58dUmv5pf~tZ;sp zHf`cp_opKDgU4OAwI!ml%j_+YT!8v{jvqb|c7cMwHGryex-ctHYGkD+isXk8-N*BX zi04ZsE`wQa40Z3gOd@*dG6O7QZr&q!O_>3<)bXv>TbzGrae*F7CgZp)H#13Mv1~NI z)RQpyODF$v;N=j*TflymZY$JlTdC6Wyf=m8X9t-mo8lbJu#1W z=IX?UQSjPV-rjOw?|V&8xSW%lsB&Vsd-rY+mCd(LvO6Gf&)vNtxe|_9=+#*?evC~J zvH7}BV8DC#Zgd!@-Uu^|pq)$r08PiDEP?|Y6>r7L`Ojs^Rae;iZLw<=HH%=UQKiuY zA`{?#w|wv<^cRI9&jv3I7t1XWu3(4d71I^tBTf(!UH7)w?u_R@Tuf4}yxw2z2pe|T`ygoI z{a*9Y?k^TVG>=J0cZ!sZEXjT$7fdOZj$SIDG^;U?f*wGhQ3D1cXV#p$$ZIL!epPmvqX)tpTr8Uz% zH|DgN3d4PH93K=;ESzF$CJjy3T&|7T^#TY&U4V6E-a_-+k-@VgIHpT9V5KAIj|E@p zjduOC)WY_ReNeDXqBeX_SeZK~(jWVT4$&>QG4%i98W+hwF%4V3T zK*QEP(}C;nSW{&~rI_5g^$Eg#B>SiHwrb8o7EYAg>hJg7ZwX;6zD-1>ZvHvrEiV(5 zVl0o#YcLa5+y3g0s@s}6XnjJxoLE*;ZWR*S{-@xGli%d<3t!Q(!l^@Sxi!LHO2^`M z4|WPkUW@Trny_8$z*el@_d46^Y1GyXTIujWF`JzYBRxy@2DSP5#mrE6{@dpiVKMyh zn*F>0>!l&<$l;ug=@t%2-)px6HXjWa^-LqBq@0E~NP0^3q!+(eB%OFjtdYmi6hITn zp`)Acy<#$p$!D-f_i^rhvNKw)*fML(7LeMUbk{LnP7Bm3{FWPC4E1`Y$)hjHn-W%T zfLalmX-c%}5KLOs&w(jb*wxsJ(!4ig%MSA@KTpA}nehBwR(b2sEVn40oNz@dPzXS} zZ&L~VX$kB1n&O!)v#k9B*sx)qgg9Oe$v_lr zQa?o4$tJ}LcZis7c+I6XYM*SIzTNYsVVBcBAt#ItHawDRG2Fv#`mJf8LaR(i8QmVu z)4C}cxi;*#pfdC6ZRm!08=o6SIj!Fh&R6rJ&fkHpdS+g zis!&5(I7T5T?H3xv#pk-Q&Pd>-019j3hA&k#==#*@tQY)a=6%+h|CMWi!0WqlzdZl zu=?tH3q8cmbD^Sz10o#XZ!jYR4|?A@aQ(SE`|HQ=Bo@M&3`$>enFXi%7FST&+4xmugIUImO| zhyZz}mPnP_5E}RIEoj@mH_1?BYs%l(QxjWz&lyYc$bSmicme5FDO7wbn&M-9{mfTs z8h#D`E_-u#)bU3r2rmTt#~#`Zujo2qg1wPyZPz&a@z1xSPKoKwT3k&l5@CykCni9r z%1t`dM1RP+*YI4FM4QG6o|n)}pEwx0U`Ey{F;r5S#rR>>1h)^FF zg7b%?VylgE7e7X-)P&ro@h?V?ytiV>ui6hOvEmX|seTeiUGYuR$+%!LcQ1GUkc zxl5y{$&uMS)NH19Ls7K`$%nGbInELWvmYKH54z+R} zWr%mGon~vzKCmb~)za|nEYJ?nlcJ+xKF+RN_Dq}_Q4~{d<|-ICS9JR|8+ZcsK%28Z zqBwJDsIYfA7>pQYsdg@VJvhdfdih810WmP4PA_Kb^^P+#+pEN@1QB=n=KZPG<@eXY z&Yt{O8~~n1B-bNgPM^3^bvz42fy=IQvYc)FUe?W<^RPGo!Y8GZU6+W}LG=a7)nPDI1uH8i_wn*c*tE zGXk;ZQJBOA(i=4)m>e1)=_@ZhLZJQa8vg4)3>@Iwt@5z2E0FTcY8Ct4_F*A5>Im0M z&OXb@9d&9JSlhuGg-W6#*PYdYGzUu^xs=yjc-L|JL|@$pi5YJi_oC9pmme@=ihMu|0R%78{eu&BraTtSk1~~$<5D5vt=9u5 zzOT4&7#`42CmocRAsr zjGpa?g>J2<`zwc6cZ|1!tKQk1n|Yg_RXLv9Xxk6Ghkj$J+upXErt0jXXx~UV>q$JC~!hw7sw9dG70TL6%sL* zSdD&QteUE-UhK={*qq}k$O3cP^>zkfh!$K-jk%@cHCDA8+Q%AsWt8-^f*(P?GatS281&vple(4IVro8}@}_wivv0i7oUDEL3*eXCXtf-V#0T;~!K!+SW< zf$ITOVw}8Y1BChy;p zKZCl*z2^hB!5JQ-@XOuCO~8;F5h3muc6ax$N7jEdoNm5dWH-z$;I`}6j~FgOBJHyy zHoC-cGF|$P0Yb)@(@qaM`V3+Ca{`n5ZnVW{>9{_JKY0!FYE7@apaYt95jXV6(v~!= z!eKdPzi3(`dUIKK`27M`w&gN_YHI{q4ODRMZ+0Dv7BaweY-hDZ_oR6>Mi#|F;~v+LT{Q(_YFzEdoi|+IbM@ppMZ{6BL@xG{8Q}vxOW7rw}vTx7|lP9R4ya$(Y{K( zV-Z3GF?~@6QiJ9^8Z=Cb02Lkx8Jf8TWKh|uN{ONaO=ne-v)%S^!C$@@f?}d5Xh~Jb z+50RsR%R`c_sUx?Xcva=%Y3&a;a#OjDMmNf86Rrc5+V!4`3yP(IZ^_x@k2k>^4?`U zy>HT&Mp>L?3s-EFgtKf~jTr;nrVLp|2{zW-e>Fb+fK8Aohvy)y|hQ!Ni3 z%NN!9^@tyX-<0MmcA&Hb6i%G*+^d#*2#{PfP%1%w{rHk(_O>I zQyQ-NtCUgVnfIcT2F&#Z|MbcH`E?#AIWB5;-2wTbarvJi{)pKV*sm|0ez1$H%{~_Rki+ZuPjw?s<9;$e~{UoxiZ6&HR<8NW>E% z&ZIQOpR`eiFvx%fB5-nb_!rLTA3l?xp!dYep7(yg)J8Q`BOspn)X)AAgiD9`z-`Yq zdtHka&N8Kfdx~EzxDJX>`S(3@1!&Dy=^$!{D?{R$IydbXbI+p=zkD{TzL-!kQ}?^G z=fB&%zklMIZrxV-r#i$3A>bkMd5=IR>mW78svngiEOQ?KUIm`>{XT5|j}zE`aI><1 zb22K1;(n37807!1M_7R*@nTo^KlVjY_&R>()#cx7DTe2RT-jf)zZ|xaFDCt`MdROh z@(G^Ang)IUn=QMKzqLwhC}KH~6+EP*`iHjSBAyeO83pMk`t^-#c$|jVhe=U(aTeUJ zt=*sPzN4k1O*_6$9mcvjpJBGm5m#%re|Y?Ll7} zA&piJ_%Y+ zeL^diG7qNb@Yg-o_{hl4-X6<&1oM=+bv#ogJ)$r3{^j=PSElZr0BoznzSB^|%P52w zL*uzRfq#e-z4zDIG3O61Nf+2Z5Fl^dxG_DpbIxP7`T4J>$m8y%=8G}`pfQLIo)S+{fe$K1oqvRl&h`ZT;s}!NZs>a54(rvnN;~; zXb)2EO19h6B^Qsn^AG}bB54gaUHTtB9H$LE>YUSVi=RS9sj&vZpEbVzxl3y=;i% zs&G)NK;54wuKE5bQDw6{-x z$7wY^xD>_>C!mU|a9*y^Rl!L<(2b1cx4-89+ftks1|77tBt8&M7AQkx7+MU}Bpm(Q zFA*pzw^Ck8{D&6`?CQ!%FiyL5{0#J8AX6Yz_qZH(_SM4EYHP!v^Vk3HxBYwlk1kJU z?6}8T4m|)jxdmD~9K9iZ;Py5y2aW&J8zT$i6v*Ys=+L7QuuL=l{Xv>v`~7p*MtsaT z%!k-0;3%yW^a!LR8HePW>i+3ldkef3a;iu0PG$u**iS$=mJsf(f|C}E04bb zt%>9wsb}(0)N~RL(ypsy@;gH2{P(N-`(mE~P*Q9BS^Ix^^JCzU(2)Cw?)~UJFq{Vw z*qJH8+mNG+{1HV7dmpXE|GLBf?!zZ|ZhFL{J4bPsj5au`UY^{VD0j(k$GHdFKdJ1i z(D%8ml@JyKDcJX^I7b?RU5{tLbc zyn6Bs#A$9WzdDu=PHN)4J=YG!137^Tsuef|-~sxAJ8v@>ot2GX*GhF}JuXeVk5(J5 zHmQxf>jVV=5KG-A&`&M1{!LOXgF)inc?`;k&1hh3)vxgV!52LMq7&^V)+@FnVSRYPq#}nxhS*VBD+rV z>w56^9xrP#JN(#C^yur2OugcYzD%tS5yck+L}UV@ipC*DiFvgjp$3fzrdX6!F>wQu zGUq-NJKe>`=tCL`W>U+n=nCkS7Jn1-kc<2{UFjv(0;p6B-x@EVPOzxb)>e%!+2gS) zqs!}R<#Df~Cy1Fkf_34MZQ37MiXg+2;DR7&im_AMuA9>>Eb*X?2n5h$R!Wd? z(0$d@KdXCWIi7F?sRArhU9i}F@xmHSGeVcTw$KBE%C6M)dV@+|jpb$WOdk^Ju(;u3 zodZ3hbD{WNtL^2=95P=!%`CS?Z0Z{t3I2hL45jaHFsxo$b76<=@~{7x{0q*ZHiT?-hM#F4j~nqL7M2NsKqs6x|x0o2-= z&``g;E)K+MQ1_G-0W!h8DC>@UVSrMA?9D&A1uyO;onD$KRyq}U-Q@gDqpYx=0)5Mt zu-hs7rLIT|wJybZq~Y5qJ=>lre;eqbxH2mX%&qe%2KbVP*9amXU)I)a6_}{h*QL4h zX7#UV&;TGVHDYl-+t>M2=CkDEQad2o+*4ytBjC7FRVeDJtkiOm+$<~AuixcS zX+x4SoVHf zl+*MaXekW_a{S1+WooL*8G+X8VJUItWlsJZRx%q`KU@ysMs_PE)~}53_3uc5CN9Rl z!w;N$iw2Zn_U?frWO$%FFVJ4K|3 zD}CRglgQ2l5fhB{`Zq>{^W|Y0Z_r)k^DY~^TNF~jeA9)sYgP7e0?Z5jtKpSb~(lRkfCIS8#D^UJ@tz+G{+fp?#n8mJoNo2GDAZ=_?h-C)`%x+TQG5 z9L&>;%+|=yJjW8kr(mWxR7y0&xeoC1Ar`L%MtGXcL%;Q#4J zJjyP9?$WH9Ov~=J|KPIG>K#9#>5YH*$?h)En^#-KlS8w2Re&Jkb60SVUhPAN<#9H^ zzTnH^F*+aavN0E4L>va~IyB^qW%>?e%4ZHQ+x&~mv1SSC^MsBJkX_|I@xXe4dbB3q zz;F4mbGy^l4+x)<6>M45QwxfH?`j2_Dnn@YH(7d(`eVfWtgCj+B%D%hBvUMiKY%u$ zJd4Rf9rqo}Xx&QZ@ge=M$weF6IQ`qLyG4}#c&Y`D5cua_j zTX!kBZK?p{XHXMoH+D;_#-^nGyK#*@W%1V&NJ#5q{+&16cJ_z+1Us(i^XAS^ZN(0v z1yq=v^x{X|8hl5x{WpX-1Z7^G<(Xe=aOs!;gKWjZreAF*+ZjsL-gbW>%Acp8R>aF%V z+q}rdALI{gZuutSRw;nfm!1tLu{BNCAli44sYwu(-p+gxXq%~97+2L(RwGwuTo=@> zKAIJ%9ySbTkMZw)@z2Y2#05ZQZ=Y91bZaDkU&`t#)AQ;r`5n2L!HoIs&*q1!R)A?R=jY?iwK zE8s(5TIM!DO^RaCKnfFh^~-35WxDH5e~t{cAi;ilAw3InAjnJawJ^4|GaT7#xHWQN zz&hS*f5%oXuya{AOXNVYHy(sr5YYq-lJf5NM7cy~WS?8ZOzt3TyN&%>@F|I(eFFc* zQZuj~JIHCsD7@lto3*!Z3tb#e_b+lEHcz)Y&`bEhoTHyHE?KQPS`+t>OJarpxoWCO z%916MA9arYH;OIa>}AUn)gZU`Isf9+?_})7wQO);(%J8%4dIcQ9`sEmTA@tjpR!IE z9pB~5x(}sN(_|Hm048c_Di4+)A&hOm(s;2J_Wr@_IMIhNE)6NWsE;*3>+M_`G2toI z99-YDq@dfH+4D*UM;%X35D$DM8<3OmkC^omoFgq?GJ&;(3wdMEJ)Ax53eFMnbSaIJ zXWtHC`)5%4=Ir5Y;f|Sno?OF4Cc$sa`TOKGO>|2~J)zq#0_5mNTgQzZi+D?X^qemQ zU1?GxTkM!{)nbi6d>*-`Chl$Euv$G6!EJOFbSNjvsT`j>IZw+s4Sh1XwK_&$FRk>fGB8g*n7drntt^CHYqzpHt)UVQA=@&hwY?35@>rLpf@8)d zZF9->-bjtwuTtZofg(2)$j8q>hHJB>h!^L$v-_}O2RAlr_PUW=b3TWP^O?sLx3wn6 zwggN!28)cxrRG)!DtMS@IvisT37Y~Nm2Aq52Ry7u?Fx!qywyY6I&xe)OaD|}dz(bF z#JPIC$aH+-P`Jwx2zDLOpD7G*^zpV~V3j5Ocd{ed~a6K;fZSSBatA^VpX+OQ_s;g zAYiZqZH!SLIC(xteIPBa|NPkwdibFSn(^q^+cIh}F)C7x&Tsk&asGTMN&j648!;7h zB<lr@*H6T;}Xps8YvaiIRG*4xFH}-S$1M-Y2ipj-`Em zYOH)y>WW*wDwIcNraf4~A#42rpxJ=Tp=C&YCitw`>^mtbEu7}|9&5CqWk6xxYWjx* z(f4`7k+3e+LA&@ug|5=RFJ2>30!(5*%oYcTSD(hsUF3M zn&&;_lTbjeu+fqBpNCWM!NbLHV`dh;SocI~Dn{vTi^ht>uT`BO%DR(hOSe^7Ea$bR zo~uQ=AlRg#jQ-Zp)>C2NXr9B>mutGOkM>y>a%iC>N#b5j5CId>SeBJdH$K)|E98sV>DbNd zIo9Vm;jBg+{W~f7pO4jRZ|m-F*({Z#U;*Q~w&qhaZM#NIdM-qFb_cnv4;SWO{kD42 z4{GP__uuGjIZu_C56)Dp$NM?#IwdFQjZzqGdMAXN^}=+JInH@((azTxZ?l8!OugDP z<(C#+J)-cl!Lb^3s4>UKr52&wUdf%;;pg=n0)ui+ZPmbyufN(?B_}z6l<^ zO8{sR<7xRV60G;8b=`GJp*0CRxes$Cq}(d(p~)}qs24A?TJ$-ton~X7iL-7Y(|6M9 z;M;r^W8YXaPnNPvOlGN8V!o9S#_W(^=)Tcsx}$tRmv1uG63r89w>rkxLm?-sR@{

s0KhU5VMv04j997{db~De-mC%Qe@QP)Iy@86!1XqpM68##z0qH8*8p zfD`Fv8W$`nGSAks3c2(+9Ev50RPTtbzR-S?iO{K+cH?^WD02>sr9SQ4%?PQ`NWD$u58D#jf>F z$6d}!<6vQQ@?`E=wJ6a=9g;L{Q&oXFN*uM4@uPPhN3Q}26q#qXE=ED|LbG7~# z@BTBd^g^ZXFsHdrRF6?-fSvoEh+~P82*OCTmd^sygh%HY3@SBUC{9=?fA;m=`-Q&V z3`7}JU}w@?A#VI;h5bwD+m#TT`$xa3<|;T1YxR=RqCcmVUuA>dwW)ueQ`j{&wNhYw zx>dlTIJbOUH!v%{e0)d1bpM%&T3pVk1S@EmcZim36i#wiX?~@2lLAiIZ7=woyI?TB zEPRn5oRb&@gknwGNTf`^m z`@}BJ&(wr8AVEW5CsX6@biQ#j-d!8H%(q)kjfQXYml}44fzeE>*Yq|kt@g~@_;#Lt zG?A)&iJevx_a7J#Gahm*^|r?3C`b^FMOV*N55)WZXV_7A%gE)E`7N$U!kVH+U#Cd*0i8mTr( zl;5?fm?sq@R$T-Ky0BmavG7{d@~x4(FtmjzMan3D226*VRY#lKxc%Z_{k9Gc=nAB| zJdV8i<(D9UY}RNYyJ7R%@5bsSEU)7^$gropBSo#Pi@X$kRg7s_#L|HHEa+^UhVpPk z%V}!jM}+sC(TPc3=QIj12}I(E0z?K2vwKi(RR zO0*Qhh5?0`56}0<#QAsjCflIv$%Gwxx}wV;C`BY5E@H>M!*v~?Y=X>1PK!h&)G>w@ zSvyNS(A6E2T;o|9tZ6SYyjx;db8Gu!4!kzBp_ZK|i=00r+c-4q$)rWNM!ptTqC)iJ zoZ3UkiOh{>jOIG31TaFRu`3VEeaF7s`mPCcyTI#N%gW|{HVuC>9%Y;DB>~gPtq9MZ zYh-iovlR)J^UYrk8neLjfep~(wrRg9dx-?xavYIU-qrZNGY*h8d!Tl=Xhru zW#VQpDOTHcjB7hdb;rB4SqSIS@5I{Z+;u3C8$GgXx0-a{M_{}(Hq}^F$JRxPm*@Ob zRPCt!iqO+n_%lSeBQRd2KcdTmPU)^g%=7Tp5FTlVY|gBg2YX#r`5mQTA)NN&(@efP zqh@kh=C^Kd9l%sN&<^nA<5nkpXUo1H6|MfGlXRtfgb=W-R9|h0I+IAwC2xg!LggeS zf=G`187+Mp-_$=x$`K3>eb2~z-1D*w3vb}_AP+?ozZFRw*QR3*f5CZ8;;8uX1_6@U zd9$4t9|GMM<}~QJ<*Ia8wWrFl>=01g6n>Y5T6P}%HvaLn<1+2F3rpf2mCLi7KtxbT zh#YzED!sy{e+R08PX`6iY?q+7Xu2M94Cd*h`0yN%o^&~boY8u=17|zCGv-=C883;1 ztM|mE>DjN4$0$mjM_W$j7`F?s#U75Z>KsmRdaFI3v-xgR?l2QWl6RW|uoGKDZ>D1o z*VO0REX>{AH+EFDa7~G`S?hr(2YhlLi8V>naB{(`dWK5--MU5hLae^J%K7e9zsozA z=HVTT)|s~O z2tqlrBxBZdeEeF8+pD!hr;x7e@qF+PN*f1(VmedZ1XLI{-LfQM%~^MVpo~kcJUE!^ zdGoW4@F=GFzhBkKc^K8^hh+B7Uay{i5$aybps_;PBP~Koa`mcVpW_1WJ)p~DB;$V7 z3WmATQ(5EZW7%#C)636{X;xs|Hr$K(89)$qZQ>5>>wDQP98 z1w;&*0i;nFU`T0c1*Jo}L8Tc&Vu%?A?mOoyc+NfdJAQw$f57b7`;GOk^{i(-YXqPG z|3W%#(^6FNf|z|b+%h$xRcV<1nDl{ny|f+SDK)E;6Xhl*bqX-wJdgD;b>SOL!qpSm z-rmT0;A=1T-lw&5{m4JL|EZ@T+RfjdFTB6X_ktioO=jP)>z1HSmkKPNIc&KNM0d8( zNz9((xBL5!->#d-NU)YK#>4o~pvrm5 zC_?Qc*>0j|0jLKUEt0w2V%=bNp90F)Leg)(=#E4TMgw5~y8f`|Ml_7iuDbw1ZU%7I zA;d7lfUkgAT#=Do?Wof5C>cgVzsQ{czB6FHW4po3I`?W_K;kuwL%*$&Hoi5X)Ed%k zRj8Z)BYn|Zrgz}&(Lj4C?(9Acdac=vD(8C(Kl4@ z{H;-86!d81MEizyl`|>W4 zuR)2%npW7jr(Z7>dtX)#DsIq-ff??(l8eC2VQD?)bFT+aPUQ*?nrc8`RSamjyzr3mKo~&+7M|Mm@ zTscCBVjR$)(V^N{xeHOjw6GGQeh$laVm4p@+2XNv%gGn1S`Qu{r4C^(Tedg~J@zhI z;_iB=?Fvdk*W+IUZ<=4F4%1+eUwL>_1BBp|*O!B(+XADN1H3nt0esCUG`!zYWZe#A zL_!@U0JH#r3CW%3_+KZhvR8(OMbI&xG+qbBm8;N(N{9FckChTXgU(eDrLhU-O4t;7 z0p{vwk3>OYy)VO?Jze7#=$(`jkE2s)<`+Fb@;d_lOqw`Ap5vtbG;SMzvXoWeE?BYE`f z`FEf$)^E3VaZtW8&E=qb>y@LK$9OWP%5EkPj6NNA5Vj(Cvm1{(>>gqi0z5A&gwDGV zgSQvXQ;Ye@O0EdDE>0h|>}(rOw}W00VJMb7NlrD2;Ru1V5_Zvq#7m#S&hHf2npvmnqB6DGX2r108nz2DOcT``?`#kQgM#~8n zywf@vT1rXJ8^vq0*SO`_Q|jG7?>zg1*W4JO4U7hb$&TYyd1Yn`-g-ZZK>BDxkbl{W z=o<48jp%)72EK}dAj9jaw!^IJV4XJyI7qqv*ciu^YEScb6tqRw@p>uI<^=<%g1mlg zbC-}@Vojg;#}Z4P$_yFe4tgn&r@r4rw5%Yx;h{|2L523!!|hT%u}puDN`90f0qb$T z{3Iy(Z&$@v7Wp5%2Urh`!nM-rfjD9&&!;U zg~rLA7yN*CUQ2o-XA3~DXaK#s12I7;K-($ia%Ay#4C4R^V6=&}7xTt{b<9n~cRrLi z4hY*ZJf7XLpo-lgem!MDL@knNly^ELyx1zorY|Q|e1D|iN1Q|^l*5B?X|lZTp5yA{ zn!7{<%cDhDc+Ix2)`6_bYc6;3M~AdtnuyLVyClP#ZyVVs(S(8qkmmfOd$SlJNSFhg z4h$M~+FL>7^Ze$Jff(9K7DLE3D5dYaLHZu3hJjGF7tiTn7s-l{m+o-fiULEJ5_R|X{e4*z1|_Mdu`#tz<-;hD(2eb#^2zG9TsTuLtH~Z< zdh^+d7B;aIjjpi0=l4)je0-4~!K>nwy^R>kE-P<6LMEQ*HYQM8O zlEQ->MfTpJTv!3Vez4sIFnRWV*i=KJSMfxuJg-b^mPSi^GOkXk5c;-32*i4ZHl{H%yR;? z8@=+Q6~5WMX5Go7nY$s>KqFy<;w52_t6r&hiq1V|mq*WPZiQoX52xHYE`lB{e&)^B z`P0qUwo38$J9Y2T)?t$XQ=SMtmzc7989BeS01|V(^FgkB`i?)kz#vn$s^oq{gzID5 z2h%jsLJqUUVS_I+?yl=NIj2?@zFITw9|vpD*)(Wmiiej#g&MW~ZhP6E<;ha*S+82r zhXWbYcLjAu6zXixe0RN}&a;C^mQd86@euZO9!5+Q>=KK(fP}E=(L?MB__|802UMP_ zYLMpiJ6-Nmy&a?tRKLk2NNY_cAFp$fzT~ihcG2!sUJ+@~0kJixpZXb2-x7ZFX364H z`jl1tP`2>2jZSt?g`Y68)5Yto=t7^zmvcaCTPnlr9sYHzDqo4@)nCeYf1kTNbw>Qs zb3oGU;q|qWIGS0JlS&$>!r*CZO7xqXZM)D+dJ$FRZf}Xe<+Q8B88Xv|D8FR9dipG@ za*Syf2Us*8R$n9(#MM*v;*0JDBe-LTtV!-kNlUfOGLA z^3eQ!AbuT`Q~!3?ITtw)BU<7<*&cCyD2;ax=!Tj0BE5#PfNq#@qttlCdR$nNkg?3G zmR#Lv12fyZYup_Mg>h_V{^t3JFPgC`q?x*K^lV9@jP-apq{Oz1R^J%%gOL@sJB|hB zG)r$L00FJAa_>nT>GO~%h zuPPl!RKnp@iv4J5s~iw40LjVr;8zQl$g!{z9o(;-FaWkj3wt=4yDh!c+Z@~$xn(Hk zVc$Pt&m7Z`NZ(LoJ;H!?4m;5+fkEIMsg zlH--3A-b;^{OBrXGmrFnO@pqo&ENr7T~hJ;n(jGTc8A;W`lB>|%Sl>F28gw_rtM6c zvBklswo30SaMS|@q_=G zf)`GrqNFm7*Zsu6WQy|eVKE-=cTrZ@YN1&}p;aga6L3yQxL|}@wL_V2R4Kh5!%R~s z_&g4YxW`vHaM#;7Afja?=uY|gB9o|(O;XWEhhnx)%i2Sw>=-ZaWor$fX(zX^qRWj3 z8of6eO^TKC29m8+C>mr3ad@F@U6@lA$CfoaDT@T4AQ4#OP~nMiCw-kX>z-tofIW_T z;RS+}v4%xJE{zr~1Q~J)g})82vsh9wkKt#9aUfkMz~Usw468q3*xICWpdta&M(rC{860GBH{t1YOj(d4Ed}0Rr z2K>H5l|yb{5NlU`3u@rgDUp%H@m(G@-6)F1$Yk=Hu0U_pI)>Sw%Z}1)2Ta4=AMw;Z%+1O3<*CO(NC8=AkVH#$f6mH!m=Yte^4B9kRN)IochubZI? zx_G^4Y!{0<0aWr+pwAe`nZV>d#2IZ`_f@Ev3nWCwN#}e{AOjzEEIrKXYb)(NIwzj5 zkx?_2!)tX@?w0}YV)Wr#hn1!w4X@liT{B33J{*?s@sM$zVKZE&;Ws!;s$z)Ais}xPAxvh{NV{SivttXG8AU##j<0!*d&>;BOsw?# z2sjKlS;CfwV*x=(epxPrg4thCN9dqxlnI(K{bbW9=B#a68Q+f`d(Cd|!He}bgGizX zooi%6%;r1ISQR5uAKJV<&7?bOMyBeW7d-=yN4%3P-3lp=l z-4f$fK3ip-`p&}|I-EydEV*x=t2OkothfnJ>EXkNneN^AW)pTGU9ywyhgHkXx7FKXSUkqnbcYRlRS&#OXXcDKdUXwA2V zsQDb`la<~m`bw~y!}`?zkSlcR%umuX)z#yJneN^>aBS)}z8BSs=%Zi;wFkgtSmi)`29sz&%nXIc>15p|pBuf|#9_7g4K&qr2Fx*_idOE$ zpo~>L^ImAtNQS;z9>Id;q2&A{!uw_UO;>-)puHq?Se)5+=AE8@OVFKXQy`}|lJ`Mk zpHVSjPOpcykHOyh5NAA{6jmb>|qib;A*)P)?;1 z%QoAC!%wcsDA6?CVI{m=XE?eu*>`!tEMAckTC!V&{sK;0*?!91gnhKV(~R75}@fy68mO)cgjqE+;!~uG(L-kKPVqJbz7;hh-<5NZAr{_y%^=c6LZdZ={ zcx3-8UF-O-A9k?>L?@MqlW^|^wpy{V_O;kxsP=!VrZIU1X&&r1>#<(<~?8m+R6*r>AGHJvGa5Y^b6eW;bK z@5W~JZzZ+=W9ojo2LVq=LBjG~{Ih_Qs<<=Q6qEFO{Lnx2R!%&P_ch>?HI%)-k*v>y zkyB|TOVc^2s5trmzqtxH!-Oq(fpDKa)+?qX#%eKsH%^{8dDcHfJ5sTwr2n?!qMw6D zWpsWcqV_)~<3Gny=?8cSk^`pM-#+(U8!S+h4Z;B z7l56~uVj(m{)uQHxTAD5K$T-HxCL2Gy1ajVA)OA|?B#yM^K`~p`Dc|aU}!%DNz;iV zlr-6CxCKH^7Ln3>?9{B4o6nxS$RAx1`(qata1DN2SYM;Ck0?$dF813mPiF4#u0WAE zIV`&MbN!>4XC`%CT{p)q+5#CZ+o7j?=ZA|FTi-HuSnQ_yZkm*{WZsEt9C};?sbRi! zNjMYZS7%BOq{p*ublUV} zhyHe@{p1O?toF;ObSR!vmdDfs`f^+S-Mg8-A)sq$?onhEklNJ3d@WDSVO}qDBL?_+6up|>nGG_ zuFGlf>;JPXE8d{O?QlKc_<}73|GWH^k;WW4P{{+N!`y z9&4>YddTuSbkxc?s9Hm{BR+8vMV)apR*tA%iEUObvFNrBcezpsX`uJGe%)12hgoW4 zXJ=^c$B%PWU`KN22@Ph4cNheL{m_JM;tUw9MSDQBXM6@F*R=}d)&UAjiWNb*(i(5z z<5T;&KYgIUxN>fz@is3OaEEnR(yd98TAxL4SAf;+uu(z8 z?>lr4g#twy`%HFmIf%@{mgAz-PmFha(xDKcoG|5Z$=}w?e~#3D8%L#moE_9{Ao?*w zy#PRt+VgmLVR61_q9xk^98b%U52@9)>mTlsiyvHQxv_8CPNvl;=~O>HIuV=O1Q@e} z?6KQ>X+n;%Z2Sw$fN!$I^v+*+#wcU155_T%)P19zQQcSjD2R2~jbc`Ne>FnNvT>kj z7FFK*Bx+ebX-Lti{6SWbVrhM9QBP|e*Q-#B{k#5A+l8#9()sr4u<_(EJ=mRI+091CkzFIzWb@9T;m;9DNZvkf6A-NnqsPIH2f1bcpO`d_`fXVlUGmbg0!Pz!VdJGwVK`3HKcx; zeVuympPp&E++Po>QW$6^K*ZbT$}~+es$Mq@)TXSok5}%WU)&mW zgz&RO^jRWjc2zHduMD+!NPuA%)G)Wp)V8Sx( zQ>o+uQ$sMYo6OD3%m|d(rF{evbt02i1S%;F54A+yXdi^9h#u9Xb;x!`$f~!VNZt5ZhcXucC(eDcPjJIieG19MK&}LwtmrQS>75XBAv)J zE2^07xm=|hJ1TP=7(K_fa=Lk!_sTy$)ntL=U;K4AUrTSwuP*qpw|5f53To#MpXF3bupOOUv9EZsw0pJ19R&#}@ZFPF8bOhC)(h zLyttl(F7_$54PPq#Y+sL12LA+0cK{dm^~IYmJ!?eA!(+*_E@@?7lacpl|%mNAte1l z5_^W1vue^6%>uT3QTv;Qh*qSXa3?7RzwT29?w16Yv z<$D)JtDY2|ykSSao)?7=8)VqrCQBE%TO&1ET0g&v&!`D=^*M3?w8P*B=%wRhGao!; zKE?4V@yAn&24WU`Ls=am4lwQDO|MuJIn4+ueZwj51%hl+S89HT#_aD11VUgwOo;Xc z{&{%D&!g3KKRe9Wvc0panhFreUg%i2fu09Q!B4MO1{g#M`Hvak6XKLeaF@mh+ zzAZ(3py{p419U}ElBJ@|>LHX~&I{6Id8S<@j&*r4dL&RAA#5K7FRd4BL8Gl>+DU1l zPbR7u&6_Xs*^E(@@_@c2;jgGAyi6)6YV6_E`QcO&-yR5=Nl}jVyp~fh{iz=hUi-N^ z2xndS%Kl-X_ZQ9%P>8V14*_Mjan?W2@_7|p+?^;NPv<10oIh>ssQ`v9uifs|xsT44 zohX>9!CX_JZ-L*%PH6C*rYo|LQarrXvSMl7c zpiNp=u}*VD%Wttb6gz#;%y6j zPvzY*LTb1D)W-Q>K&pCmAJwA*Yni^pAd8(n&aTVK0+F(} z-Z|Q7N#naeAYba&AOl|~d)N&sVDKl6ubA44+&J@OcX+(`Vu4Tnz0NR@6pkpU**1)9P?P_!a5Cl z%GN7ibpgqE_RF`7c|zsz?x%$R($0)Hlie3$z4mip|{tN5XAlDQ^XuIKq# zXDOiC7_%-29f!eD=F4810^>Tm9&x>Xxc_<@+H%BtCV5O3R#SX>IH+0Oq(wbB+ArwV z>#K#Q`7L*15W|j(^Vnk`B=5rrNbk?rA{>&J2@n!oMYSWJ;9>Z`U@hVlY3Fk)lY-3a zmIs%s><(Uu11bfkvhg2FIX)!9=}uGK?n~1x+p@l*W39=~C_iO0J!mo&Ay&cf9>i`v zL>;U(lhxWi*&vqNmxsXHluppL^4-sMEBO&Zo8oS-UBm19Wle1fRBMqNq26JPHy_Ns zM2XvdPzip@9FF1%g`J&nPcXiFJ6F6El|7EAkz;Y?Vt}CS`wE&`H>a-t7Q|gB1CYQ3w;?A3_&I+fz zmpX2(8ClFQs0?vBFoAd$+sKvjeD9`CsC)>|=W~yA_9Y~}>s>mi>Nq@B``1soTVm-C z3LmCGtYJXFGh;y9;zZBwl6?ZC{T8nOovmUFeezOubp3QF95fM`zToriJ5!aj;2ogS zKyS!r9lqYtvhRiUI6C5sltAEuO(s4PJqo`hQ2lTz+N91|&8L4~rCz*>$ziOF#*9Hg zK)1H84tZpvq3vtk1Bv8pt}UTBQeB*>xU?qixuz^P@26DNk|!#s{VzDjWtv}>C4HqjsZ zE!e>i7+}8}kqiaMZh@0@>WPr-2ZmEMXJF@)J z`AHgR1fYKHrlf2yNeK@1-AmE~QMlLvAK zd0ETh*cYCr=&*+aU>0V$7%l6J4H!7}#?GldI+^=Fx_VNF-34|MuTQkto^)Y~!hGlgs2Ek<-*^;Zrn>LFM zmTIjgNOHr()K4;56Z)6(6b6IFS?$w`;ccLtWo}}>1GP9$|;`a~52{hNDUUQ0gxDv1I%-vtH-P}7>pk=H{1 zaH~F%<^vNQ8%h=dKZ#ixYo%=oRJcqlRQ&E#Xj2I0bA(Y!CjI!b@i-*Cr-h|eU3-X) zS?`YR_8xo_Jtc+IeRgq$t;%JtOr7I%oX6Us8nb!=ENy)-KB*0~a&VMHge6rX zP+t!xnKYywOMChQ@z?S(Hl>_qRks3zifSad?1nh|b_(>B>yt)$XILzLylYL@O&QZ@ z@90YN5Hk}KY7W^#5g1!^S6@zayzzRhx*3@ zFq-z9*r+0-_&hIb)mhm04#QQN9c1MN!PPvj&@U!kN3t%dkH?V>^cwTs*N!*(@gZ@1 zfc(3q%g2!mtdDGnpb<_A#6iFG-?6unD8G761vV7BP<3J-#YvP3WY~s3XlXwU6gQ%a zml=8zuWAsen;m+@DN{%}sjy11`C81W7-;BdY1GW19cBT_@YegOx2;BN`oNDUb8K>9 zz80UXoyRTBQ)P3ZgF)D7GOi(lDyB>7=z8{%h+Q~btL|l)<6vVxixvI9UN9kD_kR8O z`(vft|KS8af#6oyk8}4opt^0m&cLpVfDT7>TsgnZ=U2{1qabY+`l?`<$pWG_`n-nP-l98ewYj|-2HUy~ zIo5jPwdfnwmR-;Wlw`i@g`QHl41CX(nKky#y1?3>8qm820iX$sYGr+z(wztIX9-)` z7P?Kp=rFO8R^10$G^lRN-?~^;5G!@bu$BK+_$}iT;n~k~YQ)nu$|9+QgCE?~UsUDk z-j;!bI*judUJ>&+@wB~`tP|<#&+chmDqUcr^H8V0zPYm_guc(7Eaoo*1QEFo4@VjO zr)r~x3N)bK=;`T6X#5oCdG=+59A94KDEG~ag>q0q#at+-6m|Me8#&48i#G zWDTL@Z%w9qBZMpUB(-Jo3}?Xh;{9H6Y}$|u{uMC zWrk9|w?XtIk_rSoIs%$P<|e9Cy?j9WI}>QfX)nFmjBeWV(N+Nq4hD^#75R0uQ)fuz z=|I2QE4g->t(WLjh<|w3<{+rnH~8jl3VW{UoXfMz-qEhMc(gU~l=IUSC_c)kmg?r? zJJWuf?=>uUCOk{r(CU~1`XLcPzrs{7w#J}Z&)i{Xj+nt-(0r$|#Cvz0J?`cLe^Sjh6{|tf*(A4NSxj{~%yZ_fm5aRmQ_efJCPVq-*XB@u2?v>L{NasDq!LQ=t~$uDnPqqCS1yx7n4?c7%>$@3P><`PV^Sb&ofqUDqie2&qv~Ut!|R z47_dD1EKS$m2d=UUVPGamZJIrlb2*1-ygi(^!h%)X0EzY<$SOwrS5B3Y85qXT+ax8 zMt^L&wNx)!o(7Fil3Eu~*9$Cy#5mQPF^ajbz34M2k=OFfn0pn?qVZ(Zy8MUG6q(#v zWHi;`XqTU!ra$YF;KQ6`DG#m6hW=`pu6J2U|1}R%MEOip0_dG?Id(GGe}AR@>cf=O zE~q$*k@xMkUSz4WoLfc~pM?#?8xhe+vN!j5sqNi=KhD#y9$pB%rI}n_>tiDtAb%N9 zwUk#W|kBSMJj`eG(%k}-rMnCzak3dI=$*dHQFwy zrS&5zg`L_yeCP?C4C7A1tlBbJ5dTOnw&Ms;^Pahbb=8kluydIWo;4MGpg&}QO zz3(IOFMy7^3LQPB*IsQt1o~83etj3E)o3$xAy-Osduc!FN@v>iSbm2==$8b11`+qF zmwX={Xx;-ykY67Js6*Kfvj~RtA0@kaR6FVfz&`j%B)i8a@_HBAPgTq9=H1ZS?WVn% z(%>^FYWVI!g5#?WZ^`IJJH@;@4 z*^U}53B=u0fs=e@R2|B9UOVsvNjr}!Uw(O@v&6pSc6g4+7hYh|Z6N@e@_XkT6A#Y`s!WrlTwdE8;<3RcCl5y>(whkPU_iV;}ciZdp z-cYP$PzXYbBIiPm6sx!qtOR-^b*8oWa`T#cP1tY@YOH?AcP?fMl2VaJNEC4cM?Ed! zR5qs}Uz>S^k%Yh8ayn<(!ygr>U}v67Xsia7ih(?qCM%DF(RBZ7*Vu zTC)nIIzaMvTX2Bz-f3?)!3JFBff2QQ#Rh5qJ*!DxmDS9_yd%vzNgH2Zi;?sxb$X6$ zLhTQ~d+Ok%y=9>lg zFN#P(Vg1sE!wLg-Jysr@P@Vww^LR3F=)0HVC!LC<1HrUAXSop&R-Jv7tjJ=@1(Ga< zJ>OCJy%J&6LE7xkGi7gx9F&Twc@MmAD;Yk=1kW&xdDi!JG`s7NNOBD(Cu!TlfLx`M zy!E8bD>O+!&Tn!n#Bq7DAma*`cI!&&Pj6;etbY0y4W~i#ip!PiZK|9~M}3*3LkUsU zRPj^c+?Z;&al(Y1K@*6(&-e>-Y1uO{FT!ir7kVHAm5WpRzHIwW?>ip#ystN!4*GnN zl5}TtbCZuJL-ux>A8+qPF;Z@*8H$21ys6qehL1utJ>CK>Y%zb0-(hI@)?J7A4E919 zII@}0&?PjbWZ=!=HNG4Tk2WYFEfKww=S*~^;G(0 z+^(rKi(!xDNE{9=AfE0%1G|4Y`vfld`y-H!2|tIYoaUxQrk?=9y6pZ16-bRcMAU7) zs3*^~prjV*)d6jt*}e^goo@cSZv|AvN96`hPRm2Y%H#W*3l)?-Yy0a}Ln}R=rkgCR z-E8Y!&ruZa?p;TTS{_&zs-4%~bybAx7YW20Y3N4qg`5sKj7yK2arH*pMqF#yu7BIX z{;7W+)9Qq;bt<8k0jDeF*;|uOYC#|`DV!}L82t2UD>}t%siF_06^0|G_|ccIvkum~ zXq>O3&RysAvPieMoi$OA44tik@Ie<9`kdqX24;L*!gJRV?n{ybamK!Y! z*i)=~5^r(&w56)*0-*yn1m!l>2-}=lMpozJXkiPL+CT>13uwEwqaBP9CD~Hn`wb^! z(xKU#p(h~Kzx(6=d59^o;Vf@K>hK6+T&#JXBdqaXL+VvQ6_?0Sa4jb(WMrYeozN1=Y<-4fkHN+PhMg91KjIwD}_lQug(OK zw6`W4MKAb~3%W-edR)Q3+R8E4Awj31>#qw^2uly#y^x^1$~_F_)B4bK1iHP!&2TLl3)jqL~ zUURT4?HYZ8;%txo&qS-P)v8V24_0u#$jIVq>#xIIW#Jz9HM-STr5$NWY zkM|@ikSAZH08)%W3in^0MB>Vz>A|v;x8nQO-tQEA&Sfm0?EnEqkY#v(O*l{w78V&Z z^UzmP(c8Yr{i2E*%`O~df)x_HfZH#4C!y2$-KSNarrZY2H$Iy(HI<*GyoK}xiGYMa z_Mr$DL(A>xJcFRFt+yPEw=~-+IPEs>n|sc~6MNNIhTIXt6=gnEaB+-9*n^(YYKM^9 zXxqXy|H6-@eX~{7Z|zY;hq{##x`(Is-(kjkQ;QCBcwIj|m~;XZnwxGaXU~R*?$j=d zqIQ@6rNGfA;n;VP+E#}eGy{(Ac}DmUo%+)Gt*V@lXrAh2sPn0;Kg~}?e#ag2{i1h$ zZDM0jBuUVQl_1r3Y;_OBq8;_8LoAx+(=4YCgXUkOF&AS!=_#)O!0PU$*T+oWCr^9< z-m6d;X7{{R&=oqsv8b%6(#2ctStDs?lFCwT*E~m&8jb|^-J3u~WFD*dypP+TH4;e zrNC+F7nh(Vjv|;lgYBC;lw4g4FYG>2_-TvD!&BbA6}!r5NZ;Xal2JqbhJCL8Fe4Q> z6Nrw;!zfFS;dfIVfPguvl{-0JR3!$f3=Ro?p)pogeW1Bt63q<}%`Rrqz%ZP~k#e#1 z-VbipURZh`(gsYm+Vga1e~IRwQoogOaeU7b#6*tYb4(Ke#pMSkaVY!vA`K&(v7A(0 z|NNKZ8$-=O;E)3$!cx{q{nt^i)X5GInA<3IDa(rSEwuk+*y# zr!u}3Zh81hMsVLJpY@VZHK--K%)x4z@)$@&*rP48efXVOGFHkxF<0U_uqU5WioV|uYzW#Jh#n8iQQ*E6i;~BvfGcz+ph9X-Yb=6q{ zS~g*)36gbzDUzG%A(?>L<$W~4lhLf+H=`QI%04OPPdsO$*p%}q#&Z2=XUP^_Q84%I zCDF2TeClbWqw$nYU+X7c87&{G1>+#ap>$mkc9tQ&s1d0m5Db@Z5qKn%WE;4pq+*`6 z!$A95yWrqwh!U?U0EB&at=n_e7iMN9W%2uCC@%BiDa68ypFSjw)FUOO=RPB_2tuP+ zZavv*zwOkPg)vV$8TcK2>xW;iWv_n9Z|Z<@vvpAp$CNa49QXKp0vZS@0X3RoVy*kn z0>7iey|R7DA~YU}hak2*eNJ_JV4F$tm@0Pt?9KZoQ2X1rOU_&KY7*18n=4;t@-~W+ zviCdx(gG-KNkk@DtnI#Xb?Kten?uI&B*+2ojK|btgT|rxm&eHYC@qum) zcI!x>i3S7M2W49&1#Kpzofm!pCZZL>Ys%*oUyjG{1343QGE^l=TK97F8Q26{|1tCRH>B;qox71n zxWIYA5(A(43hCQQA)ktC)g|tGG9fd}Ca~6*iHu$Y>yqU{k#^S~_g+OrTO=*wA~f zRKb)1U^TaL;Kop==OWJA?|CD5K>VgZZgz%%ugH25(Y&7G z@QtLMK&F`n5Rxes;T9nS6ZX!FomJ3(BZ7C0b+F}}%5NEyOp|~CV2TN-ncV3XO{cXU zjp{9%uHTd3;q{2o-c`%_m&quu(Wb@v@U%tp#n(4K%|xhdyj7ST3rpJE;t{iK%1U!f zL;Z5^^PPYeJ+nrd`4ESmk0uSf*{!rR9w!>DX_daxY2wlbmARMN#j+yGP6v$zI5!bO zjzS)W+Ce0M?zx8CA8FfcklKa6m2?J8OORMtdaeVT4skyC#5P*z9f9>Nc;!`MIk`cN zhcF`dAqJ(kM0>OG;B+89g`Q_-AW+DXR!$QCW3J{{|sTyNNHlxf)ZnN6Q%3kDRh6>k)HCuiH7;dBe zv>S8w1S;6L@+BD?)yx$>)IR8GeuTSqALYc_e4+P`iPTRBaOWw^x@(*FqOIIQuk~q^ zSa&Q7sFu47JY_4L{qc=qe!)NgbbCAp16NqjrMd3d{;}kIaGHsm)Z?n^KS!N)a}%&L zF-s($e%p0iPeB*CiK`889*UbP-1KZnM^}Q!OtWGFY$o^dJtum?wcCLVv3uLLq08hU zpS_#UR(bsFMoX?y5+fs{yeN|D?%4vW6y&lv6OVwex`B1~{t#Q{%(a=v1^e=Pqwj5M z3Yo1mRf0d&Ckq+7Q`XpeIu8^g-`5_<#;}Sx!d>*zbBZp-IiKmCfbbL)MO<7&7N42z zCWRRX7kCj-SgqxgT0ZIhlCLa>g|^nfPJ=gtlw6%ltX+;u+TDGv(p!3-b#tm#9igKx zmY{sQdTlJ5BEG!=1s@yR` zy^l9~|4bTHjkL(wIgp0o5%N@L*L(b>mr*eyEfR}hSC6$VZ@mFbr)pQL2KF2&XbuEM zCD|4Bmx(XKIkMpyzZCzmvcWrB`!v(~5kZB zhe6dIhu>)}ryDF=9+u}pMMn(V=^K9)ba>%bug++EI@Kq7m;Z zIB-c;!J7|4P2bFiA(Th^nA{9;Gbe&G-WscrCb9#3W_-)r197)P(c`eLECfh!yxws5 zvE2IlV<}N*FePoY4(*VRqSws$$Np>6_&q14hqx4wN?!Y}jMW@{y+?j;g_CqTWK21C z!^I7Nx9unY_h+-UL!vY&!$$fX|3DZ}t`&k456AqKu5y|_(#V@VciXTCK1OnX7~W0m z5g+_2Zqd_mx?0P5p_>$uY~7j2bs7(M`B8tiA6t29J~=|<;_+~3W2?AlF@(>1Qee7D zxyIQ$ZIn+(|1>xKfBmn>KoCF}25PSjRfQIJK*tbR=ZMS=(tbJQX26D5$v|gGpeTYeG1Gg{4!NtcsyWxT zUpTy#3A$)rW@KyEi)c`2zSmynQ9GLrLN=O+t1%Z(fJzEghX6ctU zBVnUk$gXM@+~Idnd?O?;kY02HEEZP7!sD`cEZxU>N9nvE>1*v4&?{jZ5fBuDw1Y=~ zM&_Y<_o}M3$FzmTSdX9c1aA1dt0xY)+P<3Yf?r*E=-FD9tJl=sq=ubM_^XEj?HaSy zW3!KL)oP3754pl+B=6(vo00TV$NSEme{rO{lSLSCFIIdB zAsqfXxB~hyLpD%}#4_;lN~ex@=^B<(2MW$J%>>!|`fzQ-XSjaXKffe(v-Qv=q!H!B zpopT=Dv!QD;4@3}T2?0Kdz?Od{@ji6{9pjGAbl77&~@d82L}h4W@)LLvRhf3nlMFV zvaPr&q&`t(5Sa5EJDlfkqDK_!W7uUH?mr&2gm#DJ~p41M2z|4_xm}; zPbAK0yJ;@_$*K)aX0%_EflE$`zUB}G)@lVUEKfT8Bi1wO4*mu;R1^rcmD!A&?i{Z7 zTw51++m!blJ11!vq43I|STOB%xVR^d#ZISvjfQ*n~W! z!3|NN3)gR&2v|lA8>w`$*eXRe1}g*?W!+BUmU0-$Y#H+1Ay2sJ7s@T>x4kl{K_#X< z*ZPfz%goa-S4v+^R#w(xE6!oFYhu0AF3-ip16u99Te!VAs92Bo3#hl)cP47Rz#|ZY zPb9$_5?aM$S~BpyTYoa!{j>HxE`U`)7LFtLuP*?7UjzCN$!cKTGS&f$=0ALQXP~#Z z@EN`|ru--pLd5gPtS-|&F+I55 zQH6I6i1-k^4Qu;qiFIEZsg7n!K5JLIvXvI%kNc0(a~ah;RsiA)K%TFs)VK~~fP!57 z{>NaG5%goK22Bso>`H))?n?M4ZkdJnE-A0JhpnEwxDZyguqw*8NvV@n%|fP^Al z0s<04h;#}J9U>(y4MVAjAfgD;Ej=@obfcnlcOwl03=+e@zzo05Jx9I2&-dQ@^8EKO zOgypI-fOS*4ghi#x{mvi2>$O*+`s+T=UnDk+Q2hr&W?{-_B9gvo*^#A<;!<90t1^` zB$)KF-Yhp&nsD-9_DGZ_t6b$nV>xo13O(G%s%`vg(bX1{`L=`M%Ce^u2Kcs-BK!}& zJus%{^9`x(sfF{@5aRe9UYsJME=W9PN!E_a0n5F-gQ3Fx`@pJg4OVSbN5e(QsXMr? znkdf8B>B2lC+g|#pL3}rz{piezyGIF927* zuQ;E2v7^)s`rW=9A9`}0sdIkbF*KSt%jXyDgehzzeDsVQG)Rp1G-@F8Ic&Ca@UIjY-JAZ>XEFBANk52ndYdbUQazmHjMu(T{`5o$fOB4SofqB za7@oLu9#XEw1wdbaaO{2wcgD13X{LK^#=ZD;9dRc-&I8ts)jZ*d zJ67K${pev?VubXxzu?U74r#=*`#;$Qs`!z_BAHa4=yz>m86*);FJ~~29 zQyc(8bNjBA~z^zEuX!om4@;n#Q0U?g=9_y-+NIw@YUqely)0vCRMqfJ>Ea= zbg7yKZ0C0*HfXrN5@Z`J^WiP=Ys5i;y^}Pg3DaEfcaTMJRdp*7WLv1th$0G*^#t9d z(89t(bOCNTBl5ZoZuv#un;*mW&T$`pHtnARZ_iZ%nC(Igvonoxqk2isn>Uyy+t}RV zEahcqzM#y3Cpu06k4(DX*(C--xWcH#Oyu5#lEBM5Lgb%?1XpN<4Y_qGB`O6 z_hY)$ZO^i~mj+biyWuO}P1O4tw#5`?jvp`Ht2#&6j_S>62K8Y$ul$>9gMxsx-0jq$ z+`q`1C^IMk6!72b>z-b9MO1^?3{_Y^WB?Sl3POuf%~|PO6Ld@6rf-2#A7#UDI)DVi&TH;*>x=iQD!|;Z70$99v}*(WVK>m5j<&!>@iMzEb_|(SAFifMXO0qcckGrsOhu+l z)+At56DwL)^QU_;3F-rH;aV(m=ap4cM2%mNkoK83y{2O(MGbY+<92?JkWZ7jHU8p-q5IQp}@`|une2QxQaw-grIT(D@m(Nb?Dmr>~#aa~HIK+Ww% z-ple&kQ$#(@Nd*O9jlJ3Ev=1Ln4%xI9xCZ6J~2p6PG*aVZa{(!6=u7wzfg!r(#oPNaJjCJ6aulLdj?1wB>AISJblO1)2gn}5+HOUd$ONF@`{jgLHki8V2n zc7NrT%~<9K{eMqgKomu#Soa0=-nBQ{`Dc6GCsG`ji)0j(KIEX%qE0*}6~YL4M041! z@woE_ua?a+c~OpwTXh^No$=?tS#DNs3II^9xXbKVsZq8$s0Q3TrvPk2UbFcPz1TEw zc#D&J+va?T$6oNpV1zs0MrTZJBG98n(vWz};Ba;8@#q+YfI2NAB`A441yj(L+W2mvZDdt(tudM z$z2&4X4h<$ixd=$cP_1^29l&WjCMzd7hX&_QdI} zSEfy2d?}vWL);?P6Y_LmnvJby!E*6|3ik>0;U=O2*q;(VUUj z{b0x0VWx5{PxZ_%4*v(D5dk*KSW;W_1@qsH`ycPieDtS}Uv@;=K!cMIr5Dj{j~Ew- z_5&q4OXb*G1YOP8K|S|MVg-0|W`b~m^_#(y`!hPLcCc}OcRU~~b{V8G?s~^6MKe^$N)Fjcv*7cH69wY>Dye+R zZ&L5T51?_j?M?1$6kmPL{=G-J2tDSlUf>0c=WQkJ3y#0rYy6R@!40JR9&Z*9%*4(B z9C*yVpsaxzn+KuBnG=k#k<8B;bfO+Hbw>vRO#<|=+*)4|vpi612ywoa{9~S$@+BS- z1GlC$V!DlDpE#HU8Teg0^Q7cI+-ZHAI-MjL`{;<*iKuR2VzJ#8=$MbXDV0`NWHjP;mWectUb_@U0HT~b$7%h*l zNFO+~|2pw6$AxgOGYc;N_{2^ee;K15&(8uW=tG=}l+i#HNU#5pCD4BTk9FoM8U9@A zW`?b(bByf^gaZDyb(sj&uZQyWI07vX5i^#DM6-L`H@_C1ArJX=yMF6vmm=iwnWs<3 zLW4d4i+4Z!cVhYdQ)}0vH22xW0yt$ZtI+(@J1l!{1S1dz*r{K}Uj_fV{|an>`T&~- z+^X!|{1d}J7ro#7b|0dVRe{$48??`3rt|AL30){M)MztlP#dv`h&%Hm2mcvozn<}b zJbD1e)vUw6y<3Xp#4eGNH(;`*>XK|{5&yjBzZpRaDotx{^ut_I}rp?okYh!&M_Z@U~y6S{%?57C&vXiM~z4S zP?X`%Q}#etxb4NO&r!dEe~k&eBw_bS<=0EUcl0z4)J4B$$}s#4eksK9h(S%C_~SnY z{lA6#|CuSJ_>a|rRj!2E{95g9z6NVUrt?z4ud)1lM^Z+|Wld<<=C8NIl|U~WVc1;60fUjq+b$&Q)cJ!$cz~Pj(%g=^Yivn;?fvIvV$9q#+VPBnw zt$(`qL{a=XuHJzt1J2I zO8>{v+nwV$^Cyk_^XCJ5@)}&rsF$;N(pcT8Oe0>^2W`;jziCPF<7w`K29{-mpqbpS zpB6e+0L%DV84N$=Ct*T%P288-8$QsuXT`1SNac!#}jUFA4VTi$Ni#O0fsGj z$=&i{%CJAvwKA`jHZNjD{f(i)bIQ>|ceSpayL0N86mi>RDw17GPX1EiImQQ{zgoBvYPIOnb~_las`OBR5XG9R!SGy`STIs z)!2SrpPdM$Ei2@eO*cy)W2?DE2maMnZpL{t-A0Q2||Ycqx;BA3?|~wI*FJk6Gzq?5+5_Ohwg+0C{R*l3RA}1P>|2m`=S^sx z#yyxRY@?0|dim77l{Ch4nq21;DF2kcEDL{ju5Tg5^~H004=c~|?qVmjtEZ+~~xz4on(y+*AIsxBnvndZ=#!9yZ?f6m_L^qxKk+F<1o+aWzOM z`qdR4HV1>-Z;%CFmi=q6v^R3pQuq$ewe3IIIp@fbaqy~;7MKV<;WDmLdP8^6xO#n= zQ8`dlFsrYBp_V5+KfdAGf$AAD%&9C6-NV${o-_~pby0&|SM=d@U)hpW6= zE3Bd)jXL^Hgpk>l+fPrTVV$9Pg^i- zaq0N(R5iM?Bn5!`BuuC>xP1)Wr;P{L4RW@gWqGz&DDO8LpO>B zWd_3*eVfZFc5Ji~5PSQKov<_M$HgaDXLOo37WpkpQU+ifvNl(F6(euH#M{uXJIwlM z*nS?+2UKJ7$Qza})MBL#E-E8X7uYTX2di3=(6Hj;W}Qf0b&XPE0m8)D3)%5W zp`+Gz4Y%s-htuKnopfU&EZwpe-a2_EtnCBD?{;pvNs$9Uk)MU|SJDdWaH=XO!7I%P?Iu+TG>^~K%O$P5-)Yl-9apAPa|xO38ErtI~b1p$`u&^Mj;V6@e4E~{OT`L+h#b%$l% z+GEowaAWVCJ(WJ}JW5G@lRZgB~92P3jK(=6kn zKI@bDfZ+Q+rg@Mg07p;j=X&PC4&m5hwYpdxot6~2m*O+O%-d`+oPsa5p7yU117)SC z`bklx;yYfAs=fgJ;WnR3LMzyNXDHV|A2pZ_>l?Hi28#6z^MAUVcBF78`n{HI4w1xu zS{_mN9eE*8G+u70#SYx+avhuJtIm}PNLm|JXL2xj>}pGJX%{^i?kP-7OsD?2_Wb)8 z{Xh4ckLP>;OseI`vZ9q_tLDvjOTc5-4Li+CE9A#~FyN{NW7|{kU+Jlir&e!~1ijg= zdV#&my=hU74SIDiV~V1PqZNN-e~k3RHwS>#q~#_z*8A%ol$JdB!^}}Y74HWe8{y5kFo?AXW#q-TqDAO=!D3aBahbi53 zX9DgMuz#>YYy}(Wk%$yhO*oY=tI_6;%b`-5K|%LdT6iP%RF_&3 zWTAJeyGxx3J$|(?Of}-NyHs{(g58y9;E_l(uzSigl10@XJI>$xnu~R*JZUb#wDJKi zVCJ)AB$MHl)^uFq5#3bkj46h0X>TK-r6IPkC6L@Ac4q~Rtv0xaLr;C&-wtS3wDbkM zAyDW_w$Q=@zU@|4WwUAuyN{D@Xj}5?^_8J2)wt$^&Wdr1#k0gJ6>ep4_PJnhQs~MV zoH)UID=G=tKWa`S-sR6$X9Ze&E)0>~D#qm_R?8wH z-m_7M?&#Xg`!k`y-2jGXdbdrZ7&%X$nmtt2pX^~$B0>K7gOuq8_n2oi&}|GUXc4Sb z7jD+do;#g`Nb~DVSHV@5;gSF-z zYGo_TWMx?o2|_1VaG1nMHs^GTb;9BHbJ5tG!TokNGrZct#N(^P1;}k5;IMiDV;94* zQ0LK}JHe2*z0&R1Gc~h*A)t`j4&r1b>@;;7+XD>5j;#T=V4eFu4Mn?ihXdo1-l(V~ z3D>TPC18(K>Tc>;-am)Kes6kxWYN1tS!}bdQ&JgEBy$D!Me9CC5X>hU<4Ntb^x2y} zX*kPA5;nGf;Ji`SS~o>|!(q{oo2sO64COJ+;p*@ zXMb>761hV7gYInuSQ8KiiMY+iSbpb={pp09^ z&ZW1u-zyw$U{3Ycp`=Sjt|i>IyI$k55fe=Bt8pb@*83Ju`hnz5zghepn^61ldKtB( zJ2pC|RlXlCbL@q{K5mCbA;0ZgI*p{IxUKiGrzeuP>>eF%hlgDiF(_Z(Ga<{hJpo%D zuE?p~ZEIh=o(>~5(XiavRR;VAi5DK8rBo?TjTRz` z(|xV(SU!ZGD}%Y-^}MciLj|BZkY3$>uCTD-uzY2tU;A5Yg5W!nWlAn?F^^a~-2z3# z0Ni5YJHe#FTN*ExN$J7?w7_a7_n?LALk=T1i6u;|V#@q;US8(h3h_OdB#YzRGHdyuN5wG9u{acaOI=3&d9D7KR?Rgd+2mgQWlNYQI|JHJ8$ z<$xar6CQ4m)M~@ErIjE$M%i5nOgZYwWjeyCV+FGP!rBTqk&uAJgYEgme6ckfY>N!J zW$cu(Si)$v8zz|eO5&zxLOl-vEx8NpHv1Yac9~k7{gQV^lqB=0r)TeVtnUlo()~mA zm{w*=6BjbYcgzv9DDum%O?~+8jd0)at0>2=IyJNhBLXCmAH#gSYE=}yv+kcvouT*i zu`>v!F(_ZP(hWgPBj53NS#j^;;73>Id~04;Omowb0UghmII>dC0iol?7GtYREU^52 zd~FZR@9?b`xDU&Jfzl8IAL}*qX^z82O zYl#(BFbP-@X+_-i8v4fNQ_#Kt0J_N?NaP`M)^Ba{fb>IQHw!9(&9FMNL@{nol1_x9 zQBL)?1BWhGA9KvXp-Nwn{VpnMwWPqvpV& zhu*7$a)3gUwYCy6#k)p17*c;Gwm^?5wQ!mOm9#vapVKbMT!zr~D7ZmagCTY|^$>de z!lN$vm4Nk8WI^4Tc<7EK@rxbIq4aPSsh#gu82S9chG$}Hj8nsfS(?{v5S3}Iv9nmd zaqho!gED1_-`Q83yZY4cfV^$wWxq-Q3bABvtCMfhgdsK!^dTeKye6NLmb4hu)YX-$ z#j8Po$KE~;tVWWFZ*Q+8r*n5<=Em;;wjTAXg0!e9dzGN9lzam%NQc{(9pCYaL}v4* zu!4;5Dk!NzrXb~!_IkW+hPQbYq>(RuS~IEP0iOUX6ro-0g!O3di(|;wIuaQzu}<}l zTTA3NmGcT7lL>ytV@$OF{aZ0J!?EFO=i%qC^aA64gsLKws5$YYwE>aI^a;?+!i_ucKvxlC~{FS8r9sI0ng2$3I1QUAptWlR9OXK$%V z>6@N!*%C&XYDGMcZq)wSOoTQkb{Z!~zSX9pj0s&gYZo@O+)WCys^~`EYA)dz;Ik$) z7dvlPSeZNyj)~-TLTgO+4Tg!jt=lb1;vVtY^v;)QIP8RVH;m+dR&4Jq*6EzS{r;Nu z0%;*9Jbvn#rnl=-Bv|&yH-jds>dEjsM2<^>3MJ-C^zr<7KkIFcIZ?eC|0mPb&)M$R zhOCVrndb%{CDYOfr3K@4qNqzGwci(QB)^gg3qGghpe+p#7ng=Fwr%lQqOPeS5VC$} z!%1AHMaBDzz0KJTMzn&y8U&y^L`+Uf54qsundEg=`U;{ml@`2T%H6{i3=cz};z(#N zSRJ|T_tx)vkUQc0sR`R-H6Zi)rR_C2c*z96d5Ssmd z9wkVWk6pv{jQN39fn9MYrh3HZB;AFt?pGCq&-;;zYFAiI##()-NQ)wSf}s^PARIPs zyO24JD@O2Fk)S38?`5Tqx>B{o6jsD9jtNxCASsakxP6#@ty^?FRlHA|FD3oKY&*A= zpVRPrlLx1>+`hu+5hqH-N+0Re<7ISDbKpmk8(8()cMOBF8?f&MS1m}6B;!{F@B6;zf%dKXg*;YP@-xDk3G5i-WX}6S^2zDJm7+%9_ZOQF)0a$TevnzSK9RKA)a-n` zQ@fHZ&0c3?t%m>fqnV(jW17A_AUY!ydR-bW|98x3R!4F_ol3OV-~ z{Plt|L*r0-p60ifO9oSwU+Fk>R`oh@I_&^c`|2I@i+F3RF7%*Ud-xrq4OWF zcYl=$z|1Psujv;*?)}w$lOh3i;OamW8A;ViSxWEnVvO4N94Z`MHC4wi7j=144vBc}S?gCH3J4VD8knpDO&l+THv2k}Htpwg zN;az|C*?<)bCJ4``{kv}73J5&-R_mIPf*@?Ea3Q@$Yp{|T!iz~N;2w??#BHsC7heh zJ%({sM0&WTZjBrz+;~nOq|`gcsohg2gU6jhd+{7d4uzGCOUHm%(!_>x?0s}=-xjW& zH_mTw+x%;eo#klTV7AT7-P4)I?b(F9#gfcA&ty($Se^^$CSRv~OZf8O4CU0t4wN8zHN<7}f zE~IiGyObR%MR77h0r>oDlK!9V?*>QU+&AnSEae-LGVL~`J#xS3^P%z(i2k~63Rmt` zNdwaLX{LKEZKJHFb6p3Yik?Dbpm}|(`T#ahE5gubhD+6gx{A31#XO5vrCv(^owP<1 z^I)=~MmE?~I4a?);Lx^DVGd>k6wiV>+6{M)ycFsQ9*KV3D-NEk143VLU6Pe6821G+ zy6;?RpR<9xv!FC@yAJQhaLK#*7J3g#{z;y^Pr7$lIk3SYptR!Bxjg5RxWcFofO0D5 zglr-+BZbx9E=xMf?fRRI<{NAWap~{E;|C7dfK)7eB=e-wG;$ib0eyhU#?cMPd?iwE zebqfIFjs9bMd^-~IPeXmQ|NcfZwnC1Oyn~^Q&SyR==%Pozvyz#S*Q29kO1ryp5LrZ z1%g1o4{tR@M5fTEuhI(v0jU~I-D!Hdg{Rj>J9qbgHK~8prvFb}0sJYIcJ9v6xPN!S z`psC_luUz>wxnx%aJExD7jn2jn72zc=0>~<{d#<1tZ{?C%#;mnp1f>uCa^cNXv$~6 zRV4vm`}ieb9&hT|2-+s@SMncd;Y9~C>tU5lwPxSGpLhB8O?^hNB(^(QWQsm~tp!I) zo8{cUXoX<`>oP?9^mV7kkO*^&t(ufATbOs_8H~l3qEAx^ZKc#D7A-v&tl|sGGm$)B z7b$?}d|5qu+jH7JMx{WQV*2%`w1|wRwoG=dP8*o#)+zW_UXE|db}bM*k%RPrTda)I`}duorHh&%d3CG z!t+ei8hYzyWloNgE!v1&5zCYEM;$r?I@74-ZXQC!ar1mHCKni^kgo7HcN(}0y_-6r z274;TmW9-<%Qyjjd6BtTDC0ApDod-vfA{2GKg%16)%Nl!`9vN_=eW`i(d!f>Q*;WC zQz25`13kt0F1t%;duLOPTC&4rn(le7Nmk_YhIiM@9z%dq>Qkn5Ob!As7F;KJ?XB~> z>_q+O6n5U%G(*ljLUU%Pv*h0R3=`^eIVCYGL%rzKviI+qSX*XN1Nh!yP};FHP;uVR zMI16KbFl3)IZd6@9Z7-P#7!@f9+2h7;=ie_Y9zYrorVjd+JL?$r1$0nP@UqW^yP0z`S^L#0#P8w7_vx91)zwK zvVRTfKfMc|7K02F8p+vJLwt|-zaQ!o(uSWM zacTx8QN6n?h7J9Cp_J^8)K?>zQ1pWLCXqX7g_Gx*&IavSw?6}0r)o@M$5uJVW&L}T z>Jxz=UK`aM#IUzJ1uVNB_zO?2iXtsVHe|XSJNW-G9jtwp|V%Jp)N1ea>$*-eYJ&GG{ zqJh&iqLa&8k4@9Q3=h!04wvTOTG*hV1FkVtyh;(XSsgsYm%a2P468a;(a6C(rn6+% ziepz0cYQqksZgK9C%bMfhV^0J9Rs%{9uD;m zX2x(6?hN3LmvlO)s%;s9BP+PBrW}5DO*2^3(Nq8}M{e5O=KEJ^Fa21_yfjTrYM{!G zxV4bu*_VZ$OxfmJo=!n&6Mhvs#W9Bc^<(<&lc}Hv?9v_DYh!(660jh`j=iVv6TrZ0 za@}so&n7z=YL|D$nKCLNPf5V=JtdMC*AM$>8jf9=lCCSe8B8CIN7W2HqP*QF^ZwcC zOh3*I%9qEaW5fC>t1-+voa3oQw__7Xdg$8X?(`ifGbKfP$&qVCSD|aV^9lp@E(y7u zxt<{GtxB);^cLh!;6u?`etV5VjZgHGV7*(ioS0up4`i#R8BQ#`-B*IAsm9Q8flWE3 z(@flVoIezL$9A+JaWp&!rg22^=`8Z(^!f;Uzg3suyKdoJ2aZ4V+dt!y?yN&b%j{BZ zw$%EEvn*bwn+cM8>SO(7j`Qej<5)6o)qN)?P7#@Rs?#w2Hu2qDn?;guZuSRn#3Lo0 zatARY%R@dPqO40D{GJ+!>2DFhD+y+HWRk26{b2vT+*uk5oXIAJJ9z)v1L}teRonKX z%4j>p=|VV6^4_YXV&2E|fW1Yj{jOZdE4hmoCV-^vFc zMorBrRs3r6mYf`|&ir&<{n<7C4`t+Ye>^2${Y60i!!=mZ!X9i2eNQ}~uD-%v6V~J)UvRUaeF(z{6k2@WYR!>O7AEgbDUtE`b zJmRnArI#4?2JCwY+r0HR#>>5@-0JL_b$ZEP-3h+Ls@^uIuJgEye=fB^??$-z#jVbI z;Q<-e{Vgh?b_PwSS5uW{4!M|}Fkm&(@?qNlomP&TVp|kDv+`oc*2neWt9AL0SBJQJ zpI(C;r@{Wq=mB8J`Mu}5Vj|0Hz(xJSg=J-PUzrYAh=Cz%>Ctsqld3F7?EG&d zm%?Ks0NOsm8>fm9y_Ovuaw_nKOVSYnf0T20kyR}n7xj6qeP2eh$93O>A^A*~(D1;? zP}~vhJL_6zpbPBLWO<-(m2(O8Fi5X@gkD}^6?%u$bDXNZfC#v-*@JGNJ5W_aqO@)L zvjvyL{i@$O;mdbB|0Yp+9wYgBtOu$4dX*=yXY|;8IZXQgOtXto%M0(xi_HP9R|owP z2nSIFhf2Va-zR&k&&!`&oI>Y!SwLy;`3}vk&nipMPl*M|2IK8J3PzwD+#pHDFb}_a zFU9zec$9IOER#yA`yzwu7$SQ4f`I69fuE|QhS~`pY)W}=_#e1 zc~@GDz@x0#U@DViHettT&K9BNt4m&K+CzefLooRoH~XOHQV1EY#rC33miMZbYQxz6!3E>Kqw z?@5pl33TpaRPl;VF>qUE87}aupA&6V?CJ?W@a#ktZNprTrW_{;b-=P_&=!Rekl(du zdHjfbhm~z%D!iw3z z9JN1A6NhEU<-Dl6R+-W%vv#_Y2LD{mMWM^Izw+CS)M4d;DE^E1Jbk=A+^{5+I=ss# zPJm)8%|sQhvu|UE_5B5B`R^ND3W^iTx&;x@ss4_*bkmFC{%vmCnv+yCLlJG=QNr72 zNokqn1MtSRkenjhPO4&y(VXjVn4u_bps6nbht?It%pMO>w9g)#0{T*S0t;_Dnr|>y zKhxBjO6CDtu0rFHKl_3!04MhDVaa_U2u9_$hWgZ ziR!?BesVh;;nEtYJ!%1&_hk!4=xwXz8ZV>7VtHUD2DfYFVkK;8%Od8|F+V0gQ{ znqt-!-^9(ok-==0BA_AQK6DMUahQT)hP@}8*l?)yUw;8PfU_#K(pGTtiv(&$bH>C{ z(F}YT7m-gI)W+>gKI%G$_+ArSh)QUF}Ju$9>K@G|`(8IWC{|cEP zbMa|EkT4kUBbCGd+$}iV?0elz!oY~lQnM+{k>BxbNx=sP#|;-i1=x*l1%!AL)DB%Y zO>L&XYE`l49Yjz@5w-W|i=ezQyJ^;(I=3d{M78h|=$?nn zN$(l3hOau*2Etc|om#{k0z2oK5YIfCg`5mueL*Bxc7NwjN4nL4ZiCNDNojb?6ho3U zn^ShLbLXATY;~HIvd~!fqKG=v()GS^!qV3$=f;~hnj2UHJIN`Dm@? z7o!Xg=5LCrPZ~}U;JT8GEn*<2j{zoQaWcfKu&#U==3M@WvSn2h_Yg<%zzU zPTTdpvwm`XA)s5F)4H3$tPc359o_*Tb~ifxO>yMcY!^jjgkVjS7>X%7C!r5W-5|YQ zK0cezaT;cD;DoL?NRtwLVRbN{JGBu1x!_ioE-6vuh1-~JvpmF+V^nA8QV65L~8i3{mMLk?OFPp1SESNPW{@>_RZPA7)= zV57;G{45qJd9*cjxV9&fl3vquZfA-J*T;yB6SiIvDM5+E*aX)nUusrq%9a>d1ZoZ! zn#KxG*>ad7gIWDZt8pWdUx^24^Wi)T?IHXB+Pr%nqvFJtdWop?T_>(H+4&p!zMzWg z^K-#s(}+^$yJMTt1qXX;r)R>RQ>vV^9#r5?fp|W(-^Z!R_I?+dEg)Q^l7^>qunn?2 zlZHpkO<|C>h1+|?xxz_qw&N2WE@wc^en--~zJ7_BtZNr0p8ZTSHDEIf@asOWp}#Jz zDuP-?l_%ZePL^`4WnU?F{%y4yo*KlPK?z!U-|It80Kwk0;(TGkJ6ADI-NJWSsDpBP zB1$gCbwlq$y#=G@;XSA3aODQ+6sn--?%XSw)_^w615ddE_Ro6|Nu2-op<9CN(Vo2IP?%{gAn z8^TZect!v!)yRQoxJsg+ZvA+De+POD9z748p2mDrUuFec`-nQp=TuRlZ|s^F1RQ9i z4}vJuws(ct>l^}J)~LaI4;DUpKT)PG*Nl|AVB;<42I#3p2ymEao=+LmrJS`m^3=g zC{onjwx5MxntgJkVA7}Ub2wF^!=uI*q^rswfIeLj;!L^yzLGl{;`D+SipcEn3y8U? z9Me&{@V8#Yagxg3t?v$MdG9kSC9s53S#At|m~lVf3nKKU<6$jAhYrZ2$uej&+9QY| zV5tAo?o1$TOcf6xU<`Mi1`k!)3kb&JG@psjEqlZg>bM-lymx7J?tD!R)dW=@R&ST3 zVh;w?_6s5$t73vSKZ1($6#Q`*{j4Y(pfiqJ3?-e;=2edkxNG7#@Gq))N#AF*t(m=WxG;3DhkaDM+SlK!D#L)*i#x3=>nk}LIOYwK{KH3z&*a6rsV25Y?Rkh|yln4nE> zL1KKYh_W2@M6&?e!Eedi^-NZhU(c!t@=8^Ox%bMxEV1Mb05>7;oyX1vC`?Od-0g6U zktYA}G5kE9uE0jQ8@pI^S3`%9B~Pdxr1#&o60fLsIPaqpw#VFGAP&#tpHoJOy5qut zpx5NEW2dk2Q_=PH2AgGSf_79W92n~w2@v8|=G+qGcl>+b+W`@Tne?eJm8%dvwCaT| zm>}`z{R^OKVctL&OOO$N+ER^@l@mnjTwL;-@6=6=Wo6(s$KDj(Fq?)qPPb?a1jn8H7O_p_%H6Ioo$;0() z>fo5qb1p`M{ThJb2#bthZQGvnU!PSs!JqTFRn6dF9zI% zz%YXSqM4;G`wz8}TRST8jiYjsZ%bvMkBjL~xSq`#t8n0UFsOK3;27>htdy^}4?l0a zB3BLTt*zQnesgVf-93p(+8aW?y23c*BakCP6~9>o9*cR%vU7?UwnE$hkvA$oCu|y< z0zr!xc`oNPYG#ckCx$!TcI;3%Haf5=4B*nQRuOgE%G=(QXJ|8LW+;md5}qZgq=Ou0 z3@=6iX9-TPmB$n5;Y;m)*hyS=;nxhL+yDKSMeYsZ+iZGOS7y7Dnqv2Vx1~crQ(KG zMC<;7WU+L}kBX^;HY~gLw(T-JZnbo*=zg8lEK|4gnSa%Gd`jc=*YY%G3!cv!O^@#Z zyik^VFnTw=SLOggWiYYt>0&3ngg!YxvA=D_pDqpmt2v&sp$630-gUHh1*7fT^{I7(4)QVkP{5h7gmT1B7xpB&@zJxvl1>1=Fgb#4K9`o zgH#aS!R3+s)a6)*Wz2N)tqE&P_C9fJ3zexO->O)kCe91&-M)5&D=pB~B|Fvy=YAg+ z@i81OoV+y`i1Pp#dycPGPZ=bqY^dx{M41KkrC*G~J?eGA98~u1WI86JG)t6ZQ|pRx z>9ltOg$d97KNBhKPncaLo&_#ebi-kiR`=^ncWA{m{SWZ!8CNIXXU(@Kijmykc$Ln? zmQ^v4)hxALV|JP*2Ov(y%QyHNDgd-0a{f(h>AfY0eItuIgXy=baA9w}EaqZbb1eg# z>+gUJLS&v2WaL{vm@=ANZjNd zE{sSbhul2M4ZiLk%?`P3bC|ueyy*RL_2bh(&3r3kfd=#7@;-#Dt3W}3_|Yt9vBNl- zK%M>OU3LTqkvDD9)BJb@Y)}9Z8cm-BN`!7>m1g-BgPr1@8mRtm9p(0+_0Hn7G$W3i zf`*`1`7pZDeD3JO+`5hH~N8 ztSWk!)Kldlhi(rFZz@K9F}X6OpmI9PB%lOtKepz5>v3cCU?gX3&a>DON!*`}v&;}D z7Y~MaC3JItwQwd#Pn?H<`RBq{D>k5@CQLmAbFw;ilc zbRL>X@~;y`ZrqNS$uC59sl@Ti6dWdMuT-<-q{}V z#LK^lS%;YAS+GYQqmI{OO)ACUocqhGEJ29db9qB^j!P*iw;jsYCd!68{J%f7&(5qU z*ju%`AoY5vP?T9Y^uzNZ|JZqQxm@0xyDL`ci*?sSJPs4yV6$u(iLU4NvJ zU9;4I_PCX?KrGr*WO3N#HGwadH2ZsIr&T@^Im-f&I7!yB|L*EktAubDkBAJi`tx|t zpOCe$Akt=mn(DCMEyUkJ#$~b_?^|t^ETTSzrW`tNRQ<5be!5DTpw(cVpHRmR%xTK5 zycA$t;CwWHNOw)xB`s6QyRfcau=wvHhn42pvp!e`N6_*^{9#K>dC#4!*PGn1*taat zV0}LkdLZPa{Qd;FnCI6!ztd0n^Lo)jVm`l5*WG?9Jl7ll7mI{aoB8p4u`e*V2hMH9 zVvkY1=?iLS1sD}A=f2e3TyfDvzbtk!MMYWlN*)Fcid^KAtzO^rY}0d?JEnU_7Fw8Rfc0eN7Tl=I_#fq9X~TTOq<9d zB`xi_*c~sX4#Uzth~XtzJdI*QdmrqRVq#W(_lXntt2GC}Ca<%yf}xA(9(RK9yRKyM zKUu9cfXJCiOnsdU;j>7^0*Mj^J!))FYHN3eN`pM7BOao~Z`HAEPj4$;EDc}2(*<DmoA_;DwJlwyi}z{&T3(7p%#6ox`po2 zC52P-U5IjlS!j@#IZWfC!|Ps&8hx>chU}>n4FjL(aj2Q`Dj>=^P>AYY4`zzcI3TsxQU# z(oeFh8hRARhF_g93@Ydw>xbP$fNFzmE;NVUWxY+XWD*7CRK$-B-d&TynorDA7pp`4 z`xpEibLc>z`48ObnLQiogwH5kztHZfp)3(+?CEb3*{i z^1;ZX165w$+)!vVbOCWp`B0~%=n1T_)fQ#E1v}L4{dB|b@!71ym4mpK{R_c(`kLci z8mPhsS7#3uOy6YiEACip(tXs93Ov}(Rt4%PdfnLUdv9u{O8MnWG(yWSVZ#GUFT9J+D?G{gPc}0p*JLTP|<${6tI6 z%iqS*{3q|`OFEDF6x$6;_`E}>4FGFR?J|z=ATk%sMz3sNPR+Dqq5K*wj-EiXdJo9OXA?pR%vSV#9($*_j8!8SNxIgU0?Vd01cJ;P>7##(zM6PbG|H}Gt z_!9hlvkxzAX=a?cUhPc)B$WJ(2U?>_ypMHLwpe#Ar9RXAt1Gx?er53BD-ejzv~CdN zb@SGmluxboV<=NA5@5>-jJ_mCQhHSIejLzVGaK}w0R^U=6VQKE$=ZGDQ+ zfd0UGR(zg+>Q2vxb+Ii4Lao)rK2tT?3>Zni!%NcNPDGEFWQf0o)VK}g`7f84%TT<0w9-Gf!eHIG@JeQ*BiGuO2AL zTp@u)`4pGd5{j3mr5nd$x33=zainSHDMFgANSg`Xf2X#>Ajqd6!v@zN}11mOE5mb z6Q`(*_@&B|kJOiYY5M{m!3p-Vt#naZk<6+$8t{Z*qVssJagViPY{qyP&BSm4SvJ+O zMjOy3x9G0NXD2&q89?LNzdx2399C{~35P{A-uI#X#PLU5_k}+Oq*JdmHI16xsPm_W z=5sb6X5HnWPsTo&=b+kCcX6;UbWtv6w|4|}4IDBa5DwETj^;ke6z?N9)*Nd0d?5gJ z#V(MKgu1UoTFTMAS^0Q*cF(jt({8m$CLbXmF+v#Eo)`T8qwK2#qR^VYuSlzaC?Fyw zpoD;efPkcmgh;bA1|f~~E+rTsARW@(u}Da>DoRU-#L}=d(y~i@XYX?_%lkg}y&wK} zT->uK=FFKhGr!;D4HXXDbkyv!E1-)20lUEc_=|jQ)%0#Hscnx}C+@MvR~C@d3tt}1 zwtuEyj9%fU*Dy}F`<8M*-*zMjI9F+(MEA0bgCxYnx+-#+>6p%;?S=f zNgVahbL&a2;ZGNFDeasK@1naPV&{Ene}_iQoN&)o@bOnBc)x-JvTo36&PqB4Eq)t1 z+(E&7J(ux#0Tp6AG;KsQV{26Ys5@hdD7qrei8~g+I;Mzk7%I&+tElyyN({=Pq1ENR z?Ig&lm-V7LEy2Dijw99HshUB_7WkLdef(NQ+Hx;nd8vvefgNCM6{s3QIA>OlA;N^{ z5LOv9tr$tMuaM^D<`5PMg>+7V))F#)fhlkamQD6AH5Oh8D{&AP)>YKP;timBxvKP=%_{ z=1kD-3j5ctE+E(PJ5O#NqT{2*92DwB_H%=tuX!Y)?2K5BG`pq%tBq8)l7u0ze^dc! zX%5wox7Z-?uQsHsaaYek^44TraxT#(skMznFg)r{fQ0GS9~{hIq(QdTpGo>+yy?WC z=)urDr>L05g(lMAS?IFVHbT)qW390cx^!>U8p*>yl4zq;b-&2uyAi{TgvahHZKDd& zq1XM+NX@9eWii`weS*e{TKnp`Sl%J#TPQcrddC;{iJ-f*4!<%0$phPW_vF!ssiS`U zTCd|h#GG!oDq=$ioxSkkx_eZo7dkz&^h=v-^pyJHx^=Y$AX%qoxBeYn&Xb-W%E_EH zt-|4+IRV%d{jI#e({#cV$LTBRFkMfrLq=zQZtY zoJR77V9l8zP}k;rPYsh?XfR6ifIr>6tk#J@(gK+ikr?@A(>zAb^Q9^_Dr3@`U~2UR zwP`8fJ1^F%?XE)fb+Bx0FO<D^J!SX5)Rn8oCBssXC~+|}Mw zF$bEjAM^OPnH#mems+mkEv+)EPJC!9#o=CmnbOTqR@V@6=K*1Qt=2*MO|z}Qfz4g= z(J0>9A58ov-vzsUJcd&l9>`xz(G8N&kgEBxo7R(~tv^^|%^Z>I!k`5RlwaJ0)g~kR zXgxS}GfHNyFc-HL2e=&PT^atcf0J{WE}RqiD_J$jk#6}T&+>dy?wtbEgZ6CDwW-=l ziaFpmyOg=CjtSH^nk%AGEmxL0w|McU%X?w+wy__;*fs77&rS(aVpM7$ai6Y(GRi3{ z^JpWs_-mqx&tR=uNI1-9Mm^EJ%jT|<6UJ&A;I|`PR}{yKxljGabK~2NMLOEH2AcR(?eXTHG?!0Te_d;i5IM$d#`;Ocnq=;(VNLiC&r?Y~~n zJ=!>UCic%X^2=O>_B+iQX2{RQHv0N$#>GO5f^8wdJ02;#P93eyJ?e2s{yf)&o@KqG zmU(B9hSMfnu2;dn%6@i{YB9`M4paq!EtXAoc{V=l?%64LZ27+{@>n+!U7P~pc+99Y{bcIA7P3xGqr!1KlnM*4Wu~5cp`MV$1YPf? zVZDQPV2+%7%LsgV+arb)Ij|lzrSQ-|c2d1(eoGBuZopbzRfmk8U4Tbk~REC_gmU%$+K=3n(Rb*XLG^ zj71DyoIfKZa$6Iw>By1&1f%AA zNjo1Z>47w6{UEL>E-E)Wgz;7T$bdCHWDIk`Q`2WkkfMwHX*O9aW$5i}*F&q+S(E(iXqPQbl^8ucFP8+I&G{Vf6~!vBuXS*U})cgaNA8Iyem0n#>)G~VE)?W_@o*{-lcJJ5Nj<~wVpR#aYP%}KqJ>*zicB4?K(CB;UfX?D3 zKj-#!)9VZGfKND6UBi*^{~wnQg@7JTvrrw)U(qpq5`{egHR`0&KTxJf-+5m zOO6p~*te@0(rI*gX^U^W@v48MEb3AF$>4k)!KlEcp_nTn!1Qa@I@ohoGw-!}7I@F5 zmG+a}ilNf|k15z=J#r9g+>(ChetR9@0=A5RoiV)b&d8UAxCt8mLqLUeRu3Wwg4!D2y>o97=yX&#$E~Q!NN>`vFt2ecYh@`(LgHl%YF-9Y`1{WO=-E8 zspO1cw792qj{a^1C_XKl;>tQ*)1)Yn>1(ZxupDb0K=jaQ7rcMev$vhOX>ktH3qxD+ z0JwT&UNpZ?&J5*<%S75DqK z{SdXpb;s9S-9(Q1tV)qR;f!SlSH|trFX&x&oVhgc@oCgRUGP5g`;Px-WcH|!r=ssl zCF#fE%mNzvEd`Kl@nD%Rscfm;{kmX5^lZ5y;F=^5&deESlZ(+RGHxmECn&XYZr~A6v5KQ1vty zbm`p{lUlMbvs>h1zg}1l7YeEmK@4rK>DHnLTl=Zf9gI1)lNie_&`t_FPxc6lXI~ER zrwywq#jG~srvqE!@~GubSKdJjFGBXa$eG@u%HBF(^V!j|sA_bxg30h6OA#;x;Bep8Ud)4*JlWFi5^aMo#h;ZBEalhS6@GR?rq-67gx{KzE5`P zmQjx5oGs+93){x^lBN^8=q*aqU}%e8$aL z&SvBwWjEYhU?FBMDXF8uOvvc3^!6%A!11#y{{N7_ z*E3$K0!VKNz}^;;gW3+>MNbAUyAFzV~@-AFF@HDog^3i&ld6nxc`60q7Is!Se@fX`=OaPHR$TNMSI5xb4NVjPBvRg+S zq7A*ive;T=_aM$?Q%zw*7=Ms%Ys1RfVCMK_!FedH zQvt3W#%Hr^39YuIb=3)qy9HHc@GSBcgWG@tALv}(q3*M^Z?luam+JtPLb_ZrHNdV% zUa#s&f%4A{XS=_*()te@prjkN((!2fSp5F|E7$Ky&Id9+1qf^je20iQaOKMv88sQV zcGTLni*0`m3hY+Ef1>iJX{~o{2YtrQFb89#JX(+=0t?j|7OgrK#h2In^WgWwD$R!^ z7jRp+ zzqZ$mq0aLT(VHVgP8V8qk}adVx+daHRF|`_=4)@Elzf63i#5b-rk1FOgiQ9%Qzy-*Bu|IL~{MYXagl=h5h1FAgyPCHvvVfS&W+al2*6u%7J+z|Hs8L^GBM zj)h7*f-#c(1gb8t*_l)G9z3At#k97BW_{-jmPp)K>=H4jds|oP*qaVAoSQ~b2a&Fn zS0gR8i)CYOj&P}bdn}Y%z3|E|ZeN-T{(6_d!GV}q zh)x~-7ysyy0Ui2GgF8Sd8Y(;qvO*_~t$R#+;cANW{RPyUXFeQOO!&h~3tNN6qG$29 z%eeT4f~_S)DcN)W8kw2Muv}QV+M|?As>?KL=QTvqz?}C`qQ<3$n`1aJ73ZxL#jj zaWTmW&_KF>126pd^?-l*GWgpT#~DeojY z+;_Y3(KEMx zQWG79tz>%YW(}mDX6ege1%VmlW+~oeDyvwBSII@&Ii1~7$dT|c7DB-yitU z)R_cF41#yq*1yJvaDvE{(hij)$!}FpIC+&gL+;F7+!SkDIf}bJz zNUR$urR#7?*q@wGe=Z^R0-z2j&4ZsiVHZ04+f?33z`}e^dsIB$NCL_d$|mkj#NEdm zQTK2Xcf6LQC8>)69z4;y=jn4%a0GNcw+<^jF;R@+d7TeWiBW%Hc+G+Qcsl+sRpRxE zlj4Ff2d6w+&qC0o%;9SPRAN&0R(+MWD{5GRQv&~@e{jF*ReqNEl_`BTJCzVBp$4aV zzMk`vB$0oM-J9*C7WUh@A(0?(#{WY)FN(@Oa>%ao5bKlr}l51vT+a6 zjzca03Q49uV8ebB5f;VCyRQkq60;KkL~Jj{nnfgcENo=?Iw8~<^T;iUHe+agu~Gtv^dc1Zt9A^Q`1CGN8Q z@ju3Yj1BmD4tp_7-R)F76j^X+z*vQr7peZg1NmQv&=-4``ucU^NY$C7z$GcKoEiJ_ zh^!o*?Em%kqyO}Bxdv`OhQ!;IM|8_N_Ffyg^>N$wXC_OL<@80@GM}Cj28On%2DcfI zg}sxfHmSYd^qU2LDSR10zY`76lqj5#J;$0bfn!e}m9=QP^KfY*No045{egRp*o(t6 z?b+|uHIW>~N+rf#k{l;Dcv@f&|46(fQAHj7BHXnbKV$gv1KfpujICp$Z~^1STZ!v9 z@?`r1Yb{A$Wuerk6*nomgr1XS?K@r549It6qqexEc^{Ge?xpk!?CFbJXB_YjrGZHq zC@6yYXp$;JA}p&wvp^?xsNHgYF{#2dB$)%e#(ZhL09}h*7pvS2q z4T>d11>dixz4vYMi^M^r*jq}4`5uncBHF%V3kXviRzL1=B%Q=pr$94-Nug_6;9JJbFyV^R^i4E zE0PO#tZRr`D%MeKATNmzP<+aTv#aDeYx+k7Bp_~3B;98-{3L^DgETUL zwri;6SilkN;quizvG<@EabPe$U0%@rU+#~G|G>N5Cr9~nvaaDlV2HeG+sUBZ!k%AE z{Kg3FPTH#Vn3LUp3ttei?}4Z<q zfG$EgLM_|esx(Y1viE-lV_Dt$Q|&F6MGRIJ0C|aVN8yM1HYgLox69X7R~T=PFdBae zr`<{cY$~jV2~4gc8TTC|=lyxh8Z#~WnqHrxsafW+8ob%Hm=AH&j2$Xm<=Xj~`ppCh zF|cH!Y8D%J6s)F4=q8<4ga9Gv?WZw>fDHnMUseXeJQ|T6C^62BX83XSy`RCy z(=0fV`L~wj6h7f;`h0`(1!zbioV&!TQyoA8#vE6%D8tpY2s@yE>?XdNRVnu6MPpGt z*i_`L>Q^Gt|Jz^kiTBU69!(=ZrNW zq>DTJASYe#B4R%jq~ZTaoVn3;rIsKa3jwZwm|%S^N27pUAV)pmY}8Ld5s(%H0NYV! z8CphgR-}d04fZpHB$s<>5;NyrG;6f|i20eR=zN zJ^9+-Mot5uq4zO>hFE>Uq=O&VKC?6U=r0ziJKW;WC(=$ciVmP!JQ&@WVmIbSIC@|t zR)kX7ubk?Iaqn#}ViGJdw5aWQn&c$n_^hFKdpnf0(MxC@)A}iuTIz3&Zl=mP(iIJC zRv_Zw0q7q%OxqG7_BcDn^I&RaP6FB5rJ+ck+JHQ*nun3x2d&H2AOe;SqIIF7?3uHL zt(}H}O$B0@JL*502&co|A^>_)fi$hvyBC z7-6#^sx4ZdoUum%0Po42 z3e9U;f26+|5lv>9=H=`sqmCgpI8S#ug`J5>{@gv-N^tO}gSd01@8gk*g3QK3JApw} z{HATKaXYP{r~&sGp_P^SV*T714N9trQ44pNKtTL0MNn_^@k*`q4&Qu}sE`84yPR$v zuwQEkL>PU`VKiPz*mFwEP`oIQxj#720-LdWsp@qBdg(gBsmdpPaD-LJC1HEn@ADdB zA6eO_KYzwM-hE&@fY&Ex4GPC=ZkwxbS4im7a(eE&F)Xwt_(v?e6E^LXyL1~A%Q$8m zmVFFoH_!{bBn%k4@}jx{amc|4^7hA;z*8W|JE?(+J7SqpHx@{7oH=j)m}A@JWNl+q z=Z(~pMr}zwjm+*nQNvFRQW3V}-fo3}-*ngCrnOS<>^PtOnikyS^RVsM81tNvONrQ5 zP(85GUP=72u)1Wk*xt|Xi_v@NOLEGn{7mXHB&*em+0t?(a|YZ{Tfn8@ z%J{T8%~7}xH@rr;f20^8ul)94MqKk?CUVH#)d6Z?t`2{~YiO;^SsVbWSVRTW9vVl7 zm*77cF-(g1c^F9@B(!Q6nRD+GMz4C~ly>2>w{xD57HT_4%TUe7Uh>v6r%tV#w_Be9 z%P~ZSa>Ff$Dr<23OVx|b^y1C4kCf$kA~y;6*oIhp`$ZsSUaqkLd7MjF^%XinYd#AueQg7@4C4J%ZJZai zzLoh*&j;~48Ze&^A3hX0ADj?&`;|3Sd$|l`h&GyzwYGOm^;rO_4&!*qTd@9|<D(oyu2vst9le2EYO`OUjK2`bsZsfAV{S|O31rh%apX4SN9^1JH6vWyp3G+* zHOa=)CSr4tXl$Wd!@q4hApXKa=74U#CpGaa;Ig&Q*~bW0pb9*zU{Ib~G+GptwH3(t zP?Y1AAyL*B9+RXl|1R(8obw*-P>n@UHcX?y(kxMWqh%9t$?6I7zuTY`{3L&)+=W4F5k$o6WTIawlArT8>GL)-w8?#Z01DJbI%FU zJp&sEbhTKsVjMw8YrEgZxH{AHk(74|AGh?KbfEQ)|Kd`Oj9cq}5H5%hiZLTuZK&CEbyTuiPr- zx_gH1hZ+G~o~JlCU<{jp7%;}9dC(}Y8DUJ$>zmIOzNw|<*kqAMafb$ZOJE@NkIi(GC}=1iP^ z^^BqGr2fvnPh!nrKB0b}t8yYkUkdcV9yPkQJiN2!Lt0mH6?rgDv>etPQQ!bytJ?k6 z&t()?$snhAuHkeDJ5_y`TEnut;Ha4LNRX@NNW&JwJ~wq=x4Y${a}91}d&Ru9D#@>cS}WVy#U4vjYu@K#*Qma$^0qZX zVSl8wUuSxAzcnmKHq~MQT8gpID)UHvfI%F@^b(NXZF<%Da?f$|S-(c$1`SMXPF_i6 z=iEs_n6Mzv!#9BmV`sl~EG^Hi-XBDtR+zh56hM0r+8!KS8blW>5J@$=1T520QFrEo>g>KjEtW67zMq4XhG~`ms zd7;A8JzC}DUr?TLRbQhyM=u&es;Ry+9 zp5z4@o-}z)Hp-l54gzO`b-(PK%K0*=vQRjKqzy-F6P!1#U3=32+m?>iJc#Ub9+|N= zq5apZ^6>g3<~#wi%~#&}sgwJPG1Bp}_3S+~ZV3q-qV*`fZAu+$IdG6%@N^pQ6dXr_nWrDA6yB4Y?o?9Ug9 z$jMM*;0`%^)}Vpx8z}6W?9zLI|Ip--d;{DQm3Nm;kIZU1C(Ww12PFDO-@NnD66rBO zag{LTd563h9>p=;yAd9uo*EYTWwY2%y zb$Y5JnPqL@ zB4!unRr-Z0HO>2TKvL(Wh*CO_2tf4o17ini@8TMX7geg&tL@t8^h#Qw6H=!7XfEgG z=AE}^oVxp~TgTN^Y_b6Ge2Oc^3+oD64jZAVIJG)lG;($1Ubk*FmT^IvmXVlN5qfdD zLEb?R5rsHX=qBw=RZz>oxKETzW{y;5k1$AZ{F?raXRU?%49r}`UO_bUoT3@h#<3dT za$YmKo5S?fsxpS3W4@b++l<`K@q@_m!!c(r-00f{oHrZfZ0{}#26J-amrTMD8jg7a z^X^e|xm=jjkcbjIj=Fp<%B4}y4og#w?_IqqBDYtA&ANF(hH!Xc;q~H0u=kL~fhq_9 z&Do%5X-vPd>x(_`@!3eLGX%^jW!t6~!(3>^taB_!SEJ@u6N#2Xbgj)}?ygKh(yy>b zP^RV&IXC_w>0b0IDc`)3ty$`q(|A7J;{rxkT(A6V0iWYvhSySUG<|pxVakD(pgMED zJQX|~f72#cr4E*Xj#z&yTe;%+ZeD?Y6Z+1#ku6rxD$+(XuZZhS`px%d^s&yzH2zZY z^OC|5lSpSJu8PTzUokFu;)$LWdLtOy5bf&XZ}~+=b9@#Ndj%^Nu-$Q@w5H87OM+td z`)<{@N7Kv9y0`ZMJ9qcfnp%YX=t&LuL#w9btq`g!-m2Q;oOWcm*XkSy3pI!J{Yevh z+tqYfRR=$P7&Rdd@iaXbAjM%Sc{j+l>ql~^gr~SYsTqBEj)O6D4w+i4FMOZzE|G8L z7VR@13&O3XD^$%90u4&b)AUsG==nrt)fUguE8D}HD^r4Ub~VKlp=4`+qNP-gtY#Ne zJ+Hf%O~_Jxk#!UC6HV(}`evHmr_r0I_#KnO=otwy^z63=G!Su8=jjP)1(V9AS^)76 zyyi_%{4<~Q;fUu>yA^6O^$%|BIWn0K`+hcYMTHq8xX-JY{YWaPX_qt++(lPCh~D>` z%SmMKUfvfph!sAtnzKL(q3;D1hLCy9DJN)@%K6L?Zp|&1CZ=7?WAq%3$cr2LbxM9t zylo+~;RhT-qt=^!AHhuN%=2(Vfc{2W5HJqERM3ASKSt{%y5z;*1WYFt6)ccwSj$8yKW{j_c+AJ_A2Ij^TY1S z$mrDZt)P9%XQ{49Qs6R8-tgEJaub&+#Ix2RC-w<3t2ocAuf1r(8RZ3-5(w zBt^%z*Q~3rd2bfa>zVm*e;CEn{K8fBUe<&JgrZtb)|J-yskH%VQ!M_H0r zwmCHOd;tVqA<`4`vgRQq&Q5&x*Hn{QSN`^+;@Y<`$&IieeOT~^8WrkTX7-40JOC1U z=ok+E2)u^XV#pIIv)1(Ku-@hodj7(1q-TGDnjzkHLCATV{IbZs&EmeQQN#EfpON7_ zEp2|PRx4Kdwd^#7L8&LXe1a~i8OD_Th)COz*ALuDCR!=Qq&q5`_jJ^XD+snjBK#tE zc~_g*=o4JD3M{ln8Sbd^cbn`%`tpaKRo)&Qe{QC{5|dVWwW?sWqtouu0dqGzinMyR zsI&HpDN|Nmn0a;ol|abBrzFp39cunV)}bG_OSVcYT0QIo^Yk{uv{7X8XXphCGy*{i zKh|v2u;837CD#|ra$RB0sL4Q|O~agOciHIFsNCP)fn+EWrOfwbv%yD=H*r_Cl8H9k{>UrhB#;?hYg6uV+!5|-m9q(ZQ|p$=a8UHQc; zr1o%x7({c|zQ*VDEHfzH{?X~P=;4iZxQ&Nlgd<``p^k^mbuBW!y<*j|jd0ta7Fye{ z4*ITTk9r%PeS>Au=Uc5X%01|MIA)(1^)dl&p>8Nr;`w7TE^+A+H_5zLDY`b~<92DJ z#X@h0Hvuowb*5AJ)HRCmW;~h@HJp21RsvXRBOjM3LPU+GYVM)Sw4_dX-nS!+4y8l5 z)*#$PhT<{KA2TVpr%b^Sbf zF&J}pvlg9@Rb*4QNwgJU#?U{DqTRgnG9h%WYoThJ_8O=>N4a;Wp=34HAxvy;Z-Y^= zzQ8W6{r-VVGNwM@LU<07k{tk-paE&3tV2I)X|>!mF7bP~fzZOJjrmJe*fk$=#v4>& zd5jJlRSBxPtNHfr9!^xL9FzBD!PD~nUiubWF(8RE*G7C!oo?(*ojD)pSsa(>zFq3N zuD{7L(B z6)TCE&*an4m&|uM`&Mjc73Qc?@$;>YH};{wDIkIg{3EseXY@ylRf=I6X6`7f`Oocd z8&+p2z5thv^sk8<4u?0O*9J(OoijO;&Zhpl(<6e;zZPj^C&yXtr*FSFn_;2)w3q(l-iJH&0f{E) zz5>Sxhox|>ihAvk`Gbr=jpSn6?uI=(;2@amG|e(|B1{jwBzC0+ODP#BJwbJMG? zF;{K1h0iNuww9t!)IC`ghhn zcIh!xx7*hguYXaBY`&@>&aZd?9b>4N&d6x%m21nqr!3F-N!Xw!FTdgRLSo9iCQMXJ zyEkgXO-KEEi!aa_pDpU(vD5vEu~XamL-(Q-C317zg(VfzIN)v+@V{# zMd(aNrdo+2H* zYK-1rTi9C5N_H3P)8*>oV~Q_c2HVcp;v|LmJeD(BU7Y zcD9IR)dEXeNUm@3gr>T>c=UCc@2l6Zp_`}Dm@ATcx{M85h}O@wKqWTD#Ma<+dp?Pp z9ta&1M1+=9UT=I^3o>5^pAWyg!LBOKPqpuJTtt#v#Mv@5+qSsV#La<9E1$W(0xLR= zrYxvkENg{$%}Z5K;#DK@-9a<;rZaaihq||FAGwQHTv#uPb2BY9K|i6G8hT9MjHZPEN%tvB4mot3uHtlN!ZQcmyfx6weA>`%mzS)+w^El&cc0437qy*A3>d7u>BA*#k{`P%+2 z6gVhbZtAO5y#bb%2&4rt=NUCNk}z12=Wr^QWt%H!M}sR{xK623I^5&pJA_#+X&JW) zvyd1YO^)3POFrNeyB_Ko;N@N=_u9@l|J#dER5e1f_808+oadh7nW^jsMp74DL;Coq zuI0!YJNtDyH^=p%$hgj*H_lDNSp*Q=OMe=Cx@Wn@j5dcJB? z>&j4>A>nEN8|(WFPi|Yb;ty2V2Tiq|@oaU@3Z%QZxS zB|$@(rl#4d*(o)7rVrz8b0A&GZ&+d3cxMaRW1-tERraW+s1``fOu%V*Hda>I1|IhZ z6bg)P%A?DwwQxHnT(WY^nnQj5spo92Bne%b}F?4sQg+>@gQ#36UfTjOr`wPtSI zTghBW>a?4hRdF>$O-_8ThIi;OVU&2-OP_-){Lc)a%gM`}Il??Yt1H=@M#p$AwnVDJ4#&H{Nchkpo>h z)l6lrnF+Xl7Av%%$*vL>7xrQ(_)3MDztxW!TQvhq9rVyPjW_%@Tyi42^iz2PE7zI_WfO|59QC!TU zhl%9Wsp5WmHFD_kPWzn%E=_i>Jy+-Ael~$~baY(2TNfTJSIEs96s2vQ4Ojp{`$>(* zM=}hhSb~u+RhKDLWZrnIqL~Z7P|XyMhM&O(7uOB<*pYpZjGYPYL5vn@AA72Od0d5@s(_Z zwhkSZt-dGb`PDw91{rjWSB;#!?@KitbDm#s4b?K)+iv)B)27UAr;9I3i98VrT>Ntj zLPk9<7?Ev%50yyI?Ha6SZ;`9QT&tEWYkNo9!Bf=Mafti-^k>ZAuX*z__v$Ho?Tw~` z)}mt0XlvQ_r0`;xF7TVXYjWE|1nu{cb{_SKuIL*TA3f+XaU*Wq?%K5zM8&tC*xG<- zN_lj97J{X%{h6vn{akPi)#U&`zfdkaSIwuzO1{U9N}VI9e;9X#hLWd|TSzEOo{9@l z_P?Ievy8JGK>JjSq3&Rq%N1j9hHnLxtY4Tbf=A{ON%8Vop&3H1GGcLt;qj8Gt=d-wdJ;j7_E#RDZDS(#XTO2ET8AFwgY|D&LBYJVbEKW;qPAm^@q zsi*u+U6a3`2s+^bAYZdt3;3!&wkMmsn_rt)SK7%VEpfzNj%y2B|d^6_u=(EAQwWU^HE0B~AV|R4ZlfO{+!tzJXcIVfW+QxfQ&#X)q4EJS<0)5dr2ReY_O zVK9g+Y&g0gta3dKuJR(bB5OwW28tkAwA-eSBohnljCbsuqBa0yfI=r|WaTgbDk{nB(sw4p+r%(K{i_c*X%P_jzW6XQE5+#59R1F`y;O49?ghBL(_>uf z)}e>?AD$c(#H3!GlNqq#x^WBT$|pTp%dI5Za%64?=2=M31V@HLuypi?JCY(T39viG4ASeF`!}=^M}E$1m69A(BR+815N4 z&a)L?rdRY`Tf@{H-hYd6NF+&p>g3wbmJUL53%^9PDLkk-LqJSXLZkCby zVs>U_Po9DP(xmt9N2ot2#u~IGzQ6uSEG3s|j^p9pr%uxjS>fjs;o8fix&G?v{qjo7 zk^_LoFa5!ss);@mZzZKV2)`{Bp!sd}Qwdco)#b8PYW&RlKa0})e(<=^DV6`Gzb}Mz zx}~{5A7ivRSPEDQm-8en>WJ=U{vtZJFluGaDZeyQ3?U48QyB{WbY12JeORg;GqUK{}cF7n-jHClK1vvI>EY{wSGkC7#Egd&j%BTBiB!z9h zHp^xTy!6_Wu@2#is!jR~)o?f&2?@#CJl#t3d?F8$-*~fXI)g54sm7{$IkqDr)ua}k zVF@8md#Jq4S&=z7FP`PovnT(}pUM$VI$#XcnCsjH9EYK#46#vvQaF`fM}0D7XxCFm zqG|_Sm}2xeVqzv2k~CWzPnblIMb9-q_38Pwu`%2o)KofezYrIv zQ5J*jkY(`vC54(dLT#F&rUz_VVoDdnmU2aB8RHnL53J!9!6Q=H8WOu{9{917&H>=T zz#s9+X=TI|opw)|69|RY)sn5KGkkl~{rUG2SJ3>o3Y~?{ZLAM?$sGD14t2vnX(%RQ z{EZq}j7CJ}BgURD_uTo?-gdj zdT>+Fb$bayJ}ky#+MW<``5vjlbH~=*oprVfef9!Yb}B!}yBoOdOr!UbPG{t`y;56= zPwFp!b0lQpaW>rlxUF%c-f*^n1s%@gT6O=*WCZ*ebZEhh6>-%`cd;avIjRK|cBJxd3)U$KM@|-`4 z#`{N&n>Qe>f=h!V++`!fts~-W{JOni52)QXvU3x)rOj7PkKG^R5cs}llJ?yzo`)T&+Vsj6Su5;?y%vU_h0>6|O}v7gG4p>l=6%+CGJ z2fgM9Znl>i&l<(z#oaY$xY1g*YM-xecWe7av2$9jShH=;>si@sq1q6}U*9)@`&A+4 zCd?NiXXUIlTkERRL|=xqTw$q*4g+jWC_^_ELk~>8en?}#BQn#26>778wP-xCS+8xc zuRJH_wu_+!2jCeh79NAP*4_+a`FsG@-y_505&g4dleUL#fJ z%+kW@=Am^3E9^`mDKR@u>8ezSr11mB1 zM*wFW6d(gC`#c4KasLLHg76(eDU#qXQgalo73H@$^a_*i-kVk$@1yPBYXrqP6*TIE z9exF8`Q*sXb+M&O?-YezbzVMyMv};%j8WJ;p+1zwl#D}FON@%mFyX~^GS?mYTP}_G zaBR5)zz%)dMK?FKK-H+;cE04U%oN{vn6EFfg=0qb!qwh5)ya^;OQ9$~SvB{@e#wa* z)Ln?JNNNIS7|L(B8D0o+Y3X;;PPb|wshMvQA)5ZNdTyTMws4PWVO%)3{$Q>V)Fi>68Y~&&QMeb65Z8PkLk1 z>23ANWpF_f3M1@qEPAKe-ZJ}>uL9<{@czr$4xC}~Ps-=NKjV`JUCYSXUpmxYUcp!- zKxowWryc&(o>xHexcSb=uGf9sy0Wp6Wtk5AXq=ik;@bZwcmIDrlOzOKiDpSlvLlEg z7#O|uO3V)3b8|d3{Ip%OE?gG<|J@TVwjlWv3seLC`6QsB3)>k-r+XRx@g zK#P?ZB!WX(V7NgTJ;ffoUlaE8ILGF(S`@0-Rh+lGp@rML08pbdqI2RB4)pWidybpN zk(O*cNTctm_&{|S$&nNUFpCI9o|~C(|Hlk)Bmciz%-A_j@iJe2WA z&mVYZS3ze>*9Okv21X4eG=hMaGZS~_aKH3#Ep1h>ZIRV^ap!TJ>0kjMTyk5uv-NL1 z9NCyOI9RcbP2&~ZXUsfctY3MV>W=AR{d@hrELp%8hxE)JoRX>F5*@p03g#R@ahqM`A4xl@O~#xw~qRmO1w_m3dl zm^83Mli2&y&EWw4eY$^mJv!gG z>q<`oTu3PI8%ID)?<5O>rcAzM)uDHu(EB$fS?Vgq>1`2!Y^w|+IEkA%@fEK}B=WRj zt)3my*Yz*f|8}>RDNZwUC`u`F%8FB*K6yAtrdtHwoa}6m?Ns(M*Hw==!f+z?Z!O8| z#H?@9yeG}TGJ21n`+Nbnh~An^<=(1NEnkl)G2CDOt>yIqMARbYqF$W99e4axRPhIK!s@3q|ckZmtXPn-TB=!be;r% zK`a2|Q9IDi1sv{kFcRNen4OLC3Qe!)K|>}6zxqY(Em`$%Vhknc32v__oZcs@J{K;(S8Aeg zc-p-znJ7r5vMp_CZ{0im@K9d<{Y)@vgNvn!vJ=O=z$6NAUd?TgMEl=eTxBe5OCI9? z0XH+Fa-jcCUq!w6%>{iSnFCZwy5GBZxV}?T1s7m%&xoja-0k_Q5{y+D;T^zuOlkOAR{mkBz0^rwNh{B%h!kJn zOl-WjX*iJ<_4^9Abzc${=~~)mx%a$D(rU&R~jktKEucX zVEWqX>ICcGKX?MaO988sKBG4;*A(Dka97noxg7uNGq17p#09Pw!Igx{e5G>aN_s0< zNRaNmuy8k6kM#`>oPQ`gI7RzUEp{MSZl`f$q>z=y>UQjkAG1TFG|7oO zW1Gl#FdqA>-m1|H?VUK|z*~>$y|=1*80i}MsX82<_rE54WGntY!c0}9`1c?B`Kb z96+P+y^eo*WM&lQSW^j)@rD>~VgGBjesyYX{LKaMk4x&c$N|WZ7NJOCi@yZ{pv$p< zT;hAbORgiy6b8O%yhbqg6E~~R++bo12}iz(j{n;QIi3DWR{?=D1gzxt%kq`L8YJN9QjgYmw=BYfP{rLg#J zU*3l=?kFfqHrH6-sac$aO6c;h*Aqq$n~rH<`>;H9Rm^xtIFsRi0QFzzy9#@#450v z`{Q$gVqI2a(JI>WfP&GEN3bnUt?kldN)YJzx$~{QzWKHPWpn;Q0#J;<2j159d5JXuUheh0A73Zm|FX<{S1x+r*G<& zR?wkt56;I%`Tiee-x=0aw*7ra1ymGlh$2;#5}NehL_tA{KsOGU1AS4LjAoEfG5<$)EZv6XU^ zm|!9fMr0;vHCx_mWMk-@@8wyo-1(NbT&cS3xaSn{b$QpP;tjo|{sms6AJ!|s18>s9 zex_Oc1>O{JCjcm;+u980jcy1IkAGQC7`+MR(tR<*1nj%-r9ZF@Kxn7|U{o{-w>h~j z$ZDf1Gi=I1>Bi*0BPCXqJQg_ma&*-2Mrh+I^*cv6P0=@}8jK7M`L1||$9{_0)a_ZM>jpIr0%+Y+QjZPF=q8Z$up>o{?Aie@O z$_qLo{_81lv!r zZtEmP@visHjQl)&Q8c6|UXO1a=?d~+Os>c>B>%=IYusJB-FNk3w;S~S(qfyFPD{-S z&n@FbAxCC)8N7z(Xh-!!eyu6$O`M*(v7{_(nVpX4`qkA70QJ}B9)Me8IXMw3ur-M% z1^cX10ID`q+Hy(`K=Y*O!Z+-nl4C4YuhDADBhS|F^J-90Q;b5z%*AJ?w&p+=P+wk*{Z9Yjqj^ zl+QJh73M$n&%dR)7QFxg?}*A zJqwm|r&6{#>9NJZ8Ys_5d;J+1m*+(}?6YdUnR$_?PRWsA1X~VB;TpJ%MH=G248+f= zIuM>N;xOnf#W6S`QXdTU#z{CDaL>9d8ztL>zIk(tsIvHN55=Y12v@(wwKlCP z74A`>!XxpNjrOkMo1`10Kn^apaEAB(awkRCfx;8`s~n*b>Tmrh@NFLhgKz5QpX4;j zeYI0Ri(p_E%nG!FE|28dy8<~i5RF*9$;ru^BKX?fo#7P$fXP`#bR-$G6Gr=n+GjMU z&mY9sPL%czKgK_CjY*!MC{$P_h;#z-k%fDm4MeOiEXng$CJ%LZU zp}ss){JTrD<28VGA>pTX3%$ea2hqkro^7*oF6~>&07imqv)9l1ZpbAn=}WS#Y#ca> zmzICHeM;rrsg-SxduX(rJ~s#XC{WexSvBbbH`ipz{XL8$&Zp&BxrcvMo?%6~oB#Bw z`eAV{wzJo+@BjyCW0r^=ubD~_U!RJQ_3$oT2wEtxV~!}T!B{<-C&4?@j;Z5XW20sy zOXI!>nnWHaQV$YeT@}|jO$*}StCKSH-^S_o$Ur9Ut>bn>3v^qO3A6)#ejN|!1@HI| zZspXj;wdbFgeGSZd#p>o5Mc-D9V*ylvwFN2gJ(!9eB5b#l3ndygp9&oK9*|)t92k( z_-*n40LS_B01!Xv^1i-8Mn{hVAZ$2OsmAs3YBwJMU9W%ZYIWhbS%3A2*L7UcK`Bh} z?7Y6}+7RHzC&=#F^VL|j#_{n4?ran$LGCRAIa;EH?fR4K%nCFF8&xjB!qsJYfj~lc z5y7AL%wO!-PdP(h?t>SfzKc9Eeg5$fYm-{}O34~zM=S@v;NWIZ1R!GDhmY$q1NZ8z{XzysM#J{ay_$BI3vW7 zO-I9a{tfoYUX!ZjXv`|>12FRUi0if46@ck_7w&vYQK{Z|kSj(LDS10n{R&Q)s%|v3 z&c~wb$EqbQ4Oa|ozpkdrW2VUW0U630^$4t~6_Jlo39|(+@;YI*HPHKbAD=<&8%ZMs zarU&fmKB-}h(lVmyI8X{J*}!eFfUWGJHN2ff?~oBEQ|;OSsLCoC_PzxF-J;^;~^S> zoAN)gI@|t>`6Ng?@#2M?@^ejdcD3ldVfXm;Zt9l*j^3mWNR{ydoRxQ8e6C^_y$6`v zP8Iz`C@~uX=C*3L?a=_1zYN}F1t6wdP958F=xuk{)z+RRrl_VcI?uCqhf&ygisc#= z$9kM%4F)I!WzqHO3UQ7Z#-;PN3H4n2%Y~ay07utaUj-KlV3r9jxkmS8l3LcB#FJOg%OgK-0=zFstOP`{qrZtsE0wMCP`+F1}nZWc!SE@l;0Bo+1rvy z@8YiU9j$R(=DOW3Iv-kT&hvI83B?YT*|u;(7-b8_IYD^8hfek2406K4oOAxp_fb3OAjjr>Z!A3k!^ypMKBo1wZlvpx zCE?sXQ9#m4EnWWhR#II~Y6 zD3l#y>NVT(KJn$zlJP~W{!(WX)3ua7lv~z*UG=?2uC}E8{IrB>CEuc^soHgMk(OmNj5wbG-Z0b{t^O0mPsZ-`57Gwe3QlP>#B8CvaE(`g5OTL04$t(8p^E_=jVX`3dC?! zZB9>{5PN7)-qcq(0gg&z>}(4}p}E6JVkmGYCgOhct8Rti+6EV*lJ4_zAR>m}X-z~M zZ#ZX#U*oKD@ghe=Ns=%8gB5in16FiSuylF5o7%$B$JDN>+=PKsuTX9_)mMO(odb5} zT+Yw#iYxvl&$+U&LUq1ML2=A51F8Z>r3?*wrhm405F z{_aj_?dWhS^bjWm1{E*nICYSsyg)z!%QEcV4)Jn=(56L6Zq9wkQtwBTnPhSI#mbFE z(RQJ}0#hXknZ5YBp`GqzQB#wI3(*73CUfysB$R01;fb%SKJK;NyA4IrZ}gU}3j&Eu zAc9P@z>tt1S)2bIPrm%}?3-8sKD9HlnNv{Dn&WR|-Y*_#&cv-DpM$9vS(0dXI~dJF zd@b9&2}MiE5GL&(0EioDp7pCn?#vBYF!#ngmCA-K%zGC-i9prFBpV+jDw^&zm#;(qg`gJ5R0 z>PbO$>FZ&1y@JE^GA`OL!xEmXGzg*M066mfyhJbpNKB$x=5jY@a8mkhT5pqTlEqr7 zKI8Bl5c#5Vf5|`?p&zZNS*!Q3VDld+=W?@@88 zs->I3)ic~&KryG{Pv9x~pB$xI^;uf)O#?2YCuNliiaI`)wd}M&q>)R6N6Bd&ur3R` zoeLb^mpbBOc>=K!uNaX88ur+=cq!`G%gmXf%(B2&V&0@K2;hSseZ3g|DALiWdeLtC z(SEa~UmBmp)6KO~XLkY+hcXw4pO?K2g!BP8UDhW%FCMa;EAs%3zL=Er0{{>AV5qEa zn9(#%5D0V0T|eUntseIyltxK9)(no_2}Vr&R%(#qk1NEjfE+USgV~v2bbTnXq?>O6 zx|f5#lubT}#4zz~?5yEbRA#K;n%{Gudq@5QBvJ7zIpsjjej7NzL;c5#RnhuIz~91{Ts0{~Xn_SBWS8pMbfu^xy{oM{b#<)|9g&(b0^8&5MTx$6MAzrNyHR5tPHi1MbTX}jAKo>d>ZU-spQB@V+7U$F{eTBV6O8ix3hi%L3Fnkh- zI$;4u6-(fk$L^zakWYBX>0^pT#C*42#N^xt#H**R|>s@L~n!`t|}&*u*&q~ruS zl7F-m|IML4H*pMkR#{x#XQ5p7$>Qg^%(=uNVO(Rly0~ueqLe)HZJpa0`XgnUnH8{l zks_iJXwE3ldoQ#~OtN*{Lxt9LO8XPG2N{khJFoD4=-{mx-d5L84-LAAH9?Rpf z*z`a*H#fHtWcM>b`aDsUuFjTy(Glg*Y*$HhLlOF6lG^rNl{Qfj)InQk)g9 zNi9c}tCBJRkE~HA;p_vNf%p@{iluYX)4zQ+e$w^B*;0yn#3OFRzbynE6fzIIrNSC zk5@R=pSg0`JeHCcbc3IjVqhUz9@!K&-v$W!<=DA|t*sN{uG4R9+Oj~3`ps_c2Kc^; z`SrpsAz>%)6e)by@y@<~a`U`8wOekHbR*j*c%#%}fr*sSLlBx&9gNZmo41tF&3}qC zDJxFE@{v?L{)?GkNbxq7s&p#*zFzIs(d_D9ltSZLNg}nyLaIc`xd}eNXuo?*!1qYe z57?k-=>&{vuQ&$kPgXT2q2X^n*IAXnu2(iNS32|ddJyIIjd;M`w+-o8u{1_-wW+{U z`8>%SWORRfN6^-|rw#}JFe|usZ~H=A`$=U|Iy^56yrktB?{?^JRZqJ*ouu~K4903& zYX=Cs3aMi&ZG&%hdvb6iAH$P#^{5gc1A&O4OEElLl*D4ftb<8y6GADO6 z05K29`2q_{s0S!eGe<$#YMk1zrbay-@NCwfaU!FJgw?K6P)cApd3xhE#m?PZU3)jNnC&2eD)d~DiLUSb zkWXQ<3{3CqJ*ITv&Sw*wT=8TFzp^T19nm6*Ddi@{q*%#hDadW~!Q(J{9d*W{OctOX z8DLY&YkY7@xv{xfB`e=g^|h59vo5#*BwZtA#gn<^hfi`PX-e(x679XE&a6niceO^Y z5S}jtjFwh%Kv6PEQ#++vx2;b&2bqPMik9_QjzzSgV}J}$mQ1?Xd`@Et8H)(>36+v3 zJx)m-F7KX`;mQpYRl?@M?Q5Ci6|K8Lp-bytMUy=)s!MeEa(1=UnB}_Qfi?+x_uu0} zN+MpoIN`()ly{uscrW@NiVM3>wV376EIZa1%5HyoFrJC z^F9_Z!j!e*>g_coMx1iPWhWG%IfpA&3RiTCbx{dJKmbfIW2I6EE!Vs~)tna)Na44A zK@6lX;XB&%0%Ix1W)VeCy4_I+NVv93=%|u`);?vNIR#doCr8~EzaRh0U?1jTpz7LFdHnJ) zAE1K7l~`ev0i!Cy^XqlsLPT?00f4nOV@w@vwI(_Reho`N@du` zP`OILBDuqI?6k7Df}}}`cEg5#LE8}IT01I^EgT3eG(CNOPSH;GcWY6C|NT1qL45G% z2g#1rv+K$H7H@#mj)F+lDB}ts9svi1LviQtxlkb7`tsGmaDaC4*&_&#d=<5Eh!W*C zHca#$kdG3(N|F^kru{kW`khL1Wc5@27)l`QRlCq?3Wu(S`ED;&xDT{CpbtLCC+s!R zZ?6N1P>yL0&9t7zW$dw1-{S-ah5Q4F>|6^kAoLfv!KS5t>RFmM-x>#MB0rseAJ#3& zy$L9&@^0mEqA>%NVnAZcLT}}Jxj^nB-4JBfizyBHq(@2qCN(COKytIII?+wo*64^m z_KufVvd);9PxgAv!~hEIk1A7!cnR;tN`M>P{^auFe$n;2d-%<=KrPUg+m9}mFAP^% zZmPt-1=3xaj@AxsI0G4ec;?ecSo`|#g$+vKt^C5}fa-AX`)#s1yuv?m`MqTT#GxmT}i^EX4jRRBv5>1i4 z@>*X+KWi%l#E_6^NWCAMrlnhg+n%f4ZFG8o z(>X)6ZJ5uF$v)X7AWIVeV}OI?#$4`6vDIV*{c^O%zXj6&6ae^R;Pwv|!16(%{7M8t zsv8J$f6B2>By6!B4+u{|Kl=2%%q}z zCujp${y^n#V@O!7C_2_ZfQEi<(f{4Llg-hX^@sfrxN+h+I|ogsA`)Kq;SZkLL*vx0 ztdGYr$7aui-*#!Q%U^$S&usi65lxFIzo7Ea)(#@C(#Okk=dIp_8~q$8aGoXth1$;^ zsz4#|ei1y^&)Z#(JgB`m6YsMel{CdWGwsh7;bZA#HJyTyBzQ0TO-lkks*~rK<(~cP zw?8!kY%f{wSuv0yP^;)J_QQV%pw4H-S$LocIjtT%!V|M9;@^#6C6eIda1Z9L+xipM{fm?L*|N4BN z$>mUn(HF{yc)edeXIftZ(x{dYblnM5W--c9lL3 zm_MVedOR=RlxOG%^Mb@*{ru0{drRS1-_DuzwB9pk)0>W_xulww7^r!kv~&duKG!Z` zRsMb#4r!+}RnGgVv+X`&b4LS{K!%tPC`$Y}@eA3@X}zWViFlotHuX<4wuMfqtGfPx z9lh|kjd-tsM9fW{1IjTngr$G((*47cby1xdx~+j?I1{LH179B6IXN$T_#3&B(*9; zBlk<^ViFmY`h?H_{jU#MNl7|j^j#D~<-a~$%(O2jP9PHBbO#*1azu3hN23G8%+Z%y zOQm+S*?*ekVgHA394_I{v38jNUmj|^SEF`#b^{VFSX0U)k%KTu%l_R59s8DV7v6SptSIQlJ z`*3p(-~RL0(j-n~)_tp!*jZBr3_$3jTS`aoq{Ye!Qy?^DP>JKM8k9Y!f4$R3<>#{d zt*!uPCv>>&?h#1}#mPa)5vWaU=rkQ!^Ub3P{%r%Ll~$cFlnzfu+@gPZ zZ0D?P0{Wr-4>eGT=8^5AVt0tsD;TXq^8?<0devs2A!#=L9)lZ+*lpi@%W=Rmz06UH9lJr`VV;1P$6k7j8ewzhg4R_VHxk1*E6OIqE7gzPu(c?W>9Jbotb)E4ao`R!)fmI!Dvx5RKG>13QCPyc%Gz$1I z*GJI?8>h;-{>s%%r(>F7;*H^Zu|43jv>#B&vw zFi?W$C9TK~lk?N1v;BY9=^scH%)-n8rfEpKc(ieSz|{@TW9@lbpNjePP;_@PmW}d0 z4aWV!<=FY8piGijZk=26<97}ib^@ZnV8P>d=f~jdMLMCLp z1`^7szxB8eMpT^XOiAka7+|ROjSoA14S7Jg!$#?y6U_KNq8I6p3@U}8AQ@rvw-9l?9Jzv3NYxrTyi$7#g9d33ekaAj&Q z#Wh@fVuG@>y{7LH_Tl)ZzWOhaL6=25q-D*y(dL^}$EFQogVx$@+D{DNbxI3siyX@<3JcdGqfy_#O<5dole`}LgO zIh-{^Y<4uqYWvs@ldO;oy(q6l$0(;_m{?uy*qMNx6ZzHC16NIY`LB)nA6zH^L{qso zzc4XKYhqvMA!&gj=GAu^oGrJZzD9-^YT&3*JfhHq%hm<*NVB^@y_#4MS*@k_2dhWT z{cc|K2{V(s59_~hl6FKd;cK0OnSABA>lX#U zgP7?>pm4kEyvTTwsW+}q%&Hf$eX4xeyKS~&DYnC-_^R02MRnBwjCN>Z_b$Te-maUcGwSVT0|#|LM`z=t2kZrfkE0Bj}$>le)~FHh52Tm zPWX0p`!%OG{*or%bA6SQ%fvQ^>G@uM^0Q)6myMr9u2q){xNeq-ey-p5^B5a#@b{9- zEpH8RKrE>CtIH$V((S(bXH_z=PL!;e-M%IwBZW6Ym#d@qW!(;k-An+5k{3}*1VJr?XZTHW+YP%VGmg|0eW!lN+ie-OkW zegFQiE=4kBd1Teh5`7ZXZc?JaETv2*!lIaVrkrqJ164QvTR!}-lQj7NJY9@3hXW5c zcyOGYqWrZ;x!W`Q))-Fb#k3x?k*(Ejnhibck76=8r@Uqd7I|- ze*Rq=aSliF%wO{FlkNCph_v2f{d2aD@a^$P(*3jJV9#9HIGh9wBX51x)}-D3&5gq|o{5yFSKo|9|ZeBJO0?v_K3h_a*7(bi(J}93^409$( zQ{x<{cIVJVt;#(h8^3(14_m8DjXwHi$?7uOLYED)eYm@GtlXAIs*3f~ zW3|MVLiyNm4l^kp%d=hY6l7#YN*MM!eZ0z2Or(82f0Dk8B6uR~<_4U@$%(6TS@4-= zj&LP|uk`ZB>U^%zi|OP3>|RmS&oIv& zlDDn<`bX!Cg-f+p(_5wW?!OZlH<9WrBZVlS*($(Vpc&xO@C()4IIkPk4{_z zlncU+rozxH7d%Gsf6yTUBXe;DTj zHt2;gAXX|$6%+^$=DfL9sx%3fnOg@$>L9qv!~i^ zH9*pot6dEdO_Y4~^egoTkNr{0ly0GjSR<)^1HZmS@f@;i5`IZTY#e==FIB9LdNnXe ztD`_fd01#Q*h!NKYpF?v)!lFt+j3xb3da z+qKbTSGz4;n4ywI`0j3mu5F6$n~vaDBq!VMN4G8Xv_BuMX_oO>rnAc}cM02U45Bz# zzh?wlI=gCf*#xA<`J>DdABxBLm+D?VnQ*z~-f)D(^3>LB`l-fD;pvKD_qG9tnI*ei zpS_NXiPoI_P~GA*U0mr`@`cZ6pXL5X8b}v-sc#QYMf0ZUuHd-a)Z_YB;re0=74`KX z&69azJjK=pTXSfie*NldHy&7Klaf|PRb^t1LAknBgjtg~?CZd_!8T4#gGN}ViQxa{1UT8Gn5_pA$~C(=({oX9p;&o6^I-(9sb z=I4OyCn!u}%%aOY;mR6GbZPGe({^Qkxs4#I(}hRLG_DCXtYmd`QNs9%nnhX)}NeU$susU&w^ ztl4NH;vCmIbWQS|*uX|tQ-Yp$3qdQk-|J$6*0;?xY;mxirgcoc9_y6G)y}O+Ft#a$ zAoX^|a6HcJ?fZf03GO?qKUm9-i!_Ra^7KpYF!|rBn(v5grs54Q4_Q)PA0PD&-P-4s zjJdoHk`GjuE}`GxbmC~wkW)Wk_1Dio4Ia>5S(Gh));^A=Qbs)O?-+FP(>m>U_NoxG zw|1>D;!=$AyF|)PiM-Cyac|<1n9c0qcKM1g(Xfa>ohS+vQOn=#*DvYcx&-*oZ+CyO zdh-ep3JN;WS+KbwHR;s>i5wNg+g{ZJ55qcQrmf_9=kU!`=yybIN4BdPpRSSeZ#Cma z`&qtz3i8PFGo}|XQbgapJr#Bya&zSckgUhp_nM+fydhRE1(coQ@znxnjvdg`;~d6O|E=;F1+aEaYCJtSt$E1IL#N@2#Zj&(1~b z-O4#x_W43+qj!C+Q0ieWfQ5Wx|NQhq6)*RJSCkWtaV8ycJtY!;v8E!`9kNh-dtbcg z)GGU09EEATbV*N@R@Rz~gO{QPiYFo!eWXMR)c{6Y(a*4ecdc??MR^SU9k*H@u6_#T z!>qmLDnPC!E=b6$kReRGHaEqV+Sk{3Gfdt2B&-a&-sQ^(TLNbnYzsd|pHon6%Z;{P zk$OK&#atxDHR#y{)0+AVzfAFy zM?39Ap>#L(A8&JSeC5$(%HJE7Z3Abz&sR7s_BtiiE+sjIh4Z1~uZWI&D@m>P3T{`g zSoGQKSk`A`A(bM6-V=|l&#ZHvai8e=juTDgu4vaRdaNUl@+dIz+L{jcS>}VA_c-M- z5tr5jee0WiImttmm*R}QzYH)_`g5ZPuCH#C@O?f`ywSr)SehV-}EWi+N7z9x)XUdJPby)8vMu|_GjY4fU!v~R~>x3BE2)Z zdyb@~L{l(o<&(p7WD?1|6fFW$yGsq+!y)|!!jjF$w)t0xn1(WxkT668wPrH@Xs^sX zESD)oqJswWqz}4WP59`N%?AFe%u>|7F>sLjhLR60W%>Y?xeXM^jp9C-#_qF3<$+yKs~BkTRoL?(}r!}6wRcjeHI?@>%?2% zCq4H%zB)s6K60ztS;+>McZR2AR?RHat8kGn+J4EkVdxEqr`>M>eWYPc_m)VATvOOEuxMXSq%HV~l~b%r;6_RQ+auq_dL zi>`d1A^Oiy+7DNnWk*;=2CW0vygGitnvuz_aay?#>sSZDzJV_i1L-h4h;ovNf=a?( z4sCe){#iy(w=f0Y%|6{#U=u>h`t{0oeeZ^mJ3N)wK%J#ZLNb9J7;E zU$z>(i;fL{_zF~yJDMygqz8=M6$TCy+82BpOYhPC4%dO0m*Bj!8O^k$pHtjUG?#f5 z5U5_@I$pXqZ|C$1x<_30B|ff~V%9OVX!J#I;c6eo!5Xx4EQhHjJrC%>>6%Y(k|EK_ zLgM3j#m^tC>cQ4wK(bL!@r`frKI_=$2FKN9>gb;jxG!+~7PX0f4=q0zv(vaQ8v?1P zf?VQT4zF1bXgczsX0g_9qYV`f%VdZ9^LEz#==OEaS+T$D;}4zb z&Q$cv{LmHIoElRxdW+Iy>0r{Yl~}hwZOA{h5sAYIm8_W5G)^Zu8ky*x5o7dk2`HLx z^;>s1OZG-HP=1i0jPxG3Nt>)+&cP&>#7U0O1VA*aN_XSDW;jKV&dI+i?F;B^NlfXb zKa{5Av-;Sz(Of5lP`AHE@sgJt(E!UHP!?wXyT14TSJcz+D{JnU`AWpY^En8WM8wApXtOp z)P9#(?O|FnHYYAt0-i_7aFpGV1H&5d#2|-aV86X5iRuM~nl%<}d4mVv(&?o1IZuEG z3EgUlB?N9~2v)A^zi-()2+K+I-R{fabXpsEWQ~S}JSiPD#HKljt+n>HJYY ztIw;pes{B}5@+f$Jv@IpA>%CPIl*r)v?d#zC(SHT#ta_aHY8mr@-~Rp>f@UUp}Mr* z&-bGz;)0ptR-*DJlLggY1JR8D_L-5GQ{Bm0h>o2XwXU@uc(&do~)H2LD zcXTB`Y<&o%Znn@hN`bpp7`wk(Dj&NJcWRHz1fV-Do#~zQG+n1hs-nItULQ=yB z>$ziGP+~LBY6NsF(R7SZ{0K`G)MaFzuYc{(wDw;%mm{l$+%q|aV{`h+snWFJ;xozk zGxv~7lzep;tgQHAcpw+B;qjulfQHZBT|mXOMk0Aq#xca@wsK4`6}o+S$e5qQWE=(1IB zeXn_GuB4baYH{yJCRN!2+&WA0Ncis4h^>CbYcU1sLeZc_`uM%b8>%p@wyoBV)Y`yD z{>Pu#AD$Vd2^6UDFEF*Q@9&PL)z{Y39v9K|e^g?mjYh}EfL#pO$mzZARr#ng z#7Gd@-+e`>bl7UnJ!za*6{K6X2d{{)Qi8f(L zLp{`@)ttn~!Xd&pq~8|zG<_TNIp$wohn*ut(0 z=R%2^WYKOv%$8c2!Fd*A4x=Gn;=B28Q`;)6V@PU6}g zb=xuf91HpKolJ9z6gs9id$-0%V&svbP-3v>R((XAy5YF+rIm5jV!4rsdaBQ1drys$ zU7A{eI&5=tl*gwGf*-6^cy4Z1gygJm3-{~Y^!NG@ zVIIDCx3r|;25<$vd2e@g%UO#6e8b6F<@@{Y#MZYDZT#$f)%ZAIJKBTem*i%Xf|8ZB zaGS7x^_3FDbIk%YUNKo>ccfPGY|zC_?07f6i5#J95vM{Fo<8E2PCI~ku{kJiH~%BE z;cC1fT?`0)7U7xva$3tGA(T(7Tk`irKQA+Fu2U`uP{;2Z1qJn6>I znJTfD8P^HabP8u=V24Y*UlYixk@EZ&S@d}#wDHwC-)pN>KN%<4jnDgEYdi|P3DOHA zrp>OH`gkX&)->(a%FWbTvxbntB$sfRMOs`LCpV<83>w6i}e6j4l(AOmU2xv zE;Og)={J(yJ#KnI6Y@^FzTP!#DNO%fUpbY8o1Doho5{1-F!;W80UbDsS88Tsb_A&v z1M+sL>6ddO}sU#$(W#IAGc46o;{Ud1!Y3>pj1fb8PNvUPHJz zi)?$MT|Qkf{Z_O$b%Ke@-UuaMqcK57&)Kp2-*MD(;$DxtjlBj+@8SFh`0_Od3AgO^WX`VhgPGzI?dMha%IW#+%lSdEH9Kg1}U4n6U#b%zjO1?B&-4TZoW#a&! zx#>4E3`q9SLOVdQb!mj>zph#F<1l5E@RZwF?klOyJ6Kg92}1AReNe7)ew^uP9G1ti zxenheeEK4>O2LpH(^1l1@ldvSQ<=g()ImEomiH(`vB-hYyD zVGc05-gMI1=^lHoITKdRo$5KC(2bj+N$r$@JoS%=)wL8<^(x4K2di5US^z)SHFYa< z!g7syNLGk(J(jP0d|FbtU;pUD|8=81R}OU<(xgv2RYM>y_ZPLPZOSTurGOlAbr%{G zjp7RcV>E2CD-fzPGGaLccn}x6=EM%g8#xugFML<~4pM|QbJ^Rj4IX;pb}M}r%guFQPuAyBS;4VYtMCfkk|s^GwlkY**jeSXbP`fy zUc@?II7?vo>${idQZS;^Di)PFR~SLd(Dn~3{EoX@zI_1EXkPiAOv(F&U<~ly$seYS z8d4kF7~11^uQ?Q|?b_GZEl5uTmv|fn8l<2CsLu+pTt?z}wDb4IxS8gB0R0@psWmZE zKMyC^Sfb*kx8O<>qHzbDl?z!oh!piSd+zalqDW*482VWhQAtrJ9hB#_zbCwA_PWLoTs#tE$q(>=29z zP4InRKmP&1OdEdja5$05~*-iNJY5*Llqoh{< z)8cxEgm*M^*JTKm&?juaw4+v9& z&ii-ca5E-ODZ=UbA$ORxv2YdLCv!Z$xBO)CccDuOqJs=0fv+kB2 zbrhI0v$TPvsw1_snnqkb9J}&*W&AD&YPw>E#b2ffa8$QTsu$gtG4-u!UDNM-ZyvD^ zuIeZ}Px~35_(Z;}b8WuHGdxaQ9hVg7NNn7XiyWsWwcMiT^14ykIDa*vX|*laOqY(J zDO_a6%np{IldzUcffiOB1O!i3Kr5?#Yt6dm`f_h$k@|>yA&25s@6|~^0(S*hhwskd zjPsL&i|euj8IZ=k(8q|E`|<0zxajp(Bbx(9eMGrevCj=vz%P~zb?=Oo^q7yq(AL)y zmwlW+aW<@jNfJwgy7ar+D$nKW=i<1RJqcGO@eNmQ&Gy#zsqK}A)sMTW$Q<5~zW!@d zUsgjED0*#Oe+0;emtl=~Oigz*>Ek*?2ATa!35*EUZ7b(?jD+7)?px5MRtVzsBiOqS z69FM2kd&-}au*r0+dWJ_x=TmP4VxS#olFOBWO^`6$|jy`J}V=9upPG}E=3%Z#%-L( zIp^oP=+GA=ROxnOJ`|u6FSxod>djzYZBVo+gv+3fq*q2tUiJ^>_Xios|Ml(myNpxn z697!a4p91_W=VM>pGY-6AvERX1i0|t{K|AdOSS4$n>;$nApmM3oN&^NPuIr0y=WxU~49C+yB}V z@FS}1>|@a)Z)(4Fd+sF6tckSo);(d=MEM4`I_M89VDx9XR5ydl*TQH_L*$XAoZ^|H zS#E$cctjw`CcFbgr@CXNiOw(f-GmnqeRTR}^L>Z$`e1}{$-HtSCz%X6wK(9WKxU@( z)-P8|=I^%&Rb09xiQC#QTdpvRN9u912X#})ObobaIF+cDXmpwgD~LPQrHq;?)K}F< zsqoKgpZOQ!>9>}Wm!E50F~-v>Gh0hEx}-}^(?oZqULaXB4l2=9mF^FqKctr!cUGwh zuT({bnClCuS7bS49Z}iB8=uXAo3thl6KxsNw+LFvF}uHh2x}->dBc4vCki3Dbty%q z`D!Y-PX)T6tcFxg!B*6L=4XGbi z#HO8vT$)B7;h?NedKWXmJj=5s#zjyr78ju`f%%IZ%%{N{>2mUnZnP<+AVhEV3HuAv z8UFQkd1U0gbV5@Dwc#@QX`ZfRE+O(_xU^KPUhQ<~+fYnM21B>t#dWt-NtsrY+lGnV zuJTzao<{iR-?x=>hBmU&w38TRQhs9R97!k()ER~{H)dV+ieZI@2QlH)pmIarYp_x2Q9aSBAV9uYPDsW?z7X=gz{BedsV$l03>zW;Tpmx z{Q=vh+0hcE+KsiG)elf6PE+$C_wEno@6DUV+k*WzR@u#aZ^p&NX+ctibqs|~{pVoV ziT>f`Y<`_S026)YJVG`k7Y?Q{ta2L9_3qSNT?Aw=X~JHXuY=E0k&1oQHX9CSl?vro zOeQX@xl~j1>-&Vw_Lr$h0h01+%cOeDN{VK>Sr=y>7rBP!H|scV>;815h%1ur_PfM< zpUE)MG2W#_sLt|%vQEluX+h3K1Ucto>}~Ej8T$2)6wYl99zJG8Glt^t!n}zf^a3Qr8%k zs`+DapW+~iiJk1T$l%jGJwR_K;gMeFo9Zg$;M02lqo2%I@6v%3-p3uR&M%%-p`ID; zH*)P0oB9Bc>9E9^Aa}*9mPE#1gLURFu4AymCQw(m43snCmdy_1{c$d~U$A4DpJ4P_ zd|f$#vYYHtm&q&f^0r)1@`1`$^WB2xEg@Zc@77mcH|b1u1=aah3~)$$lzZUm6)-_Q zmvk%Uya2O>w6DYwP_Jv`-w>=b^kuNO+SRn6<0c(m(9+Dd&_TXD*3JV+-n0A0*d$OK z{dUfAbL7p-1w~$p&oynDbMm+CM*L^|gu@?J6}G&un|NVPmGki9^vB@+YxUcEV3=xK z8>gPoOtLFr!d;nO00F+!`QG!*Dcbq{+C!@~hEC_l$1Q@p%`SUhO#+wE$rwi;_b*nr zFB}`f?PZE;ol#yY+!Mjg>p+&`#>!m`bRgCN68qnNB()X10niTC#x=MFk#>{?V5`;A z06}=c5DCbZ`0BLbO7HMN>AUnLh7()A#ZF8)*m2l~>lG4yZRlOpzJE0_D3MN|9|UI!kIAX*q+g1r793@_m#49$Wv&-Sb34V6aWU#|&vT zJ}-yvnIV2>>^v<9p0X*wdY?ze9X-6BPeb8xCeZ`u1fu`EItm(Y@#?Le`qFQCDm-TR zPB)gIim#_jx*V5BkjQ4!lEw6U{SY7Pi3uiaP5^B^v23TAs|Mg zIPlE(nJ;;3Ji~1!J#p^q+nLQU7x{+!Fq`VVs`#;;0C#1|Mit;K30YNRN0$@* zm^l25XJ_FQ-8;fZC&Jl(L!iRM)pXOX=c_Gh)D7}tkjMbb*%%%&WYCRObcgYX)0n@6 zjxxuwyq$c7hH953RcmC%#JXSR-YF1mp>n+D`n}P;ksv3;o6~-A(T@6fr&P>wPq$3e zqh*8)jutyH(VQCZsvTHT5}v_@ zvq_#~NJja>X1oEtK*r=@6u-Xx$17ahV6wU0_9Aw7#YTbuzDdMhq+1OQM;`T3SpAS= z9p-Ufe|=tCLIr|SVz|vfJ#*Bt1eNjJt)y4HZk#6f?Z)1-;rwfp4jO6sw$2-`8pg+( zaUr?)M!jj`+%(UB;Vg>ncR{{R48ydIa#gem`p4&QXMvDur1E~AjHn(Nfa|p@WcD|H zXi?98g`_GnfE{B!%M@-MWRcF%(#3!_YvGmHoyR=2SRGu(Me+8>ieIS*F?%nOk(&7^ zzNSFok=21@=J%x_0#KGo={Y6b?zs=m>`9&S7XGT?m4(w+m6}H7JM+flIz(N<3YY(?~xgxy8BXmJ@%KL=DA}aokUp!gmg3aLnBFN!Z!1Lo` z;vc+W$zgCqZeO8JL|!t}P=%Yv3e{pXWYYNynxINzNO`E}(mTu5R&k{T>7?o;mF?24 zh>9j~@u`_3i@N=)T*V;Z6eqGBnM}SV&r=?2r<1q&ZGo%Sumo4Iea|Lbc5jmGg zQ#x07j`B<<@>Fiv;`&W}CRf^4jnRz7fz_Z%#!n#mzcwU$03rsq&^~?chwMlSfPrBj zS*H*F`0@|Z^y5ztx4_)GggU)`d`M!%W0S`}IL`d*6n^~VnPe=0jF>BbL>3T!!^H)d z);WgSAI<*TQ(0y?Stgk8qRouW?bHf;?d4DZh47;yIfNLN4I5 zB7#5qB~1M98%Z;GR4S_T6=6UVX)wF~ClH-e`|swO79O?~@p1MIp_>>yZYli|qu<|; z{U_;auLMkLotv8`kk=BA#|u1>v`C06A*}Xq8w+>9Z-$Q*9VH;9qG<6zMWt4eQGFBc z_u_=!|GnXPmMYB`p8-JgVNyI%b{4>Li-z;M;6F}~evM@6`NB7|x5I7BB!h6Mpr#4TwES&-W~V-sl-VD?(7h5G4|DPuHZRxhHe9=lb=n zed*8tiiXMyNrB&!PXTBGVu-4oa5DvoCxKCi`&04?Y)7s@e{I2s6_3in8V0|DoDScD z+Q|td%%DyK$72uQ3_;EO?J|i^eMvodth?J`JAx!9#gJxzgg0EI=9x|ttAYFIYxy+9{uhw>Q7oS3ZoUs7=DxlU3<=9zy8>yI zN`H`ihZw6fC5W0A5n3Hm=By&a^d(` zsCT2&HRku!6yXiRyEl<&1X}%1qoq=^6@4mmk95fk$}iEoMhO0SL{4JK!rDXW`{6R- zsQKH?Pu$We;LsM(xp16F%mSbuUCz7p5y15SKB+=HOttvB<`!Wg?PEYkaot<|oj4L< zguiP5xIM52*aSrh#y}wr=y#2;Gw0uZF#=;A&4H0E-40=i_#X`*e=&%@5Bonh4v;9} zjU1zJO2P)cCW=?yL2qYc{%-z%Rso*<4?t|U0|Ts&&`rA$n4Z#|8OdKEj0rV3_*8Yt ziI;7ETIBzT$zO2(&&eX*D;_EIN$T#NZxS)uy8tlsDN1IIgqi+3+o__c@yRwEJ}0VV zO_`nnNdt@v7YGLFwZWr6*x9#XZ>85b?#`5i^t)fzX8tpg{_YEek)%uayDp>KW(tiL zbW}Zx2&aN5;Pa2LF~M4`jnjP0NdK;jKQG~$)?t-7{+GZ2t6P}7K;UGILXwC#v}bEh z1|H}&{tNr3hYh0BcAY{VhSqtbMqnxI0>n}(eRj+KjXas05Mh*kG@_p+g%1*WDS42j zhdO+gO1Z#;k#0Iyi7>*yH#{TGM6I3LShX{Zu^0xBIQ!jrf?iB&BF$i9MaHNgMmfv( zZ<7sVO4?5(tAv~9t|!hBu#1o4@aOaI$#g=xnuMME7xG8gq5o}UUxXZQ6Ukuh5iwlK zA}q)9c9NzS^dxL0!$*+Ms+^F$<!?AyPYtR4Vux}13D>dnW!kb5G)c0kou=5$9J%f(C^<)fk1Ujd@6wBClYc9UcvPmFc%Dc z$trDv$MR(9H>Q&C``fpQ<5I*qI3qpQ77AQ(W$+rymHT=GBK&K^vlw9ddj#v{cnI>O zDF+5de2x89_=gWKbRfMx&Y;14(D%KHVR9i6-b#?lziDLSy|*pHByz%{U_kC*b>3Z{ zb6OQ8GgjJ3lo1yq2qXX7Y$dTrfKxmgHGT5OE|R3if9ZJGa$C(en1LEN;wy^t8ZME5 zA~g7mjWsfhf9sSes-78CAX$RHe)|yuYf^jjJ|&PE~3C)wXr^vY2M zZ0jv$V3rT0zE1`*%;WdP1AWfUkBA@87WgqaV$Xm6;>ms%^gQ;VLx3<=Fa8E-mp3Z< z{Jh8a2mA$EZ`aIPo1oxk8V`dx6ILB}2GhcSKO5(;cwMc9w}GLkw++Q1j=#ObgBKBf zBcFNT{m4lC-Q9($c$dZg&HUK;0r5Qp1NW>TI0}ESo%f(YN$NG5o>J>#5oHHJ2;Eue z>JD5iovSXL+NweOiLI#BJg1J(N=d)}0KKg1GTwT_ZNsS_>p`t$h0Fc|LQLi!o9;6o zOZG*;eUU{J(W<<2Xutw8>lgWNedd8XW?8o#AT8n*~ zSG}Nh0ZJ@EhCw%PMCcV8voD-!(;0Q8C>e4cT-coYctZSP%C^Igx4BU>^DIzXYuKuf#bFsp+rD2$7 zniga}njEay>GAvZX$stJW||z=G)7C+36Elbv$gR%y1n|)J$iSmN>Gh!GT^-XY?AuB zZIa2!H4$)!s4F{SPgGp;_;s=$l+8X#Vcz{4>^eJHCA*+N^O*IW_+ie5?4=|Sl}R?2 z%UI>I#wi2@x@|n3b7N&5;IqPWN2x-+IAxjKAzAML&fOtfn3&MO=VQJk%3R}`>7CT= z^G{jb*ZnM-D!7U!4Du#RgaR)|s9&#+5fK=9ipl1DX)jy3H$pxscrZxhlB#-+*+Ih} z84V;FE`wkMVZOOyK^S;8+4K103E$#7bO1a@eW+ZkkL3lWQ$QRb4F5DtUzq=jz`}oM zNWu>TBq9RE!YZJo3CR#%ywN~#_o6KETY>%fL#?9x2iOkc;~e@Fx<|aB*xE1uz+K{> zl1LAU<2J&kB_7>%6*jY7o$q6>rFW;wW2I)_Y$YSq_0-@@opI)3EPa6kHO*}Tr(eW#UH#!K>+6|ye?BM=uT`5tMQ-1xf>3- zwMlQT9rP%-&dTpB^ap`f!vJj$-$FQDxu0;V=0>5%<_1mBO)d~(sjbV6a*fexd$oJN zCDQ`=y)6yimbT82=(a3VxxTo){H=7k!?HRjS}HgSJQ$6AYr@Qw@(LJCM_)7q@GA`V zqHp{eIEr_n9G~*Q_DN=UF&tf%-y~h6(}`m1#%`j!>eEZD2L#PJlR?e`whnz1joGflL=Y2?mkxE*S6Y04?0sadY+i+vVD3o`&rukEoKRYRM9H^Ts;E>>jn& z1V*hV^`cOu3pk`_-cO(XaSlq_A1+)6$#{L5k|1o7DvTGeXVW0u;(Dt8Px^qNtc3gr z+d_h@2+FlVj+`cT_FW-~#ry81vaS|ykDJGhf7rWus&988adK9%F3=(?PadH8hVi&i z8nsCDRg(h=%^-y7>1xN-6=tZDv|!i>#P&OGyKXFiOH)^eqBmSkrR6b54U;#B8dS^5 zu8Q)K#V0hWeyw~qhfZK@bs&(!0)GTRC1N1tyTS0P2MK{_5=*fIpxjd8)o=m2Ov)Uw}=@TYItH17Yg6^Ce4*SlLX>u=w(-oj3Hv?)rZ~CgX2|uXNwIQ|`EBSplo+Q+sPucdKJwG}e;AWvNU{ zCgAy*vGUF0!R81g89r*!y|8aF4<}-QP~F$_BQr=ndS49V8VCPyxj!Ht{k+SMfvE(f zO9C?(ulG13?H8{+Cmb{Of=9$rJ)WnLyV#5a16)1hD;ggEU_(9{q%?^lS{0y`E{#`d8$FnDw(o?nO5dV++x#;tqwmv zTJh>}k#gwvhAL%Gs0Dl?IvIA19?Zzm?pv>Y<{^K|3pRi zNP6YiTy*ETZkq5I#5PC#STtJirOgqeG26-#8Klf3P+c z=bIAmxM4M4y4!nPAiZc|!Eq$;w4dTpKg$HCg^IrQW=m*Y0frldcZ2AmmGW|v;?E?i z+ubMMWn^aC|C3%ISl06Y$?z2A+6{jXxtfPZcMd%%a-!eTzis#0@AB8TcoG@?LFS9wyP&wk$g zxcu(a3P(YPc<_|Yh(}G8!N=pht>{FzZ%8`*D3!oR*#c@W&Wkx49SEk$XID89Im(dV z9U4$)Bg=G_Q52<7y~>eW&AnK<4N9`))j5Lz2X_hQ^;v2>>f?m?>ZmPsCbv$-Ct(mJ zP01+o=}$oP>d`^7is{(h+W3_7AiqtCGi)U3JMyWHnLUMuWdiAa=N8Esw91{W$^duz z`|iV$i$^)BVlDIaS2gtevvRE)@M*Q)%GSgt3cx;#IDO^a9E5hdg9@u$rK@wzIEaJD z>^tu_<8knbg-C6Kyy>Abco;cy#oP2iuC2h2TmX8#C@+1B+&uFZ{{*BExvQP^my=us?HpubWh?$L)3ahZ)q_Imdm U;_WGVPryZ7@V=3e}iziGbE z?2!GE_oVzIa-7|$y2o(6+Ea5@R_qVdx`cIxCN|8?Jwz>*e!W9>*~np}X83!XZQ?x% zQFg z49>~h&CIR#Z@{gqWFf@DuFI~GjQ8tH>UfGu6F5o&h}W7#GZg20{7?%5q)i(^k4Kf zvLrX0-iY5j|Jwa@x*j~7c`o>nlFkG6c$Y+`c{(kSx1-MLQ)X_}$y7>-l1gr$V%6CS zS9}a)xfgV1;6DMzARFAMeRFFKHHcZX2>b{%i@E=H=0;VVm-E$QH!hqQM2Ljop72kH z>%U@;MWh&j0IhFR?YOJqOQT|rqr$}l1_jWroq}_Z zx1X|+_&E3C5tU@e5Z&bG-R;z8*hp5$*&Gf~BXminr2LuLp$8A()<>?5 z8#xH|uO8n-xwN3bo}~E4%#n%Sn2wyHn4XO-e8oY7a<}sKgp(>s30V zpiGs#%EHGZ-F<9Uu?}9)lGf~rTg2?|W76Q#H#I|4vU@x*D8=00wC#5bv*A3g-Te+SDc`(Z6c$Boo#N1|(Gtc3ElwGHopJ>{ zeT>3J6JM)M+!}JKR#1igCRN{cNRzGGxKG|vc}(Ry$nbS7x$xN35>oWb1ziH$PNXzaoDx~z)(qxk6}Uy( zOPTpKMt5s{@q*7H=d{JS0b=16BGv!0^<7KS!#Z;`>7@mc^=mxq zJv`Lz)%uK1Nsnn8V*~<5(f7?6kH}vx60LV7_J2wR@bYSj==9FJ2C`a zckX3@M36Iqov~u=Ri;`_%&v8ffFfV=iQM*yW3)3=s@@_3O8v%;#%$IHMcbBKGhdWc z;|ga=l3h1jP5O6sgB=i@>o%YucLZg2?4UlIj!jzjR-GP?$V-09u}AAHYYu%Q-YTqf zUdc*KoMEF|JW+2&FWEsdl$vQKUz0cW!{IqMyrnP05INi#nvoI#xj-&@X{$T?Buu_Y z$|VYB)Ap)Cnw&fgS6p{7ZNf7nj?aZWb^^t>oHnz_jSV)gfAVT*zUXxs-F>6Do`k6wep3|dxNMc#g8wRiep2{(_?PJ;D}x>5CdX1==O(EYS`O)SMxjmEE7Tgo2Ad zdqGY2wzPa-#xT?HXuo`-z4o|&!sD{PJh-v@nQ#Tcg^fFs;ei_&6e90;%T8`0AE}TbF9N&dQk-=%Wv_Rz%XrVzJB2I%Yv3XkpQ-2L zO`F)6!3r1CGTGgg$6Nze=@~=g3y~(nXp|^M4(*VA6UC$_#b%+~i>?cZG#;rvVLZ5| zqZ+12Co!aQTf||fv{Kl$a(Hp6DXCfq`OtI6Vf}YXd8Nebcc7{Fj1xzz416N(Dr%@v z7mmu$*x03IwLSahFx`@2ZXB-=IhOGHmO<~SLA1q3v|h8Hf?CYC7(sW{3^CPyx`A=b zc+cD<_)oEdpLH_MyFp5vZc*8~eg^04!sK#MQ1)zir2@k2YJ=eR{0~_(XI()-< zjTmf;BdYRz_$NA`QS<2Q?S9gzs=(YT{f|avygaY7d^1);ip?!N*Qp}64Rfqo%QbLj z`+3xojcPcZcxlN*^+mNpGPrM|?%Wqm`AE)1%*N}t^@BaXyZU6H5_T1e7lyc8))-Z4 zZtguE(KI-oroexKzFnEmxapBVot6yS_MsqZos(wo+dDEkyJ@?&%>7iMDJn-!gW%C{ z^S+Ge@wDfe`h7Wav)9cO-xgC)QWlGTf2GEW<(-xLp=@n*8 zp_dC@(hGks8$>&+fN6hq)N&8iz>EkPPCo=?-gjZh^cS{^Lp5WR@^x#Q17$8XiZLQx zadR&km|oG)30#{Sac_w$fmDy?_srXpA>*i#8ES4Si`>x9gWpnG%WH&TY%8@si>v)@ zWfwC*CHeIg1f;*%GId_;d%-*u_GJi`;%vWgNoI-7z#e^i-+W6X+4XxF+Ro8exwyAJ z>ajqDz@bzjv)5N1$J0S)zL8h0ISQV?X{UZCIME;mP?;sy{_*5cyip(EUcXkur=PfY z3!I0~MQ?|Pd|>ui%FB6j-Jw>crn~~ZR4raZW@Bd?jQ&E6&dfV5?^Zf**#Es>%|y4n z#daYz@w>2Wrgp*NchttoQ0@f}78YxAP~*6J$4<4WheJ&%GNLu+_J)sWF*VX5 z^Zg6h$+dESu7n8J{3)>7$T#P#-YnJvyI1kr5H_w{FbiV8H=*QEEB8z?>qdqbJH zwbB!ob6Z(>t~;wHJRNd`v0IPd>6x;IG>HYdJJ4|HShbba9yCT;Hb)K(f4TEh&7Xf{ z7_%k`6Q?eBqFY%91HZ|}Z;c6DnWGJUzP?Qs!&|XQFr!hM6_@wF&D@2sOIjW3`s17g zT^?5SocK@S)>C|E&J16DW8Yxq&SaqDx4pG<{(Xl{Z?{cQ_J&TNCD3vLSME{rhPA}< zlvMVYR3xjwyoLh!pQdCkyrdgwvpaIMf;*C1hi{IyXHd41aeIbl^z)m`9EyqT2&T{} zCHnr4qTUMGdhSx};;YZ9+7c1DLhc+()ivFA+5a_&uXlbVJI=(>5S+;8CQ@w4GP%fO z+#Ao5>IsUd$CF}Q7|}e`hs#qk^Xkjv4qOh2^M)=`JxJQSFp*$=>*n1fmZt_mDsbw7nW3sgo#k{* zgnVbA+{eQzdpq!ZTgT3`^yTHVSzqhe6)kkqXo}=oz(RH-&g{zm zWRB3k>IMXqOoI(H`Q_OWI}mcSoI8&w~i0;e8k>HDZ zKBc*v1FHN^v%Wqk#aorvkWRep-Hf4%;3WswTk~7cQyJW{F-+gt<5dpMibEeqVpj4)d|Cvx(ru3wG)=Ov} z-O#1t9acRzO#Doy2e3}GJ5!Y`tUdNVszwVtzOK?yjW4ll+-_OkT$#!nyr>l|VEtO8 zO|_!AN@htA@n&VR`C>XOX=*xeeC2Jque6^>(e|`(^^Sp|%OGQTN6oYvMC$daI{z1x z1M{Na3tS7#0*3c|?zAphjyik49B}~ENVOg(@VHe{&SDw)n%wK{ zCNDE)aD+$%Gl1L>Vu^zjAW`;o&e|9#DPR9l);;W`;nn_aEI-~_XpFrY9Jh(NuTk<% zp&}-{di}KQaYT3e-cdvByim(XZeqv147%u|4GQ_0T~tV9&V{v!)3D^Hc`hxlUDlbx zgH^MNmwGH1wK7d>N(DF*i97sXTA%Jd?C3ei{^CNT>YSK^(NE7W1d3F}IUCz>jy33oc zmwXkp4&59S)2Lkng2wD{JSTaMJ>XL$rdE5+Gw&m=Woos)>< zl6YJY9C#1X6w~ZDy2lz2nY)t4ZM4uw?UrB-(PF;|=`~;3sC>Gd&!=vDOKtIMA7dH3 z34^sz=;uS;3@V^=Sm@`m5OQT9cNaGjc)`z#q~85qrY1}{=gxA7at``ssK z?SHiC@V4LcXyAuNmD!vj2$3Y;NucP{ligj$ML7j5%%4IorHjsN>XkleQ~~A@JEw6| zl$5Pe+m&=Ea%H;{g~OKYz!&wD^M!$D>dxf2yJcTyC#+&MQkjE_?RYyDR&&rtw~D=H z&;3?V`ftIyi<}{P6d8M+$#M%zRBzC4a>pwvm`TZzyvE7iUS2!mxQrO-N7FDE3oF9t za|jA;Q|{G8-6K>%VPlMnYc!Rx@DpW4T#%tCP;i-h7b6btAK~(LQ|5v3kVL#?RfK~t z#w9Dhg04(8GG=d^>x?^LUncI(Weru@DCZkBa^h~L_T~7bn?f(y&pvN^RS>G)!;YA! zyK+3c>#6Ikg#-qj7%F(rg!0!6U%=bKTxO zk#9eEugrbimv*ZtC>v?lIA-HSSAT2&qpO2z#{QR?{fMROheiMqLUFtHs?T1`^mM0V z3%fJ1DvBr2Hx3zG0+(HP9xvOrf6fb=h`dwH37UP6EC~h7MyM>F z!4NAG-#r5)>_sg~j_3vx#qXb5o{vD8V}C(fav}vUT?KtL&RD^-f2?btGzkd@HG3~l zRx}I|XECM&u2L#b131<{&EhW<6!Sg2lpeY0O`OjBcv^qsh6fU9 z%?pHx+E9d&$X%iz{f{mDLx75xEu4=)YtLkSb@TLL-YH@=- zNJ^3-FZm3q>~%S@{Xfv)P^~o7b6N1I7u56n?Q7%&B6>cfKAe@L@}7^Em+)=Bgzdnq znJun|1j+o3#FLXM3J{eILjTz@sO%TjgXGT9L%ojR8n|wc5q*rvDLHlKYu?arNuJnD znN!t`ij4c5;16kdbD1h_RR0nyb@t$0Q^G1eC$3O^mSTH###VWkN5-~-O`GSZg%C&o z`_`p|XSlI3CCtU;sjuTLhmfxC+kf5zXAjROsceW*cyLH>@Gv;*@cf=y%ee94uf*ve zzVXsL->CZZ-s6KPunb2nf{5DaE*Xg&Fw2fv9gKYb4;}pF{BN@XeI$QDgeYN`Oz|3U za}mDqmp>r>!~boCgo6X(GP0sho48?5Y2!6u82lE2+WTY~^Br)s7<{UiB2J=F2Co6j zy0owUDtqj~isW`F`8)`-@O(-Qbb4;(v7lci*eASYe&ekXGqJD12OwR+!c(=fD}-2gW-M}Q91in<6GZ!awR@9+iWc?71M9K`RP+s^U zl+W1zS%tb6^njy66cwQxfW|!p3RtZU2$b2)_`jj3)Z-(W!ST`%rm?>@JwM?Y8!GOm zdE{Q$6>+lQyt5cRl71 zFQ|~!P&gdTlh&B}eUD`NS5?>HPnqwP8bjIMCM!m7j9LQ)oGTRbfv6=y%O#mfk=(lF z7u0fnF55$=f`4&WmPAWJV!A_F+y-vfumii>{b&@i2TA6mhvXr%TUh#)B_9P(#k!$= z(S)#zB$;WC8lj?ep-B1YG_avgArixgj@6-Z$)rQbTY#}h4z+{v0h_;i*6~r_O%%NA zEJk#&u|QKBFg%uqoA+G zF+upNebJog*ieD&5y-LMUa|Z%iRV0^v*pXprN7wMlM3Kwln*bm_IVSS^h68)E!*lU z4e=Xs^nYqCvjY3)ybUcKku08H@CRx3!>^1rmV<)P0v5$$d6t94G1z?L=IA!t$)=b# z)ojgvw_msy`A_GbX+U;mEN}lTfv51U>zj^-%@?L^30zw>)0`rlG7s5B3|3|SiYb5h z?Jf#`&|`+tKSjBb=z^}_kG^5OdM@l)xC?|t+3IBRPlB-ziTZwqnRU|a*3q9t`oKj% zuZLDLFRyf_E1QBa1B;hI_Dm|pkA5JY>yxH-aJFb|k)2QE<2(Xj()%qx)my1wxxxAqMla90&_ zjFxXpY^2Dd4?N85A;2TehMS`8(o=BvdF1HPu&-agPTc!|CCc;vTCo2|P~`CR3$5|d zCP>+{_Y0h0^=2c(@(A|2!B3|P^1~?~YFE**zkaQD*{$5}eBZg`TVyuWRLkdXKBEWq zuC9`wI3KR;5f@&A!d>joH-0r#hPfNviM)KnQZteNM3t=!-Sx+h*7`DF$|4T)69v}H zi_`rDe)IhWYAfhA4i=UI5*z;`e5oJ(PBd3;VvNv1@wQv5W^j}fW~#AjeS|&}WBuiw z6xne)qxy6FCRr*uJI}=uc(zB~McK18i^3Hmx#a@RawP*wuJLBwK2z$1N#Rs~2NAT~ zx#O@nVC(<{S;_YV)7iX>TKnCGx}>Y?JgW-N4uVW0)Q=p0-~5S+U#Wn=yuW)PGhW0# zC{iAZYQCTKuKV+=i$*;ed+>YUp=N7!X3}$Iasp>NiCnpeKuUVOISuuGHKL1xf2UNmRAl_GFOgVhA;d z>Y-KbaSK~n@N^((IS};L>AgiEXV^MJjD@!Wl_Rt2Bt z^d*0U+9<8XKwUzshXt7- zs8aLWPJ$?+#WYpBN9Ld|F*Hk^*A)IHXYN{?7wOR!2EeBCnPm$@Wzyvi3sWaUmK{7n zioA%E7eSI;<^BuvS6Hb%FG>lG&1#&J(D&lh$^X<+^>leVSv7kAntb;poBOESE7j8& zv;jbJaez3#>pB&7GZzBs1oBT{sKW=U-6}E^dplM$R~{MD3OK)XJI1khWCz=)2d!|h ztavTu?Yr`v;_amBYYDeK_5JK@{GxA?3O@eA>`iujW$3l$`7DUkonQtwvo0j2T!Muj z7(2;=XrkvM$s7Z;*qb8}pNBbM8R1F%Un-s8UI}fq;pW|?d1p8w@5T3g0%#`J4P_?- zc&ZCON{es`Y{okp=x8*E%Wj$VPP=|H|t7~1KP;o`k3eL_9P;LYrzm}U%eu) zz^0dzWz-ZW#uMOo=yrVnhxoHNy~CFD^B`rM=JRzpsh!acHL^*d)N^jSQBmd)}u|^#xT(m0qSkcSRIDXWd26~ zJNGIx4(-wqPVJ5ADhXb-u?q}vpcT!A%ZnIpJbAAz-xzw?XlLU{D6ip_U>k5)gEqyN93#9iKpf7##xqUWV{+-;VZ>Pa- zP^=H6H_dqnkkKno>ip$4%%enpIz0YC#Zc{$7&cU2ZjSUYx|dwz0$-|HkHOLlS=|Nd zYXX2$Oobx^CDc*RGubp)K%#N=BO0)##H8?Q zTsw!_1gLTudiv_Ub3vOJv_i)lnX1k>RfP4uM4UA2bvdW7?L^4V=88<_PQEy9y5nt1 zXj{DF$V`z}P(mA)zHhS1aWrgmsoG-dwTBIlmSE)@l_yU$U|YFDS-YQY6N?coB6y&dIz(>87&3hGc$wR`&}QbIqTBa){% zFjlDL9FTDSo>xN$aobD6!cHsUJG*TXp#pG|#>@myP7T++yxZH_6%H_H+8$E5r_zhs8NJ4KsUexiPm^Jxji{7jhdF>|yg;N`8?$f5*6OE^v z#xl-uszZQ@gnv1r+GWtKc5R$vxIo|4XXCMpjog&hoKlRgR88$_+X9t{TK9s%l;jqN zoVp7<{Q8tN);(rAgN?uem(m8ZAtc{^F=isB6vLdm$24Ytjc$Loi>J|rq#F2xB258! z5rRYFP4!j&-s%;Qdbj@&GRqghxjq-PiCb6>B#JN%rt93oud7FsBjbegn};1o zxlV|wbhtXUEHX5MV4J){6Hk7!XvUO*P2f6PKm6r^PxKp?D;$@M8EG^QTL5*<{$06u zmM%9DM}}Oz1dNr(YN2%EuB9s3ngX>$;HXJv|K5_%!7uFCo2jKXP-JcGMUQf@5W9X| zB}Flw-9j~2^+4u#N#CMe#puoqe`Sd>49ZJ2?7H>Hkh|fF;c4}m?(~An4FhG_fTf-U zF_-5F6}n`{8CdW4rG2ZwWQ&PB&wJEyp`d58M-4f1lqbsG5t3m!=)h<9A69S(iyPuVHzE{p|ba?uH7xTiE#Fi&=B*4_Ww4n?vgYE*8!m zcUvr87?3NooxHgwbGEl5KK^z7Dtr6th!<(7(Eal+alkB5cpii48+}|jn1%3T%3)m? zC@LAfX;?IOA6-yn-ou{gaVZ^^8L_jwKKdr{1rXaGlV9h@2y*>j$a(3w!(m#0&j8 zJZ^47#0rj%xU>zKy7*~!G|YPREuwp!uf(~<&nd-LGslb8%sw{Q=Y90uHIfR4h3lh} z9aO^=j*-B6J9m1wDhie}3@xLexy;(VDdZi}RzFwhmjfnYm;0q$P)t zars$4YrcmhB`%OAigRV&QXOT-ZFuk4q0rg60_4`)Zw9A|pPSF)otCe6?b=R<;W_}s zN2gq+KA6e+URQ-9Pr6n~f}1?!>`ceoimC-m8FI!H#e}g3I`V3H9W{@cz_~D}?+}}F z4*U5cqNSKpicxqav#PCX3^*uMe52)*5y40x8AB`ZWx%+e^(^gT|0UjN==;C+=zUL9(M(Q#C+LxkJ* zT1Ca}FIFR2(MJSk{1to0t}_6GNyT-V#Uc{OSNceC*Y{qgH>*(#qIU0sHL6|9UNX|- z1QXSCon)p^@fJ()+-VOFU_q^<{pK*$LVLQh>%eur=FHh(&84&_NxF6r1@ht@4foAS z8ncwCtVp`H?vUJ?EsN+rcnI5Zo5%p77=H79@LI7!XO!PVOOStZPyND z{bWu5A9v(`&(6=0dhTpsEdr;uj8RB=gy0;YZJ5n?{R^Bsq}*<10whQP5h_QH9Eqlu z%GRsFt<=#Z#uih|ie5u1Aj0SS;c2*rNl+vFT7%ni9nC2Q28M5?QC7h7O=oEM=?*L8 zxBp0(23sQ|)rwyq4`t?L%cnah9rx4he znKZ9+q!d#^QmnhGuY8V<3c1;*8FY{b+$O(|x7o^UH}g65^0Y}UaBcH_`kANd0%EZs z3T-P(*Emr2$!rJjB~mcF<7tCY8rgiZ%A;?qMV zJjAz40f;a?={^UBoN_pAZ{tDs?6|FUhs$L9H|Z+^B`H;`%eC~ zVs~Bp?~PmQ8j3o8IcchQNITpm+_kIsaH1LoQ>QCxbYGZ3n&@t;o$W&QELF~N%cbME zv|vSXu&UJ>8AcLLz3R$o)44%wq2?w8Q!LA>0JnN+JLT0`51zgpT}07rwBC;XQ~L~Y zH)n+o7ylTXFnlsJP8hD4x^EzY*jiM7_vq=EG~ILa+N+LH?mNq}PG3FAZ|cs@&C|r&dqF!o z-u(Q+cZBmv1vzp^irEpu%D4E1fxf?#x1WXvL;@E9cu8L+2`?+s`x3Y!&C4X^(p-I= zekECfs}H17vFbyZ>O!csyjRqQ09UHJ_ z+qJq_vD75CK4iGmD-|9RLRBDUIdGj#Iprk?il^Q^TE=}KI}#A_LCubz8hRP6ZtXTC zl@87Z`4TG~rPAivT&}r3k6y~I>F-crZnRqM%6i=ir0ZAR-Chri5wOtki@A@EE9!Tq zrpgP_E1ET=73QmKDWZ&h#AC`Vcj93cRwh?1bM5zx9@|N^p zuukMf9Q_lrkfC*#4_-dA?)P}%T)-`9fl>nuziCq$zCslHjzFuGzY>68%&B(f8$YFL zJrBvA{!Rbwowh-xMGv%dL!5Adac@?CN^ptoA`0!?n-vXx^J#B>dEYm3&so3jaMy#U zn9GsQw3bV{Vv}>$f&-&$bl0E0+pV2aiWTf1_3Rw=aGp(1ZGHB~0Z(W1{pptrtkfsc zj{WoiBB_03Xio&OYjZxYcC1V~AkLy)S7UG8dC6clLiuJ;A=8|{%e3X1PO;T+hI_}t<6isrN2}Ak zYdoJ)mV1MNu@&1S>a47m9WCmZ)Bo063cMr<&WSvIGkOz1hoa-f-9P<*oV|5GRPFW# zYGI&)g($6*fP#X6GzK*^%+RHjba$61ARW><0yBhkhoW>SDLE+J&A<$C-#OZJ3$8_gibN_gT++UM`A>oXE{W!ur`u;-=Rfp zERWuIY`KA>)glqzk!lP<)E546V0Ly19cpx>*4>m<&WuAaOH}wMcbfF0E!TdEgx`9~ zU=y_ApLeW((vP$~-q031RFW({-L5JXlW{H5S}US#3+&4>j7y9Sl(!g>ZK2%$-$#D} zJ>YyT&AC1n{uKRWFzEMzyvQqMkDgmkDW4QZ3)Z-_DQkO}aS82N0z5#aen{VAK2)7; zOYA=Kwu+RG0b{3L%OwIy(fg>UajAL=ZGM@02ELG|0D z?;f~Sw%2V5gu{?ZEg)os19AT}?GZq1eF6Tgayq^NdhnfB0-I0&#Xh7mH(NmeewDr3 zQ!1wnz-YX3pqbL#MNw1(Qi0Qr@<;qP-D_d%vk)V3hx^Tx;20e=9gPs?J5v|f zML0ds?jF-i6e!?WmL@JhV2XWy%h2aHOe!4b!uPg^E$oFMWeboeim<8ou1*_q=A9`D z`C2s@S~br7Kd6!`?-mhh&P^4`3YD-7*g8#I#8UA3ZmqNAH=mqp!;-wSPC_Z)Z`QpX zeA=)iW4a?w0!u_m;gViAf8intn|m_G;eOKsO{ZKFh#ufzw!BX+at_I_f4#g@LRhw` zbc?j+0y&pd%@=r_ub&^-0mG!?4^H~lmR@60cxUF}@yJIZzwV%Nb77#10&$XycO;6_ zL@6W{6icO^Q2SxNtP@&QwBbaa4>4DU!ppX-)d@i63~tOxI_oLItKPj!4=zFk(0d!*uzhuOY4sR9aPNbD z;9!a4+7%`_nz}Oc>6;qGFT(~i4~1tgEb$yl%geX2P#@uR9T|@QPsHVUFQq{a<09!C z%fs=^c^e#*F4mo|tqnFZrA8|No^`6ob9XAl!b>rJpa?AJ!F~=kMx>pD%2k|<+k9~Y zORa>Ob-xlBkauJM$evYP*lY4iWOv50XK2}nNFlc?tJ~i*=fSSdbhxx{%s#(7|4Sv2 z_R{hM?J%o)X#yZ2@)WXKm3H0Kcw;O_`TkuRGvQQMa@R8Nzp%=?)sS`o(PV&Az_ta^ z0kAmLGB(x1UMjFJzQD)BHO-3c6Y;i zzxqUVV&e*zp&Y_~Nv*rg!traffMVGr3j+AasMuSw+$SnOU~fjS39!te5)(GIx{;C5FJtxxUi#yn z#}y`G4cluoK#3GAy%x3ZV~VPk-d-N0^+?8wu_n~%9H{7N=cq|Er$F@%4{{*7yu#>g zLZ-HETnBc-EM>~QBumqsE7tr=yRRDAW-Oo5Fs#K*aFEo-$ixemj#O*9AcH_#hap^XcwT(fYHx9P#lzsuN zb^dYgovS?nrketQ9NNhhy*#S7F>TS|uAR!PIW-NIkYbc`Y1@10cv@)aHw)HSu=CSv zf7b#kP<#}WS+%2$W%-|nl-VyahV$55cj2pK0fk#UERYhT9?@kMt&y2)d73rMM&90Z zF@$V$Wu5S}^ek&hx#L(Uxe;!$P|S+4nAMKhZ74Mm_Z6 zi9HpM8f0_T%c7(2sq;FBsT}HjyX|DyP1iGh0k%sXS$KkTERPKr(0v1mtymakXAenF zLK{p!?=72VrC!{5QdT-}lyCX9=45SwnApCe2npN1_t)>Kz1#Pqi0r36jm%Wz7WsS46*feYlc6Q}Oqu;RA-&`I zConE*I4~7$xAKs92ypCNJ4+iDe=Wm}UqSw3MjB0wU><{d3g!bs@M#!St{>`rfNkUDZi)F!}Y``R@ zZuVrx_n_tA;mK~>s>zB4Pi|H=0Cgw^T>fHEd8OQbDb97HgPUD~z(gil&@g@`uB{SfzX|U9wF#H?Z?*$3!wk=>PbH?!bKV zI}f+~@#&D#*5ux1D@nFmZrmUwC#4r)PjIkHH6Y>@=hAP566@~2O2|xd9@;#&nF2-C z?oYO{)!nyo1E~BY*v4=zbW-EwS4@<5pSNA_p`d=8-I=Iea?(jM`fk+T?gmfh_yF%? zrn+~tzWC(n-(7t_Fbd-+pxk}jjBE(3=;&__x1p*QDv|5z@9#3Z_k;TAl*IK7w zJhz3tdE)}E+dD(00JaIsnllR-e_iVdtSE#9W+u)?FBu&g8mej_aQChht2R+SIOWXk zH0wFd zrKs+@cJ7zozvkIbU9kVmb*DtPiD=Gp0dpJs)Qc!Aaesl0Vp$=u#Xj-^u5Dw z+$+xGzL$|9NqQHA7bPpPB$3uiH$Z~e!D%Y$x;+WK15M{moo;uK-Hie;m-FhvG5Lu` zY16WcV;H&9l8 z1)LZt71;WcPVV^yzSmS%usJv+c+5o1O7?r{K`(IvL1i|DWmH|=oZ^pE|7UgIRXtW_ zOt!k`Kyo*3`{$PxMHDvv>hlDEwTa{4!c+_q4CLh&SB}lCoN-(IcpJA9nnS^A63`Vh zNr9H8QrsX%1`VD#@&e+=ye5H>s_vu-BVmOc+rE+biD%%(#Y=#1HAGkXn2EB|=xd4> zsEY>PxErHH(^|O)7J@|2d+8<5AdZ(sBCzbaZg}7v!efvM-LHQ66R~p>mfDpj0_D`S zY2~kYolWa5$Sgg?l5v{f)ztDB%@LdNuYCoUee3e5OUHBch)qcK{`PSAJ$0^-=e8{X zY!z#1d>h|hYuuM4cPlT|zV7>J9Iev)z*7hI)p@IAVA794dU0(lASu7rLP?(?^UX#G zDQ_`I1-OFP*28zgZdSx!dlLjzWb?!nCztktl9KZba4vAMkY#%z%`F5pyL-E|NFkD_ zy=H&=^I5cfSA}60RUt@sOy6*rVlo`>q-5_$0P8_5$O06fG^9`7hO1a^H)A7!3 z-zOzf`r~g_)sM$`9^`eeUT)A|Vl=7Q9zb!%3c6L!t|;=5&Y)FotK6Av47Lh^Pz=%F z=q6eYAiT67d0aaDqR6`H6#Rht*s;UIp_puk|48Nmg)zr#u;Iovf4M6synM-=*#bOi z?$9V?aL=h+&X6cLLFnE6IFu-jM%wu*$#M9cq*&!^&u*8M&`T$L9Xii(EE;g%@|7-( z-)Q`YukG*5sZ=` zSqp#m=E@qE|EyrSMuqjarTK7o;3#Zc9(|((rus|^h=pv?)E*d#6C8%Y%ehK*pm{A- zE43^iKYpytec?x_P^zIS5yR@UxNBIJqfr6@9xOK>hSbso3WoGrT3VRBu^N}|aw}aP zkArH#d}wwd6sZIwBw@YF4N{q$CIhWN6C^)Bo7!XVplT;W$|w;KQRD@8Gmv&;DUY8r z!RC3GR6d02btQEed-Qra`-ANKu${@SjHFZEqz!;s)pQRGC&xeTvjZG|K>#)bP+~~x z6Tcj+`OVWXgi_xOSm;q*f(y?ziv(0^O=CBg24uP4-r!MatzPxE+pYIH*_|eGeqnDa z2HAX`EixMPWXsJ4tB=Iz^-#4QTU)>|wgUM!deuTbr9=62v9>*@jZG(EVb9(i)u)gs zcP&=!$~)?%<`GdWN-AzqW*@x?ayNh*oGakaZ>0n7xu`FXsJBWc;gNjVE(X&I?oS7- zF^2!n?0ax3^|$-HSzXD_h!4;_4UzS!E=C;vqy(wlhN^N{FyXPD4wYzp_%`mn+R(rm7H7eiI*xf&!TF{^@pI5){F#}+H+6MoS z4=#$HzEPhS>G2ww8J>@hVo7T@c>_bF3)Rw2|5vW!MYg|cRvIl^+ zQ6C~|i|?gN2mf;alHyJFNftMZId;4I=TcdRr2Mj(M?OfB(OA@(9r78tY=8ug-gy1v zN{NmlgATRE!{j@u9*4b;fW<5_UFgrH6>#1xuF<^Cs_`;sz;YA_7(a8u%-yyh;Q8G; zLd$Ds0bA>VZj;(!O&HiE9_)2@?X1LG&vcq`sbYZxZyx`s-IDBKUWm2GZYKhPVKl(A z-e4?3HA)CFq?Zt>FCKQr+K_I!IawRL`L;hTEpZkSdu@nBJB73^HDTzm`!_(P@EYb5yhjg zXHn2nv+&_tx7O!P@_BsudF`BK$6PlqR@H{^?yjIb;5IR=N;hZ`2h58bQz>lpV8d!Z zs&P3|;iNg>wF&NwpR6jNP0G5|>w6BxbLvxXUG>i12Akg2{Gmwom*r})kgg2#QD=Lz zFh4seBd2*6m4L0IbRzP8Zy72$2H-7Oih!{me&F8mHQqdE^=Sv70Wis&bOLF4PH+8D z+L5(YsNZ6#*(hS5NmZ^DhI#_DhdATcQ0}wy>;g*Ie9m!k9)b@0<|ToBovk#$%N^9a5`%>}J?bM3v?FDloX zHvYeOiARV}xUV>x9^WF>Le^c{ja!RRCuN4pDP(TDOx~EUW8}4)FV?7al`tQ>z<$s@ zbs?P7B>Lqjw)rEiEuaG)_f?SU7s@P#C06jk)uj(w~ZS^VJS<*|#6%(PePm_?vak@l5CAe#65oVYaW^Xnu9(KFy z(a3O~Ci(76a!q61hh}D>y;YwEJJ@b^)a7XoBwzOZxI!-OEjB?A0x*LJoHIVhts{;T zuHWVm)T+XhQ~LT0i&9>wlc?X_q8HIlnlOH2p!eS0oF(|4nM-6F`!m2I6;V(-&lhmR zntdj18{8uLCWoOK5klIHe5H0+#=omIhAlrXH0AuvhF&3NBl0U0~@*k6ueeNx)S0VzjTfn5L`6?!Jwk$hIh);=7k2x)n$C zT(WPbiP|F)WR2&R6@x5Ieq<)GTj>98vB3BbKriEWJsvCgi;^V#36M>vVi%AHw1?j> zNSck5f8%(b(5~iZ)Vc8C3*{Ob9fRXRfui`tP1~`IPmuX!Krs~u>D2@W>u$t>QbcH| z&V$11;jLMo5p)t)7;Dl7GMJxI@;E>1OcG%HSmqr>D6aY}Dne=H>VUf3*NZ$3%iejj z8FyhcjFs{16|TF5M5OTTjflHmUb2%{MrU5VdNpq$>~`Mx9#ENCkvo0|Q2^DM0r|$C z6A=n0PI5>BU^#7m*|H-F<=rsObrGN)p*@*hvnx;B@`2apQXV~&+rKCb$~Ahgc7zHh zk^}`j+R(4tXV}M}2Ra(WfGU?Vbj4`rWutvc6D3QOF>D$(1APiP84YQZF zhbHsM(-(R*dy_@dMc>+H^=B|U*62ew6kJO7*RU&LrxCVi+>bo!Z?oVu<1YXOj`g(u zqa*e~1Jty&M#WhPMYFI-X6t+^x6*Ffmt1`YzEXclE~<`<%k-sRN)mA1fIX~4NWyHY zQNwn)k4|on=a%&ba_z=Q5bs9d)$dRwmya)aCf_X*| zEuF-#+AAbmSYke^)+=NR@MmFHLes%Q5$Da>=YFZT<0TV$OW-C0vY&^C`3hN!kB~h6 zz6UO`;2z-@^%(EoE?K?>{%zndVX<5q;@n-XdQ)hIRr*)- z!9+tN96)l9nXPs+tv6S(%XsEDO}X$fuW~TnBr!w$=Z}!Lo526J-sw<%9nE0{8(kTJ zVL%s~b$fG#K6kLps7DPFdFOee4X#1u>Gj~F$-Uk4;kb|^&xX+);KnUkKo(#0Bs?AlI+aS|8NNOi-^jnc z5(cFkKc60XZ2eon`mb2~d7l&iroVM%$pV^aUP2*t`p5%a!2h z^~(OplShPs7bU-YLh;DU{?9%-`auYeUCYjK2nb z%cCQCqR-yoT~&nj#gBC2@7z{A6x{2|WZ}93{$}Sy>NNo?)clmBii*mU=Tic|bE6|K zPUCgX(aA|dO$g=ahUA!Gou8R$iJf{g9X458An@(%$kFm+a>8;+xp`5`*CRDH0~h_? zT{+U%_&55SE1K}TPV8$NTW^#JLB&pMR|9_FM>KKmD6{Z~Sc+#=**p;X)4s_#Q7g#w zd6-Y6x)#)@IZn;|z`w{c>p1J!D*l?r_U$0s75C^%=6%XT&&DdPK&c=quwvC;GRJhmNJR+KL{9L(>T&rHK1Dn+HSPe^RqoCCO z)$0aYFYw0b_xkHBf0$=XOpGJ_R`|6>A%OHmWHgRxa(gQ*SF*R`U2XlnQ~V&qaa~U8 z#1Z%KTdVIiuE2k-`WQc6?ybHC=cLe>`+UE3ha-ll9yok`$Orto60Uy(%rN7fKks96 zVE0=ulKL^>N2fdl+zYLG-oX3!T);1XclH^mD`RG+%){S{&KVan99%tyU!=cx#s3UA zO8>jpdhm)1uc>ieIMJzyw}iiEz`dG4N$Xbmi%oooiG#z*;aB>sHwy9V;yZO^z{fHd7mzx$ll`HMrMaXI8u>yp+6v#(SA2mVK2;D@{ik98bSH2#Kqt&eMED5%5zWO6m5SDZw`~ zjS<}5S?14WbAGDhN)+;sb>U~2r3q-ac@4i5(Vv=%4_5JZOAe%MCC3#fH;5S#%9dI*|mWCta_!+1DRp4>F`cGY;V*r?$vZ_qe zcuwCXHZbRY`X&1R@uUd32V|!TTFUYDtTYe6q<}~vJF5QhW}b{-X0A{6=`Q1EQvt4# z{coI(&O`jCe{(f{1hfb}e$A$OIOs5q z))t+=c?JA*KXN6UgWH#Is-4Dfy~NLgdtLd&ybd<3N51*bFBy6Rr&;m}Hcg3NMK5sm zx{bCYe=4?d_H#C{F^{IEpW|O52`G29_9zAI{^39hzrd*&s3<3)@r`^P4Q~JP!M8sR zbuPVjz&W1(8)<@-wK0F6``6(98koO#aO0`qE&Q)j9I@}?KsT+`%lGu3-m3l{Zh21? z-e@zwYJr3^((`JdRc4vAchs5xD8&7zCDgh$=Hf zz_0ILqQ))nr-_DtIFQFqgs_3-z1C5F#7SPpE$?T69mqetnWq?TdCyMiuH$D|0k9s- zrG7W5y}pyEe9%eNsI^uNAZ20DZ>=6*_s>P<*u&E#E*0vrFh(? zb4EXCX-Q0`6AOh-wnPk<*!)SjA#ULCYTI;53FYE_Ke@uczku~FH za0Vn1vbVSQ(-u-P-y4zsN_9c$9^53RlhwmM*?nJsteQZf_~qqTA&20$CWkFg>9|^E zm+fVn8}~H^@go1z1AXyv?B!N7tLiH1;LX{BF)N=BvIVxJHKPLv>rqr?& zDz^}O{P+R8?Ko+FMHOYQMX$>Cp#x2;5iD0h zViTyQru*BJNP)RmZ^@_T`xYv#nzonNU6%RXq~u!1-f;YywD<}AKdx4#z+z6SFmK0; zPtXM8u&B$n6oS(L8S8c#zYk8J%4<**puZEv>gHKJkeZ(4(Y$nt$9DF^-Mux&1vfHo zW15o~bW!I7mPXJ=L5Pf1IbSQvFyMOBjAV>8k6$J%vXDTf*459omxsR`WNVaO>Hfhh zLUFHHY28l#s||nYK-we-PBs@zuv<32+;0b?EPv3mmA;jw;Z?wd zamaJj>h=A}>>hTr?Gr}Er!okyH)m?E771!QyUIaonm6Y-Cfja3-Qs&oDF8tzxv5ge z3oVaUF)s9*g}NOcu!x0FMFFmyc&ax6T%)2eoL=&JQGCxn$PG=Fp`u$_t};%2nrizA z&O6dO*M1i+853=oe2=vyoS|fy&t^IZw7-w{(0IqP!gVen(&>kjk=!8(Qj_wU)AY@j zWVU?G;fRFA!75oIsz>AZpVjW|PQ~7qFbKWI3KgFxM}^mUsBJvx*K#`vI+7y~9!A<0RW#J$12~6myXQ&izV5gu zv#5SDy}o6RRLbj~VXNIqd(gE|_`KotE;iXC$U;C4XXr>+V8z(}+<2ZN8_d6VVaJY_ z;n2wIZS{0`FEaw9G9#<th%N^zNE zdW<=UT@`m*$GuL%)k>HRB1YXoT}n4nR=bL8I0DUQ7amNBDv49d+tU=>nblq3NBEd0 z+XG2~v0y}-t|qzQiwvna{=Oec{kdw80oFvyp^jG|BiZ~qEPV=V)YHLG<(3&tnc;@D zC%SbL5Hvr?9jAR4sjif#V&ss_dsEOgt2a-zA$4DZygyH!EqQA(`{~Qh#`%LeM!;^c zF^sVto;VFZ0CFI_4$HS0Wsr}{Bo#0dGhF>zOQngL(nX14$oi0j zCCHU6c)$PUMmM_bNKlw_&t;Nmjn=9EtC1Dn#zkH&bhKh+v41$#MGNK0iosW+{R=Q0G8QGKALnjXQ&t$kaG-8}9H;=DdPsL2lee!#Y z5aibHs;wz>-G#{CN<;JdohgPk`S=sHfyxBLioDSKj(6F~sR(vh)@nGQc!1O-Xg3sMzN1h6L4hR2xG#qZ{q|7ZxY{2flTjKXgb^MpG1Dc=MH|Sq znnG7pM$e9qlv_PD=t?vk?2HvqOuDDlzPSC_WjHi~EXOMSm8f6wp7nmC^k9WeCeh~- zqy1ppISMD84*Q_ocy0*bpVB4{jJX{`CKx($7A$Jl;wB#?$WRHfqGXe^hhgL9Q|l-? zO<~Wzm66>Ig>0a^ra z)uQIO5|g(MBk*{X;}UH5XJC>6X^n%3Tq}iKjbVR|>f2TY^4!qp9dUj7ldI1SNVoBk z$Uj|;%O?uxH=xVMyZ}W>+&esp_YbUuQK1?e_`Hz{Q zApFBNRs>+ec=Ns3V z#K#C~2_rx;%Fw8xi^5;tIFAC(EdFIm2`M8$X&V4WOMNQ~=moH^Rh}iVA_<2$E!KjGQkZ;p zZ43hmBVvcWeEcV)Fc-IsK94KN$%SSe)2>mTm+-8sa|#?oqO%msY9Dmjq%9iA9FA2| zJ-DNfT^g-%kX0))X9PT?zJ2sTQ5z-+=7aQBh+qhg0VcGk3rB?Ms++whv|lZr%!W0~ zo@|YMO1*&(K>x`+o*EGY#t@rBz;OhOc?vwJdHJ1!`}DHsAydss(x-_J6eN9peQ!KY zNe5~HJ(HhX6f33uWZp|B-UXG(eKXW{%V{h^*%2M_{gQDHjcluSb(=%;a}dHT%6klh zMfx0UZ_MEEx8(i2zM46~@)YGRa>k{XPcIzOT;S60c2Foay;)VZI}vgyk%<=SHS;+` z9EsVQR~-N@F5<0An@G7sef?v0lWmPrkA}*=a8^y;g|T;3?eS&8y*7PZrnAZlnf4tw zE|QF<^a59yA=0LVxp^6da$RlGtZXgRXR&=}Iy^0H|B=2Zq-dWTB^6pQ(KsiG-QF{^ z162Axt}%=9wWQMFv@d~h7xXGbfxJJz0Qq7nAfBysH{HH~Lw}O`}aPjinq%(Nr+z}*;m}FJpjLGSl1n@LxUD)%N(?KgspJcYkIivRv zDMKM7EUjgzAiMHWuCwVUl)8kphvtjM{i)%~D}tD6wJqcBQacu>)+}h$;^VVz^EbYD zfhyHkMC7cB?`3){J{Y34e@;#5qq$O3Qx&S8F^z7!XUcRJOjPyTf#ZIJJXVT785x`z zHrCph=?F{fy7qwWi=&X`*BEx2ehNy+P@$1e14lZaB{)bMFc2f$PoAEC4*QaIjCK)@ zkjh><*uP(cs0sISJ?$%oOmq0p;G5&r#D?G@>rC0jkDb(^33xuuiYNuVXJ=%0=#wr&Pd|1&42S(Y+8R3gwIa(w<~l4lBgnE zW4QAtTcZ@Um=Dt)nvaNwGs>E0u8(Z!(>=PQ0&NB&LZ6VU-|Zh+PXaHD4KOF$n>v<# z2-ztxK1)a#&bvj{41#Y{i*Hd8TG`rNSq4ySSSk~TMt9+t;qIlLJo|~y+Y-6FTR{Qp zO&NB`=8(%A-2!yALC=wZmAswIp|S@kw1zH+5Y=zs=jW)#1&6_GRU zef?Y^UDVD*Y9qz0YP-ANc?YFw~)%hRY~ zzRDmK`bE;i?rc?-<+u9dCqYTu%;Iu^fE7BWmoImyIXS*e9QrYRp3}JHO+;pfaTGJV zeY^O=Ry#1k$a|`Qr)f^LC^X6khMkfxO0vWePgP#`laI34nx2=eyOAOV`Q6TbV%1eH30Ep+Z6Yf0pb*(SI2RzkrCVH8vMDw6p;#0Hl?hrlQI|IhPI}7 z?Ne?Z?4yOi)XFxe<}l6xwA?SLA;!Q_^z_9-<@5jncB{Zxxi8XPyFx0QK^h@L$>l%{ z*_z$;+uPi@EfFIq%$1`#ao?kF6&m)UvSwG^LOi6aQr!*LpU2Hm>zTOz{JtAgz1u&G zu?@Iv%xvNtpJW>-Ly$AePh0soJHN?|WukUhiMYaKhD##swutSk0k`%N<&S9n-<*f|ckPU?xm*(N{xE=5L*>OKwocjz<{Z_feR4?0IGPb0t}cQf)X*+dXVK*)6ZjLvuY4U^eR1 zLMe+iVNH@qGP6_pjpWC3n!POyEbk*ypjBMCQxhRV z#qs%s!mzBpLLy2~-ti2xHK*^OnTWIZLk^321GRP=qDb%PQ39F&-n0KL?x>%_^|nl_ z=aJq9R88m;wT@;Ked-zbY2ur=d){|-AL~5vV5M}0#$>?k>p&ub-|dKaey4y7c-T|sY)awWS)N^9!gS5=*?IEOn z&pUQ~#05%0g)(U_#FMC^Y(jnV$oVZNS*LwFtc!Y?5hFSGJ`>_+Z=;_K_hi}i#4QAL<1gPZH9vKx?>abK)-R`jZ-sb46M86gIC6KT*c?gp zWzZpQ2*5ZdzFlLA{{5v+J4L25n~H@x2O3rA;){@M z7a;D2+s7f_)+Tdlf-os{Wq;fU4R5IWaf@8JItQVafgsIZ8Z9-$ZN_n4vkOL$b6M2$ z9#G)uhHo@gb_^23*0No{K(0_+y93m3b%R)yvExVmW5Sg+1#okbOs0cI7aY8CSHHdhA3fH>nVe`4w|nH3Km0S<70Yh{I1`swbf0n%`O zCu{k-r$ERv>?)(|+}GiuJ`a@@i%xAoW*IcLJi_t$udaOlQV(E-qq5|6^*();=!e?qb-jN| zZQw)LdiNc-7|=A);hNi@(cX;q6FSf^(QXs1YG-Uq{E8!s!F&Xaf##*gE&Gv+F7&Q!jfBUG)n-KDj7xeB?L zt={A2tpKoa#8diz2UQI$4+YlgO?mICRjXjOs9zE|woAg0&~Gq@1Og{_)#_*OWs{%# zskXa-Ls7+fYX;=!2atxti<>HbO4IGJEOqy^%?X^KVu19NU9osz7~(pVTLrTcFUgkn zYj{OOb(i4$^|KZqb_Y?b2?7pD*VvS##*B7LH+Rxv5-o1hs&KmQ$VrfnV_Nkdi^B82 zr8hMP#TJT34lV)>25SFj~ zJD>t7c=7rX1?7M8eD!`ffA_jY?ufsW1^(`nde7$rQN&Ou5M-K~Ezk|;%B`Dhi&s{7 zzv)ER)`uLJ+>GV01}Z7(w?jgJP{_-@2$67lX-u=r1hN5C#lv5&g=5xM>=y3;N^nR& z6e>P5p4#v;itY^#kv*}upKg~tZ`*bDDkZR}dEgK?AQ)wDM3UXikeZa;KMksj8iz6> z7lCcFc`p@a=ke2i>tbS^<>O!~9%gWMLmKnTY>QinmD&E>OFY(_SqJXxQyoB^*iSj# z`De?s2cCrXiT6Lr0&^dJqLKIX>C>X$)Ir^Qk-9A?Mfwx%C?)RpFIuXsW3; zD?AJ8zXb(H{^#`!2L^`SeVBhF*N!{x;(^Ts>#32)!m!uI+w%*~%zp)eF2Zm~g=+Y{M7w&x1+|Ww`KipgnqbwRdSpJ1G zN|LgLSBuiqa-F)3+f~M1`2EvQ9D{+M?NVhiIFe!{29NuBJW=fSrKetKp&b>a{e!B8 zQR%pc(FblWe8oMUo;X5``o-UmW&d}w=wZm?dV&+w4s*#)kK}#-tKY!)vnM(5Al4IKo*Uz*`Wi@v zG9L^aJ?iZ@6ZYS5XX8=QWAI<6Qv5ojMf5IhjZ$fh%fZUK{fz2%VMt)8L z>GtnM`g(uZM&bG>oSn&TQNefqIT>JQyww%7&i-omk%jg*|GSa}UgA{chid$C@-zXB ztY{3A|0jF*f1lv>9VgX3{3!^JAdG(lugdSd`;Pxt(D|z~|8|bi-Nwy;mjzBolKxHL zFu|=IgAx43_-}7U0dG;9_+rbYd$S80sF zyUJrNNq+t3|MtxP_`s4B{BXJON;@9E{cpsH-%Kw)_0?w5NW4j_*|EKpl=JgWC>A8M25kH;&f0qaRZ*D@o|LZH$M~7! zX`JF0z-+y8!0ZopaNqZ_KPQ$FSGBpk>V8mrXvU0+J)x7tOm^oyslQsY)NpOLIf(_*YrNIafk$4R2WTGn7XZ)MhK19BIYl$m>e<9Cm5SiJzIL{foF z`;}jFh9(Ke0l4aVSCETtvea~}R%`QCo^-qh*y6X+_(thD-1$j+)3Bf?~i`D6OypH_&+|qTW;Q!A*unbk^4aUv0KItbb_>0}luM^_q(O z=p&s@Rn_X`&kI>uzKXSCNoc)g7Y`fX6RRf8D zdsAgBT~tWSXwi!ey?r5Wcw2@d=5o#0*+||lp`5s>`I)>iyQig8XTvw$TQXp8FN9De zrS#IiNh*Kp>h2=T?{xZqwDU+~{#wa2NjT5CabndG{}~!rKsmN>HDd6dTICama%^eQg&Ka6+3=Y^o3L z*U`xuQ0MTNLF0z158tn`N@R?kH=G6@;vh!-Z3N)T2byIYITXRg- zUTY425(g7!^qLEgEqgS221OKsNTv*Rd49W#6hg1dr{^$hNFJ%S>O{4>pTAHzRi{&7 zxs-IRZHYE2a96@>af}EZdvTbUpzY?w{%!LeYEjv(F&@YKn59|vXhstZ^N3wRiK&P^W4ljm3HGBg9 zu%@LSm=PYn>#D!IyaWL-;DnfraZ{b$2+DxY@^=pnMK}SsK;-ecmvI}IW63l^F z85S+lzi;#OfQRfhR%R?3IwmnWmmsLn;@I{27T++ox2#legL4#`nw5OeiB3ElR^oyB zx)AqGD7zpp8;&R#yUbzw_}25n<{g1Si+1V#&WwlCk%O~`IR&gFNo#?{?5a|=0ydpu zoqhqB>jwfiZ*yDtNsQHBQ-HdyeCc_Vd)Hz86XjsJ?7kCC_lJbK1or7{cgC_s;^;`G zaPhV3+U`wy4+*lhsrEC4zn!7_FQtE8J25E+~CSapV^jMv@mNvY7; zvF~GLSq_5dE_~qT$oCN3FU}dUdfLme5F@SY`}z}Q$AU4c{cNJG(c(x`X)^Ef{cT?& zb*2aspY=dlLhNeD{8a(_8tZx}Kdg7K=)H=xzB}G(}8r z>8rQ>^BNoF<<`sy*}2uwqWE+SQE(!0lz~t1?C`7V2-tLf)%<4FPWfX7r;b&wvU{R; zR@2uQDNQ>TcsLkh2n06bGVQVNTwRPk2^R92KlsfoHL}b0t~7rr9i;F!DfOV*G@-gv zW8ix@s56$iO;G6A*2`K!@g%0y?Gc{S`wwRHZwT<08Lw8@?grRmT|M2EvK79--H>Gy zlNGY-p&<+jkt{<%D&xB z^h2g(B|)PrvzwglDNtr=HhB_vXi1WkQH10ViZP61*=2)DlT#jC!Il?Z#-1FjanMF= zJ<7t8b}rNQR8F(A4K8)0L=I*pDV97lu6Pa|+_hWaWT{`5&ZMX$j2`S7>C$!)sf*wh zGKPLsAaU-!4j|g)mwR$uIVumMMjnoBl;!?_h4q&Bi3&4RU5pCf`FqP^2G6!<+wmIvW((#d%a3JIXss;-|ImVCkGE;VwwaR# znz9~6eJ(#Nb&tuJ3vVM%<|BrDnHutMgjKx?Zz9tJFp8xl&|5AqgQxdo} z*OsO(nPO=iY@Z$@MFD(ov)_T6GZb4`K7V&oG4P6po%A;EZO$ia6KDk3#5 zn!1b!PvkChlbk^GPk6c$leEU|8nH^mGgqUmXsb7d#%Yyk?xxv&F6iC$xFS1FQ-vWvqHXm{=-oDkU}*~6AkW_G6hPN2l*>nOO2tFFT~a%Fy+-iw)D> zv;MrPe~l{!h)SW`ISNhH9sANyn3)J$cVZukQr++WnkxFIjd~sHw%ZIzuNSz9>NNpE z6*u$gX8(&zC+WgQ4NH;2;-gyrwteXoH*e=Rr(=ds!)uMBV9eKxhzB+k?A4T(F66%| zF1B!Q3GILIT2GLh_c=))dSfm#mr#-;NJr`8;@I>h=X?ZJ_I;HO1*41=bz&sdpdGtR zr&cY?7M8NR<83QE%Df{%c~-4>4sjull}G>?|vraeR#n7-c`I zij?6T5Q|$wUFK^d;4!L|xqTky$Bk{nax8U0IG=V1go?ez#3 zILu70t=Dq^dAj6-h9G9<=RcPsp1j&Te4>!K{x+uK89J5pZ7Z8y(6O)~;}{v*8dMO! zuP%9z<48(xju30xJ8jn4nQI(~8QYwg#{p9^M$iJMP^oXqw-29#;L;o_u`WyL^r&U< z4$@&X)LgLv$u7X``GP!h=wVulqsY676~27Q5ThTkv__B6r;y3Fu5Z(?D#>q@T!ffW z!J0u&labT~lSkD-NLHh)60IL;aFd_z;-eKZBP6cb@8(N}{b;H-UvAoS)iTJfO_Nn7 zW};zw8z5X7Z>CQxO9aVwD7-i~Si1mAXNCeU)s{(}(noDU^UV+8Lt3=Lo1H$#fLZyTbJ7Beks6xVW>In zN?+#vG-x#>W^-rqWJ?swy0D~a$AncAW$kK0yS4)iRr>YpV?FLqm%_|SKfZ3d2pK(F zYY2VMlf$z0Jvf=|t2$ZjJ0-H{jTk)%T4L!~a5VTU*7gtnbruS{W;*U~Lfr}z!(sBR z180f%yUtz~895MimK+M^&5L&-{}D4JdX}N^9$0{O`8Oj2axgMBwERV=v){5p2R{?r z@VX_Kt!5gZ6T12RHnPVZ@$G_-$h6arR93a%s>g1)+d$;V0-cGDkD`@uz3 zQ_-%f)msIa=h$5JxAGL#AkGtfsiQsB&UI(&`F_y4=#;H|(CVs!e*vRjQ|h??#YXuQ z59@3V{;1-d=vza#J^~A=sH_pbR`*Kzi&`&B@~QVWISlAAZbtou1DZ9N=x6;zEH69S z`d><*T*Dva?$@C1*3$I2*X3aO&L66twE9hekjB%K98|YjH;6ok;H}~^4kL8yzs+S} z&2dKsLj+1G@_f$(3p(fL@^9Gw`0f}f*>r*MT!gUz(G8Lh&;GJ2*Qa*=Ju=x|Y&ZWO z%Dy@*%5{5tTPP}^A}AoCNSBh*pwcPbpv2Hf#}Fbm-O>%xF?5Q6^w7=F-90eF!0(xJ z_SyKI>wM?9|9hFa=H-2#^{n-*xYxY~vc=i>*ns|ojX8Is1f~OJ=VxgyXc-0DiSOTo z#E^t3c?XMmP^gvI4)VMjJGfBaYdemMSj*W_Xgo3bL|@b$VpWnP;eRKt;EP)=Nm$!d zq-?}N*Bx8wTXBi~<@)gUJ>ehO&pR>9fpe{~KVII0O=D~n+9FQxHLxu>js*o~Tu$rf zA6T+g%p-kSv$%t959*dNW$0zZnu;T486 zCS^1<6lC!f87(Sr?C|xTyo%Hw?=h#9~!_u1Q`^VV73{BK(pw8j_WllC}2D z%5*hHDe0SjYJ(VEPO`OUZ(@J$7y#NWijs>kseV`g>G-8GdL-xwGT-b!ZBb%obpPX3 z?W%zB5T}%eY!d*J?q5pq7wKNv66wwoJ))U|dt{EGgY?RrEz%;@xY&ntr$-SybD^yn zj$Ur`8$A!(nte>Uh1ss4-ApIirujA#fapm`A%UOcvw>$vc0i<2RiUbV zBE6^+>{v3YzFB!6k^bSMMC^OU4o0z+58vJ@4Zb13*YbZt3k~?V1iK`-BMFxccKdaf=;^M2(=T{m9zP?4k0hUr6;{9msb*n6OffCi1@6rxG7*}`|NW-G{ zn~h0!Wn46DW+ZsiG&aJ_hb7BM>hvS>?|eiinCD>ne)1Spb#i50_vg&Q%Vgnxt8^zH z5z1FXAP#j$Y~y;~f%X4BXx#17x^sg4wti8S$XJgY4AJgkP!{A|Kq?+Gh zuJay-ZYjEX5A1pSAI!LWUi#W$#WiXj+y0Z=aJQi zMki2)L0?YG1rmm#P?5Xrvw>UY!n^*7Dy(;K(mLLRXW^IFDV~#>H@qkC3i$Sp-$F&T z)L;ay)sGLq-Q$C5_kZ=eOO)lV^jydUNWq*>;Z|5BBy0(I>WLTMzmxK2o?wS*Z%9-p zwTPx955?${4`8w)wnDvGinNg5zUTudJAzQ0i@KyqT#>QEMREj=-all?P7lR*+ z2U6#!gkp5nL~%BoYvEN&bv7=>xgdWvKZaaq3bx{jAK9+wAPbjaLPTHEl6n3&=nFwr z4@-&g#toTq4Dsf?}%n0LDTxFA%K`XE2Wus0~dO5N|N4^p)@+mmu>gUm__Vl-p zO57;Ivc=08AV24d5+M_&a&oHEFQ#vn7;nkYhWblC&$`t$kO=K{{UMdj7>4Z-7=R3t z;8qgR=Bwie-E#H7Be=;G=t1{1wc+b*Wl{K53ah($IsI1-Vn|%tqFLV`_cH}znk3=! zq@rJcicd*c$+e?phG?Uxm(3;tB3sFxbU>&|K>tyA{N^Fe9i!+M{!&)?>U)uaq~e1A zkX;TPt~90y_|&n^@_5wLqb8wud{YOT15rIM*KkK{={1}!4npdxHF;~sLF(WqX?@8+ zm5o+ev$&!(p0yg|4)=MgQ zN2(pacH}sv*CMKnJAm3?1IEno7b+X8TTzN;9H_7OQ<=4Ocf-rr$5bXB z&>LU7!)Q-CC`1DL3yTqJ&W~ z^B|DaC~Sx+jiYLKy%SAIM?cVY1;T=G(+!0EBQ)a9M0^d78^)375LywmnZ$)HWM^H+ zbaMoza!fQhkv&<=IR=5am2zrxLS2OBO8EBBkevE_QL&fMM`$#GAR}4<5%J1ZsSUVs zT0^H@6CU3!xK7XAJ2mf46%|u^>ZSBNttJwur(4O4AnmyrgVt4s-tO-1c%@#wQL}4O z#_yzkXcuMFqGsrQS&kP+1kT1)Y6q04F*rUk{k#!mXOeYzuQ`8$h6f@~QecW& zA!L8;^jwYO^gU&9oi0TsQ_1O?hTQ2v^QFvDjt9^Xd$RE@hJkdiIxt{GDB}*b+7XU z@qT3SOufNV>o4Mbq6kZMMQYZmh@0sP`ofsd(urXNmkG~={@<_EqyF$#QXj#ce|_fr zi-$t>TOy^`j*hKi2mb7j<0l`{2Kk`MC5neW&P=P<>|P`vEaYZTs_m=<>#UXg8T zAhycL6ZQRzFS}E#EtzTIMx4%u9eGs6f-{LtgqcX-X`-k^ZuU5R$rs3;{+xUrCkmp8 znK!Fg{~5*(gZ9^pG|TG+wFy3W`-)L0M9)po-d9^qg+HUtuItuAFwaw**z8oVb z>GP5EQwxdOx&Td_@Ntto&NWlJ!E5_&{ZIBE*XwE{<4e5R-B_ynbi>J7-{m z|9BO2^xS53HiNaL-f(fgY_e-ChiimGpQ$X$Aat&Wh03iexZWPWJxsN3tOQaW z^31x+&Wf|_y3RUfXb%Lb;MGQ%!P>2#k~#`R6yZZ^?@nw`Z(Y(RDs3IUTB-cMip6k@*Q2xhkAEQg`w2 zcMuPb89I9pFw=GO-*B0an;#zcBBh*mg>Z?fh)(p*m-RC%LBWd7E&DFSW`wNKKH82p ztu4DCV~GEoZp(X@qWkG%USA(d(DO(XZK1$PTN9%+7e(BPe-kRpHR z^CEWL@9mjx{?9A#i?kx3AAtY9${C#r)l1wP^x*|?38lPFb>R~RFa1Mx_9)pWg{?(l zL3mv;vtw2P${)w%h~s6Kl^@Qne9-0Te7)BYyz7KU+bGZsrGL`w9-|1QvG~5%pagTl zVTj_XNtFxlFk(ZgFIXKvL&@=5w&&_hxh-p+E)=Y`eE9D+=l}Uu?abwr1BpFo?n0c?^_6Pw|Fb{yn03R*UEUOGJ-v% zB)EI;@6umBxBRNKmH#Dra!z3fK9p@6M%PyRSS6wo>gj^ z+1~H${=aqgn(N99dPlBUv4CHwj3$7PzVK*W{q369zt(m}1-#Oe`P4^eOO`DMKFVJA zvlIBO=Q+#kye_#4o~$HZGvJI24@*fA_JvUV>REX7zZUCH3D4+)$8e!DxqS8%pEbcp zjqV5E`2Vs||9b;tpYI_J5N1g4$b!b10003|qKaSZ@V{2#w`E%X04SU8O(&-_P-&0` zR!Nv4U)=sKhD{0RGGnIxxqD~iSR$$5{$gznPyZ6hf7|oFtR_`G*cssl@|TFt78~n4 zOVc#Mp8kUQ`n@;x76EH^Uhb>U*%H5D5=jh`slW0!SDz{<5ql(eXMAisJCb$G!0q$) zNDNNTdW`cf4dk{X;u(>tQ( zMlJI3bUyPI-ukr6`*{G)+V>OE6&Fgxq}FW8UpDQ3R6lPStO$w>zyb5GRq_4pBuX~NPYO&aYiXXY4s%Qne^Bb0Krh+JV_Q+%Wa z#cungM+W_pV2<`j1roBs++Z^6@(GBZj_k4FTP}1>k3OwjJZW(`@z3;>$FKg^`~316 z-+(H78U4idulG84F{A(+xJ2bWa&$eN+B$%r*zC7&bav@D`rT?+%3tsyP7Qn7U zprKYJuRBs=;r!h_?xw!C5+SYoL!0X^?HjP}luK#0DdJAhIcoR?)%c0xWCwur-}Uo&NpL)z4VmQWl8qFgVE55dn>{%tWYYXM{kl9 zEOC{lDGHdfCKv_UFrL$_Xs&#fl|iXoQRHQ|us$`5c1htbWxkp!&feyo0{JH#)!s__ zn*moizgY7I-e!IX>52{K->iP7RbqaBavdWt9!O%r6BsDM&}pDrsAPP!Jk@14mOElR z*s&Y}Mq*dh%{B+h^vtBKxon^}K@(Z}i0#RL3^&e>zhACIOu=S-I#%$jCj<=C6jyiQ zs)gOk97DzQhgIV#GC{)B@1UhtPCS0S$azxlNIy?UDN*EHYYb0djz+B#=hjU9#~tZG zX3g^}U?$b#&XB&>QHMgLkZv~>v98%r9-Hy$3C0N%N~a)QcVaWT4Tu{g<1bSb%)%1@ zPm=V9>pV|FU)IEPN7aCyQ|mI1!{Co{Q!Tlo?Yyt2prIJ12Y;jk@(wJ^d%}LdLCs^- zLW~U{^k7YB)r}4whsI9SvV;(^>S$P3KZdPwd9$w$=10a= zIg#|mq;{vg2a}IpVOa(?)9nt%c{K|Ryw}jyb;oXpL10RQxOL^Wxa$(=HBD^26*g$} z;wpFJUaU*|;dVxX(d-AgqQx;myPp@4b1k+RR&we+VW%HBv>pe`)lWv` zEYGJ!mzmO(m=A^(8OzAn%|4pLykxgLyrP|dcqMEUSoD6I?(T?FyU57YmkkgdA1UQ@ z=>(z3x6KRiC>rDhG_Wf6=3i#$Oq2gVS68q>K=(V|U(T}f^G0EI3?Yr5aE6PQyH(){ zbyS+LiH9CgCppDt572A`y70W&t$#GWajWT`c6EhpRXLpmIbW8kuroeCXj?Sza>sNb zR`DDoW@3pASmHHZ!MVe|%upHyipiTs;Vv?~-4s@q7b*j9Ibo1C7@uHTe|^H8vAED+Karc<#|GW${XhW2t1k$YBO z4Vp=*&YC>IRrdFF@l8H3ueB7iXWjzU6gBW;2%1#|w?DQ%;93m5RAhpng~o=rMMgv` zr~#n}xb58~c|!l-uJnaiihQ+!i;i)TW*jENIRf&=&n`*V@XUGeIv?cdTE(?K+*^;Y z2{zg3F-RfjcOKs78Ve=lRL!&01Y_@F;TvnKeF{VP(M@fTG*j8?(Ot(VUVF9YH~K5N z&Yk0r-d~zp!EU9WBlk%0Z7s0rgVr$$~NCvj#g{BAy*i6fF2>K3rU59+1nd^w^0UvWd{ zXPUg`z9WU(K!YTLx^nhlZAYoOG-Rc6b=+0m?YYmJFOkevFyi^zQy~)cv?lHNLZ!?kTu?X5*<* zEDA6_A{M2_jI}1;$hKW_{es-_r?7?he*nf&*U1m9YLiWrf<)lglwROS6HH zBctSAe{g8=>1G*8Ef3|F8Fd7S(yedJZ{%AB`Nv-%uYDxS+{I~W+05_gt{2T^Vcr}k z43RY$Rj;axIDd)4*5tjrKb*H`^t;H7g z*f5<%-=~tZoQW03CQ+qe$3fUbPOSh?>s``b<6x`3kjCfa_5zF*v+SCWkRw~GRnO62 zu!^^^X^v#meQ4Y!W+alm-5=VUfm{{ZH~br{=l9z>x_rz@C zpy`ca8~Qt@H(e~{jjZ@NCcOm`Ut*RBe75KP&PzegX|hqD5@|9>qoYg@KYhCX^fRyG z%8w>r+IB^fn54{~$^&u+l5C4N)-BKhaYn5(H2-h`dVXB8Hleaf&s01|Ln^YTl|NA$ z%L_AUIX#JN&#(lVU}COS=uoKRh1nNcJyb9q%vG}*9UUpeqwFgOgo&B6I$3KR27@d5v{In^Oi9_2W&mq3N zU+vNZNsbz0RLgsne;S|f6w2oqA0cs-xagcEaOfg$Hhd6Q(~bDW0{-HwoP~FPT?Nfy zEz*UKYMMoAQ zgLdkY0`#M6LVPTC>ugB}rUsqYe~!;1=L`h}5?a=}a6`0GxgSNUxMYf$hUt zr}3j%yfE|8)WYgvh6ncHFoEG!Ec2Xn3Ef(9tTTcpj=zrn(#~RJf2K#5LV*=#k+<1n zG0wRKk^tY;x6Dw_(_zUS6M{5DWD^((Krm%cBkmv5mfjiTU*EvqV-#)&_IrOb3?iMU zgXHsFTP=7=*%FlW%%h0xDCdR*p)k3CsZR|&o|h?l#C&3W#M-I2e7Gr?x=t#ZGm}QQ z(m|++T*&VEXo=T^$!zZ>(7+x}m!-O^+3fWID2@40rp4-P?C!Y>6JfT5w9>Wi1Z@(t z%~Csat!bv8L%3f{bG$TLVEG%2zgJj5PQ#Qe;+NBr3i!-%zEfbjFd#xbuUTywE@;}x zJWwmwckx1ER1hb1hzTE{oO&iw<9WDEm$|1w#4lWUS=wWNp9j!tSV5U5)->%GJ45qy zt407{WJe_bMmfS!qtc)ISnXE(l_K`BLI+a{CL>|&;pDtB@^3!$0ByK;THKko8k@HMyTE4cF0l}jR zLnW#`etM)X84Vs4yQ4$R*{b2^Fx2pxxO&RJ;r4-4IpdpmBxee9MdudCo`;Ajk&tb{ z(M8Cr3kR<+*Nn*YX2`@);n&Cvsn3PF?p}DSAKT{Y<9-X87)H@&sG8M^oIIrhgXu!N zZEDO2V8+>lMX249WKqe!40&68L@hj)OGa;f+pW>9Gp2RDY;plM?)AWVWBeRRc=ZQL zzq_+8zGC7gW6lrvGi#qqBbEKhA-KluB>a4I%BKrzV|IHU!l& z8Krg#@J~6c>RK}hachUSx7%8uLF*X(!{K>|h z8wsY0Dt`&5x5}t;_7VeZ8?*Mut3x(Hw4GYRu9N{344{vHN0TBB7np;)%HqL~+MYNp z1zC<)YnpVSG`OSGAsUZP30s3_$)wN!h8VLxtHss&S7co z5v_5~6ZaE5__`27;n%|ecc_BdL{T`FrFIYk{AHyq6{#}oV1HAp(Yn)<)x6}P6Rshh zS@~v%RHwMgB8#DoAkmK#J3ny4DqXgd%dSYp3siI!JI`Q}8jP0;rZ<=cA@++MAr|>H zgK&Nj8tSOLMpI6f1qK+P+1vy61z(v5B8j&Z6Zp{qMSlbb`ISAQX?#0nX+I8)jL}cpc z2RLjVe9;E}c3^6nVUJ+x~ovhbnY16=1(Yx3Z^s4G#b@20+!$HS8_ijU^)9~;*eE{zq<1Z;=!sl3yDDmL8N#!T+-8%= zFeIffv70wHLmY%uL=@U$NGpIvapW}UmNgTT2@8Qi;G?StuuHp$%2iu|zHfyd`jEER zo{Lvl3EU7kx7)7&wmJVtN>1u<&VK34p1-&T9H5dHue6?Vj1_y*xNQOc=kQgE^Yf^TE^cm*Klv|=!S zTMja6X6@+xA$>X8L?LV9G{D}f&E^ad6$esFi!n3FHS04OzNNRZ6T)e#s1FP!t?=<- zC>Uz~1IKsy#{g4gyBrU~bf`~Q{efbZO&yQXOe#Yu{(A}*dMm9mr`1{ViUrbO2?|$@ zyq=qN_h~`m%>(1lX)~fk%56%89geJRU7@KIW3dM&_F_3MN%+>3a59Z$h@~RV<8AOge`3&^VdMv|6IEvFXizVH7*QB?| z_!_^Ed*c-nF4Qw;&yV=)>rcG?mtLLanyHF`aZ9q&_=S!C4X9l?>Rtq%w~9nKyYbcK zK3KL9JnV6nD%+|Z-ke!)Y82U(Swsa+5+EpRgP~#j(^KWFA0^-zT_x#L z$vzSE;$VGplfwoe1UHH~ZUHHdyY8l13|Y(Hv;!XN@Dx|4<^1*ewutcin|OA#h@sbA zt!#D24*+q0_-wQMa8Yln9Cei3RDg2=n}f^5ML8}g?&$;@RSEJWi zfcws6K8RoVQ3m$x(zr(jnkiGY3Z6#v%GwQVJefJ829rZWAQjGUX(DZ&MjwH@WRH`@kAK4YeV8rUhhdV=#&RlG53GM z8N5!)sWF&4d86P9&Lb|X;~W9BWlG{XkwRWNz-|<3ji_p0=^8?G4i(R|) z+IIS@H3?ml3>$3^)u+4jKh8C=>KX&`m=oVM2M`}fcd6@#gM12Jr4EOZ7+rVPtEL_H z!#MAA#xAgE-fJmWH~RMF!Ps|)jd9h%9Nk4r6aDruRvK{9`Eow#hFqZiZ55H60ZDqA z?Q3ru&YEgJ>40;o#pBj)?<}2XSv_vusjU*$TxeRiF>6{^zW|0N$`pSU%;mNIqbW~U zUwWNy-x6`BEv&v&5flie4rU4c6NCD&XtY_ zw&>-t7)Jf{FGKa|8Sr^$ZiFVkVktmYdF;UXebP+)>C#>Y{|KRVG_L?YAhT|zEMN6j zGD{txH1AEbZ*<9VmOS@6|M1Mc02D6(?9kJioOSlQNPQseb{jawvukUk`)MFs)unr1 zdBM-_co^3iRjLhZ-!oIDdG>*9-1TpGmDn6ZkP5b1&IpN~J*x;_kpcg$V0f55{vtj( z-@^+NZrW;Iwc;SOUfrMpuNPB)IoOhaz5;Gh+mXi_sI= zbCi|~oCmeBVBwR)XVxu2qP^)-_O4ECdcUnE|42AWe#h#{ROqvTv)|=agA+#m(qz?M+r_SW_TUS9X*rhh2^IJIsCg|t^VYBA*Kz~2G z`e**_2k`$Hb$Cn9ekgeh`{tkfIz-Rzf3~gu%hh*PDuD7)eSM4f;+aqYYc$}kvSc`8 zjs8m!eo5^9htEFAVAVt^wCHTf^7z02#w`PaAq?Xyod*eBEex^o6hxK#p#`5CqM!u7ZAaQ5bsPqB-g6xx||w%CnW zll(QN@X{}})Zf|}uW7)}sJykZM)=p`Q}tl+Bc}31M88Xs^XId9Kg1sQx9a|Q;{Mkw zdADGX`|*!yzx6b~#G)ckv1^?ai@S10E)_(DO-Kf1yqEr)!u;d>Ik0Azf*zgGI{on< zzy*IT!GG9WzpQ@?cKzpI&k(fZ(>>me;#u3H#qZvm#@it@cAZ>9lC$D?^8iHR-#)`| z%{yC-iecR66?Et1a5fcWFU0UE1n3t}j{Ars=@clib@@~=8$_vY?a2aJ>+5GYnXfiH zwgq(ppjPgb&)g%tc-;?~zjSd(j$HW!I4$LcP^QUg^xB~JD{hxn7`Hh|QW{is`28@5@CmiOd$Q`y60Jxlq}^&C|dGa+-r-sNuAm+rs3@k8qr zq+$;a^Xg7e1*L&-aolMbeQek0(mm~`I*^{^CcC{u+g8Ue!4Tn7YTUvv>eccuYj`@L z#wfq2g;3d4m%?N_v{hNtsx1~PA?}arp5Vt8;{`Adxh@FG_x}!)z7YVn)bzJPyua*& zNC;NHT4T+lgjNYhD$o|t@0WIknFG~X6A}sZPG7i=d?6}AaI&9*jB06~Cd?!p)a|4@ zqq*iMpd0et`+-vPYPl#Ktr9Z}9?g%_eZIeS)2%Vb8|p&t z#9|)rw={2;^vvr~CwUQz4b65>Hz-xgHrjWL+0>9gM8ACuycZ0Q(=Q!xDYb1VsLTGq zhmV)~#2)3BTFsqi1drwIH&>rORvenpwjl@uAsM4q^KpNuLJOgS0L60UXaPv|M$RwU zG_Sd>$6FAZ%^KlZ4*Zc0ip6ak`48el`%>i$$tJihMpY-!;TVxr1!;qZ{yPQ)S4`D>>maSg$Pb=luR13 z8^cY{2VZUJD>0W!dK!p>SWjh8Hbx(;7rE{d^IA^Wz!RG_j|!Us)&cbxzud0v%Dz_8 zjNw649X6^?dWKfaz7$6Z=f%}hyl%2m!VQ7zL_Kj0fs!vRRBYBM#Ji@~s^lus=v`=c z(C?|300RTmgmeo;Y9G~~z_}8~Rvfp!V~{et3!OBl)C8+Ji9ViPWT2csQfwBAj*-QN zFy0b7GTCY@7v5gvx{@Z7Sg$0uVSJiXX7zI{VX)_`g|VEcE?Zg}#S&i`52 z!+i>zg^RYN-_FMKiaf>obV>11<+)(i5!(3s_ay72OFJ=MOEx=x{Bc3z#Yuf{cqT&t zv-XD^HQqkxaS#&cy-xfv@jJrQpy7bWd88Q@DQOY8qr(lS291vxUI@%kwm9N=D?J$f znyc}-cQ8Y<50ch*?WVV(6OLEjSCMr-M`KW)#!mc*5^`)K%9i*>4s<$8E4R|>_P3F- zN(ax!Dm|Y)PbR%iS%?@_ZON|QVbXfOQMV}s)gcf*E(l}R6zDZ{*zi!d08C+3__4c5 zbS&K7MnIL_Ms_uC^L7zwM|TVRCSM~2;I#?#vjUv{o?QKJ`zV6DRI_xlA*sVH?ENJ} z#hTZ8WW9Pw`0Dzl;;qK%dsVsmGysgMaZ!TU$QzBG?{C8^F6TKI(7pAF`7l_uiuB6c-^xDRU~xl>_8TZ zc}OG@P3h7Ni9dZuv3Pyg;~+xYdRzbYfkS!y-Rbzy0w#0r{z;~Iu6hx$b@4Q1Gec;M zm(ZKiMMkYp^9aW-0X^d~J&LQ}6tB!CUf|m_&2v7TZb#@#r9fidxd)nl?=L`3MqVee ze$6v5^?uxy!Y(6)G>Y3eWUe!YE^B~w)z0d+ex|f4Pa<`{^WoBX-3J1X$%}}HE8k8* z$qT1rH=)qrgw8hlJ^#4J2lRwrXGsR}Wv`Qwi!mkpxKDRUmpMkh1yfhU&i{l&iKK1_ z>cMuo-sD;?D{;3da|BsTh4T0$;>ZZA!Xls(7p`(}B=7RxwyZMPD!hn5&eDxfpD8ZUqco?mKGTJ0{j^( zS+4usLjH2Idv18~!B%5VU-g_eLtGO&Zd%RFT)!y&`TC#ej*uLkLX{$u)|ss>xAzx0 z$%47l`(L(H`GDDl?CRfWlbp%mg7k%#THXnvS|Z)6ko>O6C}pR#KjBt^?icFPEQWGn z$ecf+oyQSBF-AHFixh2GqZ4^!I6;C~u>O6p*Ej5GB4|H!B`;MrN+gB}(GfA;y6HaI zV${AonkM6rO}M@vVxFjCG!$>bj~e0pK)}@zCp?KKt#vq(fdz%U*BXdW&ii2~)DE&w z>;DSy)h~zvR&ToHd}=iFeOZiEg4 z>jUvlx_A$?N?ui@O2&^Pxbw&|sxC>e&saUrt9Q+>-T5+Za(obluc8NCcjVD-qqARZ z5p-U3kg+?Tws$BKmH2kdx@G`}c#TuDjF?UHRbAgw8!A@Fs?8_?^0+MMk@ye1pI4Bv z?_P*GhdwqG3M-o=vCJ4q@~o-c>3()()P6_5#VFLjy;c6et&2jb%1O?-s{CFLQqCdI z8BeX}`pZC7W;Ry}cg3P{htOE|b6(j(r{gClQ${l6k1f9`Y}Mfi3}mWLI_)LyZR!@*E5A zMj^5b$aUlg*;t26L*JJ3X}}*QV6foZt{5`n?{c5@xBG4sp;d^o*DQMSz$e(`GRk2> zr`052fd7f849%B2C7MMw?XlaWT;`2tAJ$$YNj9Jy;}!~m`mh_}`5LWwTD@=JGflqj zxn(Ai)E@&|+ElaW8b41O9~RulSMMGQh|}X7Ng^J*Zf>EL>$m5bMp9P1dxLzh(zD3T zk?eY9uTcdh|I4@9krnnPtu@FALIZCkL{7>Z3I+tY;m-Ghlt5)@EfBlqYUtKXn;1*$_%T;(sdfw8b|BT^fi@t9dXr#ndsu3oyBlU_?(Ff_HOApGH2T zm7XXg_V&LKQQS}fK~WPSbJH(z3_18nmwfSYQOZ|AXK!YW;*R46C@~nb`bn272}jy2 zzp2Drg7YDpQh87{a>78$;^ZcRKlaf%W&YC5M}oT%dvFs0?`tk(h40De=hq#PxjdI8 z2N-Mp?}xf2_Je5>sn3^-N{2Q9GFUo)$nD`)_5(AOd{n1&9_TsHO*Yp?wbryZL{A{NXyl1Ij<*k+cP zEqa~uIwkFq+|h~8#$*iM5=gHqP4zr*Dlrg!IX7NkIHEPQrA`IvhtcpYP%AaK$c>NFBMkPzhQ0j*j!`^nY$HT7A6pYoqrRLjVG z?J6yHlNN*K6x1wUp6O6&E|}KJo+Zb+y0gyjB#m&xd%lUs1L-t@k-Wbe_GQh3-EhHj zxo4M48+n33VZG!H`+M@;y-d3N?Hpa4TbjiBgwq=wvB-3aXXQij*8GkuoUWj$ojlwc zp%;Ned$-4Ht4;#VHe^PZ5_r15UY9qxU+;djzGhILXV~l?$umi68u>9R&@1YNoCRk8 z6cjEm=45%l8jzD97bZU*!Ew#itzk5Xdu`&fiAj<04@wq$Ak_DRfvFOi2nOoO;P|latNBZ@W4$-{+qYbpRj-t)Q7&UmE`5=!`-f- zwxLOPYT*9Vtb6ke)gYyVAH1Ij$ZUMKqC0GQj9@HA_$g-7%Wl>`%D80toQv-=Vz1I7 zKeXC)_iMcHKVaX7mf3iwo{#4B4c*)ayt)avOfRHN2HXnlNdau<^xIdh$aeUw7V5#@QgMpkNR^FJ5+$=(DjL3S+w%mm?5_q*7!=Pr@t3Z~yUmlM zO$~+?g!z;4L{FHPjTyR6d5gR%P%)|JMHr`Fp0SXdjclgKet5Ji zoE&rM`lx|%idpwj=MZ$Rd-@qbgg|A9OVFtRpDu*Qd4YQ`X1hm%FirpT6B7OQC}Fq~ zo%@Qe>ra4t^R+vymig!!S_^pP{_1*37+1Soy87%7A@opu4@fjzwqMt!+#?;UcJ*Cn z;Q={|9Z>ThN33umM+-J&kY-8EE+v!$$0ZPyfySaOU4e=P6}E-DOS5YHY^d7j~AeE9=ENUYjd>u z9`MSIz%I`@Ba!DaezBm11;OwDhV*ovyM|U`5VomsnV@ZuMSH%=11)=@z{f7s^%Uf6 ztoxKb-oHWT!sE5+-KxSQ+#I%EoNA05ElBP*%} zU5RF{-47G6gN-qZNybv9HPB%ba%t1bNC5t4yLusZWGXcP7q`) z9t+9>PQsLNX^(3p_{ZMw#y!ju$@EyL+KF7?FT!_TXt_@i4K+3L8BEe>RpRazpv7~f zW>SuHt-G+ix5jI4^+zE0kB#xnP`KS?jCJY%+yk=sQv?aqU@SV6B57w1}dM3FjKNG3SePdspnM`*fLZjg7QeFwm;q*v8?uJs?TtlFfCcaAz$8B8u0 z*F(3J($<|*zR*Iurko#;4y@vw`1jd7o_skNy@}xOSWGnW>I|g|40}8h^HlJ{Xk4^# z^pT~{5i4x73RATCU1--mRLkzG$g$nR$g{dl*UNPD5Bj+L+q@*rat_6&Khnzx3`d3HRCs5_X4gx|J?H z=VJ<)8?zqA=OV{h7LHz3x#}-(8qg`6+;<*#s&b9!V7Sy_@A7TZCzd=8U zdZ%a?x=l!zYFhXy08Dpx7q4`1ZG~_xVxopY`xSa(k9YV6VN#wTMsBmP zP1PzW2&XoHl85#=t~V^MO9Uci)Ws0aB)l+TPjK&0k5S$K&<}TT+7IcLufAxu%iGsE z^4_BaxqwGD2lO9#515&9mwExZg)k@x$R_i9(_|K5o({A)*Ikr<%6&Ce$0PQ;3m2MQ zfzh*WiUI&kF|z$@byu<*v2TEl%Idpa*}Tc2!92}7ihJv{F)1hbM3$d6Ht}Cz4&3Xj zZ3c6-ZHmM<2r*U)`SoL+7zM9v+jde0KH1*F%*f2HiE z?)FlTA+7c}B*7?T}G>sw*n2#q<#lDQ7mW3ArFs~i_O!p{-s~kK= zEjj~?UR8~Kp%6&8N zlV?T$`_{PN7pe57&Na8f_>zBOLShkz@ z;fC)`N|KR;<=+VQmx11j7hIf89C}qJR5RB~mp#^5GZbHTGfvs5T`kw@hRo5``S}{r zNYW20=pvWaPgM1K`TJonQY0={Hx~y_rw0DKWMt!#Bh5ATI_X6NYR9+rLs_E>nt)pP zE^Cu7(S8EUX!m-}{mu5J$A%Xomangq)aarYpW<|eW1D$m4DJCUtml`cR!jeFlrkrZ?XAQ9FOXYVtGnGvVCJK9?r`HI!W$uRi%2?>2nM z)=Fi6V@j0PWy3l_Yy$Oj4|}8tUBZbn9WU@UJG{HJ$(guseM@zg&Fc2y!Po@^j_rh% z{?b-;HOHYYzXb@)<$RXdCz^z23u}f@w~38@tdbfPcI6lhdDrqxKGxL!>7OtK%j4)N zm9sE2o^WOep>sBATDGwC0QHMwhb_&y)liUYV$~_9a;5NVN$tag=V9}IlNYvPdcS9V zq+kW3U^)HYt$zLagYTVvU@2L&@U%PKXJ$odVKga^m5I;YJ^thJLc=+b6S8%>v*$Qk zOtP_f3^3<7c&zx9FK=_yOnN743^=`c^h)FT)*ENt&QkP`YjNjm`Iq_J+&81TF51kF z4SQmbjTUYF)9qn*f)&g7)rC%UKk-r10-mT(vSX6aY*?;hPrN-wFNS@-r1zUGVHHT) zW&+PoDqrYkl(gDcsz}QPsHp>vP5MXM1zuIMIy8a&0W)S|vGb1^tJpk}*tZ_S1B;WQ zwB?K%i{~mSJFBc2E4V=QY-E6T>0sg0i^T0Cw`s0wppZb*w;|pXI_RH zbtWYWKWRI&zSbGM)OMHk5s2jzJ0R^_6)Y;MuJ<_69E-i~Qi%z$(m%YH0`J@0bKMbl z2)MachOgV~)ompm!D^U4U0%Aaw|jziOurlh>Ho5z$<_bp&-OAxofvgmg=i{(Xtws^ z|GdKQUa(W`%~L<&^ZBtNBkg9V0WC;K=|U$xtJLFNiazKHX4x;WC^sAK#LjdUFJf4V zQjhR{p6!2m*S}6Kb;(_)(k#gP3c2vVorjPb9#N-GO{=rf5}Y+KFdjSK8uPC zKIdH!;@?RSvvnotyJd`VAlPwmWX3p_)&+<|ZaD z8Scb5lyAyq^n!WNVe$D$EM)B4#?8@IkyO1`!Q*tEg=?ta`1CqN=e!KBNn)d!$m18af1UEmHH z=lV;Ybe`m3_oUltuS^F%!AV=JqHVzhW_>LR?{}Gg@O9B-9^aVsLiAd#_l8@%SXWO{ zwnc&0o<6$s2y{`T%&*=En~i>c8hhNqjh8qe74il`G76Z#GgmJUUdjtt+U!No6zpHk z7AhwXz$gW36%7eLHKf)y&kAmFugzXk)2*ti;<8_lvUR?E3GLGc4jP9brM^wZ^ni2v z9XtRwY!o4ib+LQQXE)Id=9RaO<0Xj_Tz9w6+pQUmM+NXZz~DQiFk7UIXzWe|AC1mwjuSst##={8t=8?^y4+nB?=UBHxH-e7Du#xxIsznv~D8?$)~YGwPkQMaA9g zr-}k=FY*HAbTaPI@xao1Zute-WFJ4Dv9E6kxIHG!e6mGMNWa3)=GE;MB5d+@9I>#S zvJz?oPxA#HSolUu=1Uc&hR8iVt1^-wNuzBLW5X4A$y`J%(TCdRySbT+ z>KLf2n3OZtZv4xG2|xX21Hp;Qrc>1A3cF)?N#-(!Jyc48RWbMCG`6l2XDu*&$be6t z!&u)4diiEx5U}I8=Ms&r{a3^m3p*-rzf$Q`?85^zkh*q3c?i-ZS_f&m{W{k(s4 ztKM$4%zYkA4We&vJn}CJ9X{G2|Bla0fOV2L`fz=^4|cmwsnp7^S@V1#r^|w84U<(3 zWav946`;?hb$4!>qsJ$Hs-?H_uQQFr@cG01o2B|(*Sd9S4U;YzzE_BR%d| zo2Y|zkU`YoLil8kJ+%nB9|0}&@*`TNjACkiRXUnRxt=bVWC~LF^n^0!(^k z#JuA4Yrb{r4SXt0Es&3rujCzmz?de)?pm{4iE9=^_Ot zwdm*#z_!hrdy88p^!b|Tp}B1FvSydr@^qzs&$l16_aKAu&|aWbuJ>Y+1e^s#mH~mP zys~3$GawMos|2z$Zd7t?K5bZ4pc_VSsC|)MIa#$L5&m3dGi`Qw(Gz**@W`6?Vhd0T z8YE;^R3nf~JbZv0@!&%T8q9z3q1}9B=iAj(i1+$cNVLuN13Ce-4PV+Pi`9mK^rdMS_ZFc2De5rWLv7?FRJFb<%|$zdsoI04YR7PXA+B zu|F5N8l1XO?nyjuE#nMpMXs||2>IMn9_L_x%jn_TQmHEVZ@`aDzLBT=+vrIcG_eM8 zTY4|8oPxIW)VaI##q5qA*v>$~QDr5nhQ67Vy2nt{kgb z#sFFiXl-YsxP1!p<94m7O&P`@YwubELNT)etODuTHgaOQvN0~B({aG#Yju(u_`zNe zsVk$Q%T*qpvGkY=7(~HWu+Y9eS9Z_nZ#y1zUwc?4nzjB6V9tr%7!g<8p`>|FMmR@& z=H>eCfuIj@!ufwi!VQ-;NUleAfal&QJnNnEuQ4mlMF>6&PI6i|EEM{naQ27wAOU|J zl4lMD&-ouZ8-F5TYiJcQ?=YTB;+Tz*BVh0VJQ<9-1!)iGFL*LaSOIK6U+b6 zF{jGFoHq$GpWg`U^;Mi_`cJcj{{O}JEB*_;ar?6B|8Tp(p?@=UJ)&Zp3rfDyr)%Yn z07Hw?b}s7|9-G<+pwCOn;*z(`QUO9Br&Mr0UA+u;V$~cWFVb~xr`srZ-erZfE}T=> zI;TgZ8%nwQ7X^U@OwJ))S}duf&{H-`suXk1)3%xR@bWOe)|d&e`u*dE>0i7N$OKK> z+zGS)q{yHWHo;Il6~N}PK*hG=p9#cM0u(!z&+WEv7yGz9=Qnp***;2bZ=eCkkvM}b zV5d7kj^Zc9dTVnqt&xx+RR`d9raA)5GOL1s%eDUO=`@Pr$wpjTcQ!&Z{ATY@9q&Hl z(gu3e)Z2XG0@Va0io-1Q&m)>e7L|yO!W*Zm^Q1O~`))vVxFC4c(NpO3RL3`c(E(A= zkJ~g2r6+QzW5Y43#Vi>`Y4n#On3LF4luQU*(F&xT90hvA{NGB2Veq{0!W@;ZZO0f? zPWy3%Pgx6_4nyl5zxqp1VVxBN%X)Srw65WgNAQA+?R+Zc--ehhHh~PuF&C0@j61=+ zLj$_nTto-HA4o+HHn<~@=h@!^vTGfgbHm$qI$#xLN}6@Z=^Ub#euE_6K3AN}ZF^NV z(a`%K6+Sv;ijDs9f5My)T?>*lrv&R%Spyo9ch_5yY%R;SbMmB%{Ow$9hJSR$>6lrD zP!eWb{V^Zl2<6e29ZyH89}u7ZUMxo3zfSo+8kVO=PktRx498#zTuB%%kk~K;p2Zb* z%O&NFd^glZ3bgHr~d|H`e(#V?Ao#j{K zsuT#ez86RYBKi7jlXQq{+XdPl^c#GD6t%QwA2Kgi;?2>F(T3g#>)^B7dF^Dr{}%;< z3+-5coL`hcgyfXSx&Iy<0G>cyZ_O)(GG)F&Yk?#*rh?193h7$=4EI*F!-dt>a=lyX zql3qpYm1@6+4Mv}&z%y^dzF!gc&5DRHs#W0v?Ppi!A3aO7LJG5?c^>Ot#o#o#1Z#UF4Ix^C6m!gB`leI7UjXSs+hR;HduP1J@3(){6~yJ(Tj zA9QHG_v8B9R~0n&?q0Ys_)oo^=o)IAUNSFwlllzEM4?&zY<%M%*69Gmv^xnq0s4g0AZ>dPO3Er`WLc` z1^qLD+5LQkJ7^@5mrXc^pv}e94>;IENUxg9{Ej~DczrqF$lAH%?$`Ls{PfWic5*lV zkOhy6ED2Ew8sK_(T!%l6V5m!mZ8{7#Pl`;O?wjH?3osEHUGIOf@ye&n6?=H5p`{Fe zs!iAAl8{|-4FyfXRT4{-mx6T8p-r_hD+eGI^q-eV+|qo&R5?hOq-6gj8fdhTyusic zcD27!7YEE)0(>0|iq5p=KBwcn>G&R8!5;

}fSn==zkP!JI1f7C~Y+6^*B3KPr1%b{fpSqjD6Y77T z1O7+O4tu$h0D?|Nhel24J?KP2iFc`M=NLr=Kir*{|ijecdVUx?LEZYM|1fi<_m+|Bj@6dO@!@5cI;iMWm9ABN)5U zYVwTW)Di2pzX0}=*HI6Y%zpCgG(Eu(Gd zSbq?O*Aj(krqR_do^r3-ltn!eY0`4iHZaD004dPgy5V%%>uN57Lb{&NiQa2SH(O~P z$;rI<_D#a)8~lk<_#YX9BB0KDwOeBr8@3iqo~41EAD^k#R)g{#Qx~obRu@E9A8%Gw z8C6T7(hEhJ;D?VbE|us?x_@8@ zRwEZ`XFw&Cy+%0UwraqS+M1+!h1&e$E7-THb{UrrceRmDDx8<%hW82+9h|z|KGxm04J+BCkv%Gs#%Hk9Zixysn;9B@5O~hEMbloY`AM12^I0 z{t|5p*HPFntuRsG6sHDq;7RB~0R1Iq0stLKN5OJ3kMv`AHj0n4P9n|PI-y@Gp-#D-_Wq(@DnrNBm4UQP-FY;PdsB3BqwlW>t-aD!}_vLZ%ZIpq(ml0MXI<_ z$QQ)21hTY#a54Iy0{Lj>bLtIA)r-may->gv<-iA_hMcZ5Qzkw9;H zUlNY@1^&jJ;>t5#-*x-px@b$V+C=;4Lwh%)oX~G(jd#>bOiQ1wMU4J0dGJ4#?)~3N z_lrwL?NAXTP>bXKZ%+3;q{@%mfm#Cz?ZNAAtp`f@sOK zu&bx!FAk=Av5nnpx7Fr(Cfx`V)Qu7*MULsR(JSKSqXp<^o`PF{wkvL~1%W^@XgQoR zCb1C-qh?l>2RxMsgx9Z}H*$CR(W(D8k9V@c-G)4mJhzI>n*B=N))^ubnOHZ7t%*qJ zjMxv5pxjAso(A74Au>L@HUC!xn3wuTznD;5ZTYOq$SP|6EYbWb{}{^mtFvBP;WNWY z(zU{F-4&fhV!+--DFkHY)mE9CK|RC4`Rg zS04xBzZzz4v7X>gK(4+-l0SsQ6HBQ{4(tJ|@Sd-VJ1Kr8j|9hfik!tg5kOlU#v#7+ zpDorm=oFm9%kKE(4c{LJ=M(30rkv3(LZS9k z+s(?csIIf5W#u${=*p5=VQMGy>Jbw%+TbX(W_Z7P0G+!6SB-K%zWKg-8SWN% z8Cc=@-34LNW92YPaG19p|UrzSG{w&KNB%8I{P z)0MtkNgm!97uX5WkObN-lrd^5$l(fx`nC(TPCtYjeIlvLSsrXSU)!`2n|;rklGrnTyPZwyl(`UPD|l%Z09i3bDtFA-Rnbj zux?z8*oC+g96GFf2~Y%nvNbqxfWvnhX*!291Ji3GJyRWqC+=42%}Yu9Jg*NpiglX5 zpC)80S1#s-d3(Ae?ca$314t$YMS?3`?zOi$6wn-Qa(54`ZRDWIhc68TE$xwV09nwT z?zA)`SGU%>HNE1%qrU53`Wm6}5Xst4f*c!Y*y)nti z7d(l{V&45n*{`e(-cXd`mJT`2xhMY!&rRRwwgs>G8z1;X_2Nra5HT;u8%UlmPSMv!ftTpk1paR+?_ec=1ho8Ow3I6x4h@=|pgo(>_i zO0NaM(`tPuSiHh8LH2f?twKVz`*-7Rn}1|_xyU=L$5I5wy?!U*3bJq|fmEv;e;9#; zx$v2_6AfHrju$Gf>;kWw?AC{M6!KkL&|pah?Ta*{xp*Cf4aZZe83oIv`IT;%Rtgmz zsMLYy<;wEE09XOlSDA_l-JaSeL<)ZfsQ{2!;IxUDwN`e{f9@{`&(gQj{tc zCSD_+3aUdw;Nsyq=2RF5)uG32=~UwwvDWrq|=dQS>^GK7e?)^Pz#_-XTU^d&-8BgqgX z^tih-WC%4{p4ulG*#y7?ZyZ;tYi^kn9+>tf|0H0j9(t4;rC%m|pY_I$z_M(RJ3E2( zOVECDxnp=`yRc;TDwOMaLHwxQ2%J_;syTFoa z(IKyq9hnN4K9m9R^aoOmyh9-x;(O&jv;E~+`uFHG9_CZ7=CZ2aaQa!)e9vd1f4;^z zVD6LFY@^X}U<^SP+BRzS#T>A{>wPm!Hjb3IJljoa z!*~qkRuPsBY2KE3@!6(0+N8S_{`!H@d};qsXF|lT*KR{{tIa$zuB2>k$TBXRdLW=DC{? zcCJ8xh9k6fNGb&)vrfJCIku_4Hp?G#+V`A>E8|z62VTxtmZ(hA*dz&h<9z+PtKILb_`{So z!EU-HnyMm%KQ9z(=U4*%Znf;AxgDgV-1Tk?fx+O8_a zxK+1XLXmoF=WsULo1VM6m&jdI;4C;rmnCQI+LJb$Ju=)jL+s7=&tSvrHY`l)rgplL zXKF5#jo7EB+2$h%%UfjK4nJDh>G@@pTD=`U$x>(Km|>?iB7>5wYL13T`JLXOxkTy1 zdc}JMK@RIP;a>A3d{6`dDUYpPg!FYA`+`5s_OTk?&N(>r;LW3sS75Kus~*DAZz39$ zDr|#`E#72}e*H`&d{5?52WG#uG83{sHJ)81kEaRXvKTdW*|(K<((pR!SpPN*eZ+D{ zBjS$$qV9CD%OX&09MQGxqfo_1Vs)dJr!%^zi!xue27#+UoUUu9wZ&JwY>(@{nSPv^ zVHj|QEDN4r18!OsezeON+bt~mHA@De)#Vv(f!mtMCqP<)^_+Zdm8(E8w-M|owFe!f zgy#>r095ABzJP65qr*Tq7fs=sUCWOxjrr)5c}Y>vB2kty)qE`vbq@>TnCt|R$#w|2 zXfFV~Lp35@OQ32jX6v!r6nvygf29F4BNTFnP#=Yg`$a8ywPt-<0BGu~p&Z63oS9eM z5{ZjW8}ceWAYRL|)G}+b;Cm$-ETBMB48yhZ*(LoAKFp7_9(o2>jWw#XIxWP$H}uhM zoA~!S_OGJ$wgC$$Tn^1@2Fm|l=z?vix+IHjx+!eGhMg}-ebOpRywfh$ksWhMpJF)f z7JHt_p)K9<1n_G`Hral7Y)%FviM?umc76Y10c^K-M-VA8x0vL}!7F-@%RR+;Q2PEc z<|{f`hSHQ6)`rD3aw6A3%W_HWl0;?A# z=5#wm*-anXhBUHN#ee)Ocu)(nw1!K0GEJ8kkkr@&{>(IK&Z zWyW2JO5?sQt!XHfJy&W=2#R-#`WU}&U>^S_gM1hsH3@Z zw%inSDb4%n)JBy;CBvNpr?=+~A}V(!`RXCGiApSp*zz97;1E&6MsVr6%3<|f=JL|} z1iPs!4st<1X`jwo=v+Y*N4giqtgg&}U}vPT#+~8yK}Td5YKR5VEmN-BdJs<7DOQPb z%V(MI9j<%;**7UeK1s#v)CD42YOU!EK{$;+c=Xvpj2oLaAG8d_e$1ZlDJ>GKRBKtJ zhQ51K;*@Q~bQy0bX2hn9a$W6w(e-psH43Y$LP+4YRqM^g<7dDQ%<@;aD7(jS8jvY5 z>%I@IoiybO;@jk0z|j zCM{fTA527+%yf8Ro49WQ^E1nguaYN%v*|V{_y7_a2;v?7osd~_*=^?sMM#zMxtQsy z_LR&?%k5*^Mu50i)@PiN1Qsbq^}HV+=N=#xkP_7DvBmLm9@WLALmdU@z!J?;dBl6z zHR+`|Xh&|s*L3Z6lfKcbaJE{Y@ zT(RgNK`jFF{olqu!xsl%pXkxHutkMDLviR7-U^1c^ZXAUomFW-%tMiC28XQL_m%zA9&|S-^<1jDOrws zwb_*jPY>WRI{AxuxUzxU0V>h2Q$5LPWl5bTuWE@3|LLpn^Br_grNKe>vBpJ^0GgVu7^)IGOry`RJIR$WZaZSX;i+(T z)0)-H8Z4T+=e~p?&5$2RNX9Fz@JlP(FSHC;E6}vvdbb50A;6Hay173uNzPK+^%zCiJMY=Pv%pP?uy zfKpTKmXxu?R%&lz$Q2c$!a{3g(k!NfQiwUd$HyHoWxD+&WILmg!tY%Z6(H*c`v%>K zXDT*el9DBoaRAo^U-Vw~$b8C^7(XLlWdx8aMwjKh)cOpX?Xr$k8-gq^ITnz@q5Ev? zQTJv)RqVQR`nkQT22*3x7s=8K^-cTqOu6R`$LPKX+277Y9$q`_`lL3HGx$At3ZOQm zs0tEgYmBdW9Q1utPvALzsN0Bd#vVbn7~D5bD`e6X`=R>!`uNj)i6+XsSY1FKPxcbWgN?#eq3<%f&yx zjkJPvXq!gT$Pl6-#D?z!wtGBw?#RZW70E~lLn9GSWYW|KkIFw=c0NB#2eOB1-eZow zCKi8OXqiIQC}T6lfu#8l+6*B#&K{3&4q~)&tP`}d-EN!P?H$pB=~}*WquK>6tgxff z2Lhy05z!+GbZflo4*`TS7CGmepg1|dY+t@8$Mc+W{oZ23@R+Nbj6Wv~yTg>SfYppJ z>;k<=84?hBLpX2+L>H;oeWlnP>srFyvo*5(aSV%sI-a{;Jla5iInL>NO-fXeVis{$ zYdi`Pb)BzJD7p6UV8@ZCA?E{D6UlfwaeA~EG?K!L0!%vEyln=qiyM}EOJ9F!5gj?u z@-|aP@z{2674?lT7`KHyAzIuB=JdTJvS+$X9a$W{$_#v>^2!XjFD|s;^v^+qTD*O_ z()4fnVrt55dG#`I8d0#KkI3`U+Sn)qhJescXLqvY{=OhG;6(Tpi&0ww655s5qF#eV zoDCmjq4dB@B=$5QQ} zT*KiU97lSo*_VimDO?7Ow|aL(83H8(JBrULfk$z@JwbXKTZ%kMPt(MR`F zJD<~ev==hkW)#vGbG5izXUnazDLl^KG;459hY+crwoAuUL{j&dR@Z%BV^C0g%}7;| z!VkV^udk*yqj?zLLK%3xueej#ZGS3Iba>!Ogj!>iunYw77|nKiWSx?Tx_=0DO-o_u zDONf9j@6(2guLGBne78R8iJXOxIeJIks%H3aE-SbKWxHUrcp@d4t_&`7JvNo#7`c1 zc(=J;7*4MUaf6v>Q^d#WO>= zJn2&O-#e%OofkRMfns8IkH2-#@xQMDGo&>03j%o==BFQo9nV=wjG#w?Q9si$gh~oz ztPdj0QF83@XF*03%4EO3<+#vGBkYc7b6V{+q)abW2-bfc#Qw8ULF#GF{g&Bqg^m%@$zy%l{0o0W`Al7Kl=y(2_*!Uh6xuIv^Yq6J!ZJ+r7^HHyiX zx&?6*C3<~&REZf`CPvfK62<}j4QFapeNYuZt?kq;(d1csLvw9!lTog{YgnQnaDO#g znYW1G)gXa9Kvb`N-eGn#Gq1*8WxbDaWm%PdWMXu*=<%??k|0i0@d{^+h?Y4Q8~Pqz zA`BywX;gUwQynPJ|MWX=nntAdzSIY)MZ=pPU!KLdTAm~^myEs^O4D&A9P071l^12u zy|=id-B%5EM1hSSLv_HF2mB{6sYpcPI7m7)TJ661glU_x@(hP7n@&A|AC`A9gwBmZ zkTB!6G^EvJ?R7w_6mQEZawHk@N#9f$)_&*gxG6DurPuC*4m;Sr)G@JUKS+H0hA-iT z*;{z9Q>YCrX7SbRdW$C=5sxfr0~Ghn+;yH4A<3wsejc}U!8-F*?(zP*AOH~wNBGS= z4uy^>GG^Rxdn@Sc8q5?A~ z1JvrO%KrE^;aLJkG6O|w3q66VggAKGU=blo4-l<^1TpWPdO{5b0%I>tYk|G z5o;47cehftwI94N-&072#JAdMJO|xh-zXGh4OZ(dwp$0>n)3A2=6!KWbF!?8b#=+6 z?ze0o#|?W)eWJF@uUis8{8A^0p+_mrc|sk&M|MCoo)$jefMdR-IFco8^CUcU;Qi1- z=QL>6lJyP?``LkmQReI)y8NkQMFn(reG9G%a`+ zIh}B2v1^@x5&LQjzb;|If%lOlCT+s~-l}u4RJf84ccsTJ3`s~+h{I`($fsLUdOPuA zGwkY5lwv6gsBFe0Q}kQ3*ky_#g%bRMiHkEsac=Sq-0Q(;lv5}CN=O;{yDoC2Um?_Ib=A<$Vv8z!idIM+KOzc_}l zXV}eFh-|jff2kBriOPC(lp=q>Hnk*xC>3C!J_&zN)pvwcmyBtTqPb129BV%}7SzT@s(0 zFpp6^3}<%wiY@kd&Gi-k%8R(RWdd-G6_D`QkE*$EDzFa}Wmf1bueZQ|c;9p31qVkj z$0wilXS)sF%oV9habc`F^2rWI60l_lhlC9YkIRw{hJL7y!D6QFdzr0q-tlPL|2?rr zv?D``xq>t9HWzITc3d8_%o1mOIAdw4zcGmztVuy57fpa<4vmsx9}0O)!tmjz{BIVv$O_F`LD6}Bd*4qPOD+ZU8?f&svF89)S#iC^dti|1Cu?Xmu3=G=-M7ob3v3!2Fg@xn zYc0LOI{*5S?%he>#}ds|8jJ<>t+-^?57nfUG+;w2E5G~0!s;Ww1OL1y|A(p9lo~uU zU8sm{NZqx?noCu#wVUnbilx9ESxc@`n;1AF2NF(&Vf#-`E5Gp}DL5K-5&tNJ$5n{n z;jH!Cf|4w;mW0nN5IAgtCJ}_oM)~2G`?eLb`D&yhI^=az$$W`WMiCO6M7Xz0i1Q5S zQ109(E3GEMN|W=fqHPiFskEuW2Kk?9Nw}xBsXp~TAC*=X=Mm)=1MP&}s*`SI0FB`1 z>!ROsw)lkR9M{u{d;;~xEHKV-bHVm;t;=~Fhf!oUNwGIPu~L&OiQqv^<-X}0KjVqVFdK(xUz_T%0Ne~vv#L- zv7G3w_27T7Ir{p7kseoG-W#gymPcC7i$?HVKK3vH`p%gKFu#BdTD+**p-NVzb|eo@ z;1(HgsCV1v?DSxrN%?Dsj>Rv;@q={P38KM06Q}x&0FntwsNK0=k`WQL^8Z(${wqem zrvoa9j^Aa|(SIn|amv8d6F13Ukbp%Rs`eZ~IX(KTl0gI>q_4P1n_qFkaB0>;c<}Tn zb{GlWx9ti2c0I=R&8EjImV;;OQ1;htoZG-PV}hWEon*kQ^%(V7U-VR!gvim%vhRY) z`dch9R^@cgWB^nt&)#*uRE&DM(ym_sgXOk4-9`-Q_A88`_C<{ahYb;C}Xnu1g>e8dech7~oLHa7+v1VEQg8o}2GMH|;Kdf3x}%1y0I17h?rx&oEpVA?X#uB_Fl$C4 z*RXJB!i>)Z>f;Qj~Ik$-zoh}0&G_f5i2h5YeLmR!qM_yp%UgI_b6New-SIOPB_DkrD9Em zpMEjg*GYX)uaV~Pp6pG6LD?qMpI;yrX-57s8u-*A!HT99jz;pXovLS}(Czn`EUAd) zj5L!@RgkJX`)P38H=SwpZy0BgPv3CG3Mp?|D}=w^LEq*~xDh zV}EM-e^~qCY*ll^fDhEU7LaP%VlnREj;|46vstI^`_qPKStDep`*)s_j5~#-$vG^# zbuJyhsf6zT`TCtePUB7fkd)QIKES8moU4$MBSw{7jV9sE%KMnv=y*K$fQaJqT<@^l zgC(5h9C*FNp;u+XL5qFC@7pFeyV~y)>|z)<@19Jzj=@@jJkTA341dr zrp&4F!=RI$Q(@7hwtYj>OeNeS6Tl>8E|i3C81;gx=ek&(!SXxaX9aNmsvKU2f9qJl zQ=32ni@sJaL*ruKsScVyqW{kX)^x$4bNr+^gso)a+P$gUOI(Y`DWqjZAColkAIu$R zAlkP*k72G>VQ7^qR~Ewu0F;fhqm&F#{T8~8Iz-$Ls_>2y*^D!c2ynaJc20_kvB?CF ztg0=trWZ*(`ZQQ)hcY!SVw|iiUTzOIMw5=c(8Rb3vMzgn-p)Ko1jOKfp0Wx{m14cg z#uHX+#4|VXt7I}NBLdoar5Nmu!=0=RUVyrgOU}V%2dTz=gl-s@=;d>^U+x`WrK}Me zxlaN&J2EO&^OlY%$TZj2S%K<1UQ2w^UWgjS{!JSAE_PeVdVV`m`dc7lk8g`{Q34Yl zuKb}_>wm%bMQ-Y&0&e@n=azNbfch8z_<}5a$ZPY9Oda4c)2s0cmZyGo<21Na{gJ`r z@Cu*FxEo524>1eFAbu}gS5P=>J)yFCL!N(ut)jzVxkl2~^Z(H1tCRi?keB`2C0ze8 zK#KkyAVs&6)Fd6Smr*dDRGgzCco|ZY zqDmnI+_U^R+K-khzmr3KgcNK?86Z6$ZF^!BsdedLu5%L7Z3qD6=Qoi@Im`#A0hd0_ zN#jbkZ_t}PIa~h$1vdeK8tdP+wB~_8eFky};d*s;ETsaDM;5&w`VE|+4Ng&Jv=moB zTmSTMUaSWr35!648_c}>*mjm3bGnK$twEE!WD>2)X3+EjJ(ZhdM0(&h?rEX2vF!$( zpw{eIl97ms0Fdq%azAc|Nxq#07pcExc8dB*Wx=T3@snelee+a=8V^}5+zugeb2YRSFR({2xPfu=krD04 z#05P&Umy$p;dH|JDn$n+f2~2nN8n!J@IIhS%0Z{mi5%(#)QE?p7Xwv3zRnrD#9JvX zup?56#2*W5vb`aT0ExE^t@Mj6=JmY*zPx6oYN}HD4T0Iw-Qlqi$RSr_Ydth*zs9t8 zz{oM?J8CYc)2nPDc-zsnJlu|`Q2K-O3nUY>P z-M4!F-NOBWPL`HP-~YvO7jX9YW{HZbe%15t4+<|EP)zY{S9d|jx)y%>B?zPzYVuIv zG$_-31IX`j-6UoATq{WQP$9y{K4~**jNSfMt}X6o3@`0q)pjo*!Canp%m;{oTMNvt zrWHOmaA?=rEH+!aFQ?IJh3>JW)6=VCPE{~d*$Q5il3w>Rrk#6*VV9}5m1aGjq;%eX zjO|#Qo86g>`(JdNtw&MO8^@a_l_cKaGno=RXS(2>G75EuSwcj~i%{arq9+ zY=_&eFFscEF+Gh)8apIjtRR}EQ1Ci|CDZ%*I6Pl~Q-SMuGOpVL^-(#f+U0GHx^$nL z)d}sWQPb4d_%EZZNpg5-#or{Vlyn?4KA>U=Dd3vptqZvW8Y$XlrlC(%?63C*2ef8& zk)%UUfFcBSK=M8BUeyhcc~;a5b>j|%gfR5Im*=TP)?oPRj@tKT z%JVN?6PIQ06q33k3YASl(u?-dHxFM!$bN+17g*3H2(k*h0srQcB`yNBND~X!5-85- zb^Tc>x+K|rtk$JJ%lKr$-@?)xtunGErs8dSXs>`FVVKeOthka>$W?i4G(0v42-s50 z+T|SQzgG8#Gpca?NB3Ctrh`MmqquKx$l098awkn_S3HF1F!xgHR3)mnc9yvpH75t# zj4GX2j=<%q)xH4bVW7`;U?QCPpf+8=E#|A56?@zG|Et9O{99snEw|G}6melwu;w_d*-P8vnK59z0W_B=sJV_fUUho=ivZpn79toG8p1MUQF`{)fNOju5R#lGY%JC^n+0ZX}RT24|SR@Na7pboy;nl98Yc< zN)CV8ADJ(mYlQq$!ztdw!)#cRa6NT#J*#{?ywidspKWNfKH{@rE)v%=u(Yo=l;ayC zE*gwnVUi`PxrUnhm5OG95RPWmdG=P?eNa+Zh%x}mODg`&H>4|`{W9_MV$0JIjtEw5$wPOro4t7+Hwt+G(HrC7H+ z>1$(p#dHEwzmRRLqQ%>+leGE+NR?#kVn-hzZYVF95F9f!J1OHfSPGOLk&Fn%>HAQ#xo%E;&I%5 zV~V&HEn!BD`@VByc^}b1V~!$__^r^^5UvFz zqL?X&o3EnAp3{KY2VWor`{hE7Pf6RGfK@}q2ks-|RS|KraU3A`z~QjuvQleZsqLG) zhyoF0#~y;3eZg|O1ylq;K}*-Bq;xe2g?#X8E>$mV5csy(z*#gn0bCrL*6%^pw+OQw z)tb6c&23weM=lWj?iRt^*172|&K}8ea+KF`uJetLBO`dQpWN)EQn@GlN-}loqk%P+SR%u#w}QkKalED4NT z5nITy3LRfae)HJrWD0_g4w#uFE+U-;dZm+F7&K{nYG=x9eZZG{jRV`jRc3s<|$wtWZFMwhN=mORc2bM{-bCvq--4IXi80r58p ze<1$YL$$?$v8soszbFHAqVHjcryfLLXF^sz;Mi%%N6(|h%1^gObzZ|Y$Y5|A4ixGC zz4Vm)yY$?VbmjREQ8Fj|@6t0|=r;-fi!q-!G$+7}p+aIRoCzGV&irP&@lP%g-oF2y z;mcZ$r*+8Hk%G${B2J`V=dc`g(R!Sdl!dMVwC!&|?oIvlV=zUvY4D&265*q!7-s zHx7^i@}-FFKX_T%f#oYUn$@_i*)gy1H(}@)SIvD(mdp0NGNb?f@7<@Rr`KOzL~vzb z!5xsMGv0kn;~z-F>5RNJ#5+p8boC^3ANosW%((e zz}z6vjI2&1>D_i=X24I!$>S|O>3n+8LI)R!!Nb+3dzmzQB;nM!wF#-LAlG4F#Cx^6 zdaaOS-6!%x&x=F(%uN`%0!x1#XL7f4i#VG33`%IMKrdqx#}k95nXCl-_Q5k+Zfu|( z1Q5!OyBqyGmJ8bM;fcs(dE7CaMZC%H*U~Hli?)3*g?Pbe2%Kf*+J1}tvG5`c7KZf- zNwA_Do=({Z`mS$g5bErH+wsMB$c~c`>fOwuJ~?>W`eBlWbHThv2W=mJbzsK}JcaZ` z(~+8oeF{O{YCo-1{5EWUr2^JJSj?&{Si$6(PC85GN!WEY`VZ~xUmxWB?YMG&^8W7s z_jU~H7#VPtE*NrFxI~~Np(CtY#4ek>#vq}NXEnG$Y0umh;ais0Pf4ji9$MycvzSzZ z?oE7e@X0cPne(8D_cl6W4EOayxTtwkMvx_Oa?g|85qj!Lr5NI&Sem|Tua zLp*Y_laWwUP)F4lHJh`(w^tqPUQgOYbwcP^?Z7Pw{&6m*tyW^f%XUJ1?Xb~*Ib|Vv&wh0OdQ1MzU3;MetVZU{y_fIu{-Z`A z&;o*{_(#_^xx^sS5b>%Y_%ETY!Fj`0(CiS0dH$v)0JySQ{@KrgGk-L1e*oYn0sEbR z4?T4Cv!@eLv{EKVNWpjF8|(sw6!BN6VKVlQY(^~#Nrzk?xAv+1kq&G`RDXZ%)n8HM z(D!I-a%IfD@EV}HN9h_u;4+};LQPM}=4GdrUK)Va`AD_B#_vCWl&Svi zseYL3D421);6n=LH{|Ip7K9aM1bp8cPyYW%d+VUM+O1DCNN|D$cL*ASlOT-;cMlH1 z-Q9vagy0_BEm)AoEx3l@?ye0qayMtrH}mGZQ#1G0sp@~|qIU1cpS9Lck{*Z9HDXm2Ku4?w$c>@)Vz{8fhkue5pC8(5};f)JrlgUwVL0DB>_7ymbw7OMyO_F}6*9GytfbghJ;Foaqnb^_qq z?{##F>Z$dVr=cIr7L}^K9h&0;zdCW0P(5T9$Ns>maj}BqnQ4_B0st0lRcx3*mECrP>ZgfCzef4l}zDC zC+reRL%LAqY}=JKxwG~DqMi6R4#~(x5p3|s%U~HmW%4b5 zGB+FlLqfmGr>GJVji7&w-~ZebI;drVk&GSf^*J3AbcX)6@jM-ala2%N@SpYI#E$+t zOa0%zIwtVuy^S#c9LpCtr_X~auzE`QznSEJHT?hbM-RpGToJ|(Ec+{6hk5bbFWb5} zl~{lO?*HX+|L+Er0}mREKshn`H`(A{qhSDx0j$FayO;mz+rN5#d(zo(t6zT|J#x>3 zEw|eA=|6lv_nQt8hR_af-hXyN$A;MRLYlqWt~L3)QU7zK{|UhR$5HU_w*S|^eP#ht z61~e-{aj}Md`=kbbE9Afbm0H_tN-5~`PYw$JR<@rzYJ*pH7cNHpmVNv^|S>z{-fRj zbcBIDQ{uq$eD1I32R)~FMbHvz|J#WFSF8Ve#s9+}LT{ewCvW_>tp7ST$-Qeg_d1^$Rhsi{nlvi) z3UJ@5i*aZ1xiK?-71R%UXr*?=x&f*S7CXoMKPJ@})$-wyvAG5O0r{NxZ6EgR$CIbW zSOCi{zBko=$Pp`a!k?WA>Oy=1+_3Lfo*pHDq_L0j*g0@QJFnNqpWjPIUei6tYM%5@ zIxW{UHcJh1Kgxld*)wo6Va`cUcD07i{v=k=97v0N6OE%1+nqtH0$s}z2%W-@7Xjzr zQ(h1<+(|6xW{3O^TGjqg*yO#%XtUC$+lCL|w}l+CyxDJ1_=aJ83R>JwCOO7H8}wU% zu(6CAAh`O1NTFDdYs=%8ESR{Wv% z@CM`f(+^|-utTWG6V}lY|AhJ~j}s9fF%L+L=@(}E-HOE!$F=D#@e1^37grJB-9~NH zk+rQ0x}OSVtpGfUgST3W{UUR}n=8W7-QEq}0ninTzNoqLI~tj!1cCeNv<1c+S^QGy zk{Klo$vuGQ+PmDVb$$pvK|bLM5^8gU|H>tS{N+WP7*KytX0w{D6jKnqQhiZUo(EW2 zCHgjb#OwJTaSbNDD|5;3H)rXyUwN`$I*7`4U-z4C(_N7C@DtY{2bGRZp_+ds&bd)1 zHpV5}+XSl3yVOkDZ6cSe;paO;GC-7$e?XX%ef8`9_8bJjYNc4f1kf>t-}1 zAQd0-jj#)V&b6-Ig4g2gC*?O~8p+@a)$_fjnXfaMfhyK?jVKfe z>LGDR^<95uiLX&5;`J}m><%DkRz7aMHEBjBkIl;!Lxm1gk9&=o(X#EX-d^CF?^a{| z0B)J@i%58WNE9yt_d)X{_hrnxt0Pijf*|suBKb^lz$&RgrAT4OUGqn)jEB=s0g$Tk z4EMGlNqVQxysFPjILVkSkQR@C$!JDY z%wdS9RqPBf6xYox@33!GNv?dpt`7ndX53Y#;^#IC4JtsHa`QR|4!Ya4Lp8QKLC=X$ zIh?{4d3_{E5q&@>M?wx~b#tR;0f4zUoeFnuN!Z85nPr8{v;w(_AafQC@;D@FaSsn*9o#;53Wa@14?;&hjU|a zygPtMhhi|wmQPsL%Ga~`I@`S|dfYB`BAn^hS~ zwHuR+Iui%GEj=FYTo_gJe}Xuj`cwzg>0;G57kq3h!);7q$U49co>y90)#f=4vQE&-9>bZzqM8^e{G*naU{C!L@6aqq1cCHnzM<71c};Qk9} zr-ira;S45S`5gX0d*)8oZ1hdVQA_Nm+qK6#j4KRhaS?ObEi(m3V&>L_M5htB{1{eU zxQ*MTYdh;Zn9oY7DTQQD70QW)`k0rqLKBACVPW3`tlkR(Zr0T1bYjl+y3`JjMlMH*maGDcRpe?d>93>w~XzGQ79=Ek{5zo?#% zsT6Qm%`Td0YFXOZ)_!*vnF3n3V5u)t9<>>qf|)tIUY}O zq!H1GecLr)v^#(-z~Ox!PG>&f6DP3oajSE!TVhgYjkM2#`*u!=$E$YAnMBYNi9RPl zFy}Q|QYwDk6Gt5n|e!ev#y3rTShSm(o*L(~NLF33p zBYz(>nEckt+NeO|M4--bhRA3oVoL^nqY zhEZfNAuF|o3b7RTVa(KW7krvkZ+}-TZ(ugJUR}3r7wau>k8S0m14b9ov%ZbQodhQ> zU|QRQ2xm-FF3isqEsr?suhCHRq;Lca>3ZDxeu%uC2wkXgN{)NT(&eM~I5+NluTG&% z!f|X=<+B}KxGr!)aqQ;A4J4dOXTRqhalm#?Z-(PAIIQnPDQ^2lukxkmHtq?q{~vOz`wvXggkqDa_unSvnU%YY)PU=(Rl zZ})KX9ODFDM@8RoEkI%SfysYmrR}_H6T9AW_6j%n2MX3kh9-x|wh2v2!|&2u z|2kotdDrjuYf2Lu)t)P98-mOtC1n~_N~w;xn|m9nOiagfoF&RbM~+)VHTY^5(O6Oo zz$N-kHPylc;;V7s===pMz~`HR_fC^>mYgd`RW;j;zA7c%uHIxwE#btVvh51*`XW8aEYQc}TC*}glZGLU$ zQEs5Fp<0+lgbA#nLiczSO0YUO0=R(~olvD8m z6v;j+ zeD!*I^cnod6*@b&fLy?|;Ph^5!LHqAk5VRb-BJx-1fDvfRI??v?jdt3``vLvi$AgJaAzI(ax-Yfe+Dq6_e4fLyR_vY zwo9<3k@1qTWr~9uUN7ta(S5?DNFYqq#p#tWZ9aLJFgaQO6>}cfYe%pUkyX zMWJF%9B1;;kH#wyS>%b*7od7ljvg%KAr|1=SL!rLKipk?Xod4>cI5(JTO#CDO4Gl@ z5)n51j$ZXOW$mf%w7DLz3A{ZPHAZ4iX4c!hdgNet_=z2L*0x}^ULBz~vBu*xo#Nm~ z&OKeK8My4lf(qaTqVSOIbO!k}pW1wO12IZH3(#l!{*70xIy?+gKE+fo{m3?WUwK3O z6~wcI70|xq1sU|^ZjGXp%>;#ab%FG>YS24XuF~rBf8s5W2GIemHtF&9F3~YS0JuPd z>yQO-rHUqf$lk$O6g&+%_oD*q67YDIgW%T&uWbNAIJfWV*PJ?oHNVfF^+ig6JyFfw zok4_9-z)JOt&I3=kXcEl*Y#vW1sR4oCNBU~N>UDYnhC^~#vRDk@T8rIExBS;EB$zi zSs25kyDw6bHY50&jD|Xb1c%l;!V-x^Zu@)Rsyx56F#Xo{vgqxph-&#qDbex}2S8%* zb}txza#BI?@nehYNg+_O@ro85zUpfrI{2=;+~}2Tb|e$!V7cxe^+ z_4u@kxhE;b-%?~}t}V2JjhWp3)m_m6p8?+3oNKKBBCDiZ5ZGT!&Kp z*7+e>RjiVuny*10fr=zfU*L7~C2l)k3I{#Uwzde~q%Ie=z~BaXV?GcHA_+Q%)o6>11c5H(HCO!0=04A3AUDDiO}g(-)x4oHD1sH(@uyxlM-~oCU)?Fw@!%WO2ks znbKfSF8J0Kb?#zHGYGRVT3>%HBp%WxTzk6R$!=KoY;iu2KbW;kAZf5(7Gf!^#jpA$ z39}yhQtH370PZIaIR14p?(d${pERfpZ*TP>)mXm8>}TTS^SMLw`IMI=tEY-&P8$UY z3-I3Bxg-e6EEr0Rr@{si57s6oDx!oWo~`%RGObT;y;W@!q46mHA0hxk)LF;6{ev6R zUvs+69|OeLAN?P^=%vOpP6q^j0w&hCP*2EXdt3;>OBio4XCzRy7 z!Sm%LRP7kwMyO2}*$HYs8Hyq(0kH;nR1CQQFgB5P*`L1V#UmsJq+3qX5tWmO0{L*I zo4^o7I$fs8XAGQn7}iMVqIi0&QHy0gMuGi=w_o}O8P2lH0YKPyOYimjQ?ktRNx$hc zA^2*W_!Bzg_s1J@AWe>RD@3ne^VtM&8soNl%38u&|2ek?TxOHE^3l6)*uW{Ds1%NM zwyh=cFGbK0Swdac9I&KIYAoriT1Zuv?Gz{yL>QpTHHGtU@DS`@~vzXz%bI9YfCP=1qgS)ledf z8PZg_{FT+(#O7%ZwX<-m*LYTu!V^1~VOi$%G{>t-WS-pjM!0TIf0sXVK)7Q*r%Pz20(oZOk_BSw$!BLR*Y862u7pX? zgK-nq75qY)J$d~Oc13LgH@C0`%PTF0RsS=qS^3PUhfl&@-+-J(m5=4dF+Zg=zrAGJ z28%ve*5acd)y4r~0e%F4(9jZ#mvNB;bd*jp&j3}MkL|bUb1Z+TL3t$jdgSZ)LP^`0@0#62tO@#_voj5@0;BF=B2A?Uawf#=I2aQxBi=5lnc(U-?dk!kO(C^58= z3!bqVjL)>qWq_JuIRRqL31Q|O&!$$xAZXoT`6L;kA)P=kzR?$@9@@HG(jP?wP!lTyDqMGk3(tH$<{2Vi^ub8XC(ROqU|I{zX0s{#}$EhHcLOGNKxh-J352`ycIi} zdWF#5Bv|O#@7M!q;XQ6Cn)aHD^-*b}_~pw;ys6br(XH$4=mY9SeSIGQtHe$vlC8)8 zjvtFK)#5FnBAtgE54u`vHpx4>k&wYYE%g?76~1#2u-hMY))(kUHvYEC_l~6H7OFTn@^E~)5Eb9{d6cB<;?utB_7G*e ziMSfM4({$e?W|m+3S(@1Tm@WwZttLp==n&3=T z&-LhO1=i0zjL>TY7$Ym6fR~pGkP|+N*gZeJ*LwVCy7P>?5Mg;XGD%E9e}WUzxRvj| zPY=vs$~G}k`-3u!S7#S6Yz%#KV`EY*At#Y`gSWkI>qhI^zwO$W;!e(Hdv7QcDF3i8 zA%$gl({%z;R`rXQos_2n_NUc12o7>F4ik`3dcZQCu?=;7DOaJ?p#S%X$NbioT{{q8 zTxwUuLu_a#G{Vp@B<1_WBGTjSt{I~jPK4zN_I$9_n0;4n?lX4WQ0p;<+*Z8$5BQ>d zWh+|17^HVOP zfNtANXQif~$NtZZIH(P{4qV;nc?mfEb?bfcT+n_YS;I8A!I%)>KXrqQW3gS%lca&h z0a&Rl%pXVuDVq#Rbz3>X5^OJmp)H)O>ixqZv4aDwPTR5xAWg@4%8we`tYYq#3;<8f z<3moRQ4l1c=|Ujf9*s|{-dY%7Twaxyk<|cV0zf**((uLotE;`K$Wqk@Awj0u5sry{ zj_(p-Os<@Ezk>kBSi(tmr5JYl>7h$Aw7mXaPYMx+un6hoD}qVaqsefh=_<2Paoe|J zM6C9UZ8lDlAgfi0Hh_GIEtpFlp-3C07>Z8D1pi(v$_R3@Qq=kq6f#$Dsa~YV??upm z3HY6QA+2?L^3>k01wz4vl&zT$jRx1(x!-6_ubGRuoGVaW)~Ci&sq|(L^cEwdk&{jvS}c(;~zNE zO-&HhbVUH9fA@H&P-j!th{c2gcrc@EjU{VT#pr~7sxXYFom`a>L{ie&9=4OoGcK>v zZxk0iwB!dwU7#_U=0kX69;7^V`VrQb(yPmiiq+MNY6XsBM4a|tk`*m~5ARJEJE$-$ zrn31kc2dx5U+b_7fGjwzmX#|7EkziRwBCjhE@OE@_Jz5uRFe;gS`|kY@5(U&pLsO1EG`+7Q_ttYB?7*k-%mMP_LOT)Y+f? zNGe1e*wDuq6ugD6M~g6VA@%JO>GeD!b(M%%~x-=W@vzw z1b>sChvU!GI$^w1J@{?&HJ#PE4rj5m#W-%AW|K?8kJfh4UA4+rG<8&(z8;XrC1N$N z2DPb#=_rS{8BkDyyb!nPUJQg%PRS;WrMaE!9?Cu@RmdRxDxl|v; zcU|;J@Oz2x>KB8W-Ht@Ag2G^gAw(8C^%9WW?H-N@UfarUgUA7^?;pb8^Mv5ya(K3=V$_nnbi4J1t_*YaXyrSfe_gBR$7%dLj| z6`z_W>lLN(A*tX71`PM&A%N}KI*J7y7aY5fk@$Kzrj4;qO(FLwr|X>T6&LXmM|2VA zNxOLqtvrDOVJ|$AtKnT#DyQqBJ5PgN-6;+rK^FeHk(acV%FjpE@(Qg_Jm>YtWV)Vw z+-(T#e9Ih%DYvKFT{DDPuk|A?qgq_DMaJ$>BBoEXP~h^)MY1@gE;O8Ll?x?Ddv-MF z`V&TThu!kxT$hvK`we^gV}5E58!;caj>`CKWPk7bn?9vRlV%&M=zv~n`UQ&3vb*NT zzJ8^5364(fEw}GC@~gNV?ztJ2dpHC%X5W-TSLtU*JWB6)bxR#mcyuXNEU((W@_^?X ztc-v;lfmaE&r#59?jkb4d-+zlL6{^6*O`W{q76xa@LsvIG zW*emK690J_wGx6(xIXitKH}gKRB?_3zKw$@BNfa7`}P({k*^_geq znIB?^rz7YOuz)OH2LjmFOYL_bUGuDc(|7;0kwz0jYCs>#R%iomnTy+<(A?)EX1GQ9 zhRgl}7Q{Y7H26dCGZj+*iX80$kONAFnss;yYQbc_6ph*B$EB%!X__cv{ru0ts9~Cn zPaF+lB|8OUeI#Ls*h@F$>jk-D5t4>dH!-AL)jJ6wxr6X1$-r z6b#x8b*GD$6=z0cz^1%AXO^AVpQtD;7J^@l-E}Zl^|^4eS0-61PmV;p#nS}7n`x;Q z8BJ6=kxA-_?$PGY&k0Oh7xR^*wD%U}KOkC+H7IBR?hNYBsVII6NN3x_;r!eybq``ER?lVn&-4AXpFGwx>0#ZQERugVLNcqsQNJ-lFlB7w_Yz8^c8mP zOA0c%cP3G0Ni(zs!*aHQ#Voph=;;n)Hkh*A3~v`u$(4a~)~um;G!S3Hu$q1?j zpAnS(LhB`Ogmlg;1iLk`j1E~>`dlSao&w~|wD&#;(a=8^MvOgJ1-(wNb~=$sOk6*y zM9avjYgR_UMUBVujln!X!yAgsgWg#EE-%RGmKb~(N@Du<`JMH2xvoS*xT%pXKO7Bv z_XAqD`%>O?s!F!fDCl;x>3E@5WmI*s&B%u=TofZbosm5Nim+3Gmd1*K*b_g zIm>q+Q(e~x*$&#%G9ephI6V*d@w(GU&?&&30jd)VsOb@qjU-tdek{jgpLSgheVPei z$V&8SantlXpYZdRP-d;92j#qxy-8p-*E0d7t8+{q2;SsH5s~{+h*_u9zaFC(AzS-} zD8hy7YzV)McpLNuBa2q`i;`MRZm>mXSFdAo6S{7lJ2E;6=SMP8-7>XGkzet2dGTEx zh7Whqj)GAW+oP$3o#}~;qVLxHYg$f~1HP}=?f-0Zok;7_-65onahoJXT4t@7}U?pY{9yqtfi>+aaItOgUY!5~V zMgH)6_t1Mhkz;i&0UU-&nXmpgyAqw<=uz;qVNe(t$a&?_0u>Uw!|;fY0bN2s^4Po! z2-c!*0#j;FuSzkCr^CM8yyn|^6Iw0S+texe=4dx=G*Y!MW*;v7I`Z^L6^Kh>z;E`o zb98v>8IqeDNe;jGn|My>->bXhsOiyuoqfMnBaTc6@VF8brPbhlA^~8*_|}j{J~uh^ z?EVmK1y7heXR^2;{#vL&X-K=#P9>`~OA2AN)862zZC^2{U_btQB~D)~u1~U&K(Mk= z@0*tZ1^U~=_{`hnscX`fe2KXs4dzI%58%sPEg)}%bkxE;SaYDwkn@`H_I6H1Jl*2` zC%EoXzylR_&Bmtt4*iFpKc{YxVyU5MECJR6<1RwdpU@qqyp)kgAfq6&2K(_2lfY&F zk1*{70G`<(ODtke&}ub@0n>rj?oe1?@zDi&z0+z;62V{eYHXo&%8>Zyc11A7sq=*L`F2>caxEq<@5kG>qq`fq=Mnu-eb^} z!04%=D3VB!;>)Y3)ix%zTzD{Gpme_6N(&Mi)~_C*?~^CFozc)&ky`6T5=IU^n6cXg zN*;FjaEPx8eY$E917~+Ycw%~e?L#Fv_=L-(BMaKS>;ODR zh(?DlUw1Ebhy4i0rsE7JJKW8ak;lt#YY>--zw`S^jgq&Fl3~`VV}t{DTik0*GUluC zdwIj-~=J}9Lv zG8#?9@ox>3Q2@#;5IvR+zA_;W0{LMtCv~HE(0)6kp%T`Ks;%7_4pj4qcE)gbGz)wD z30_(9O?Q~o2!d#MODGEa%^K;eUUn{J6_wUb6$rd?x~sVE&-Tp?2nMZ4oJOw$$gPN9 z72mL;F7V&!S|g3o$=#%&;4^QkaREu-NB(@{U+a!&c+V-dx&go6<(hqMCZF$P8+vE? zD^U?tyDVo)BTni|EFf9@eT|3D^_46YFrBJ@BSfp7e_g(=cSf*STnd}6uwrek4a4t(DLMlXkg& zUkS}*orJ!jhl81QUQi$Nan-n70PZi85Gc0vOE2;PSe^Tc_B{$70+V0o%>1`KIL?pj zVz;_jK5X9P>Xa$23QKVE(1%X%vYGzTo8LhWXz+>vLhQpb9ryFSV9^(UnhryVyv{zS z>oV!|K3ZU+iA5keKHCts>30QaxeizRzlD{{&tavzzMp{CKf_8GLE!5xL`5c0Ich<9 zY4&Le{B*0Kt|8}n^UNAS?N~UF=HEmo;)-+WkL>`ciSJuaI(}jx9H(%RPy!4oBU<^Z zyxCB4^*mo6&v@J6DiBwP0ga4$tbtgzm>xbL8iPb8lAsj*7#QMFXtPudBSOi(5@YGf0P{WRvNcZM33ZIPFIaZ9d`41yVLKPF?j*;m#!kLHTR>Ugu3qc~+ z=;DFKSMD1% zeGqj8Kq6;oDWd1hJ=XIXLh#GBhLXk?MC^;Z0G_OQk7O)9aV<)C z158M!8QG>$Odc_UkHax1IuYoj=i7h|*s&Kq20!st3Z=I+=(ozUxEo{%SWJBR;-d=n zG5q%INvP&bWI(wTPTpZukBEwTk1k;xT6p<4k@1hVd3DsOS(C=55Y9DMA=>esNN@T{e6+a-NYxKc}`}#>~jN)1t z-t0OA2CjoeK+tZlUU`edV3mR)qnqD^#!@k4m6er<(Nuul9zOm}tM& zx6y7kB#kTjX830l20$L>wX3soG9CTClkK;B4!X>2rrf;uaH7|4D8F0#TKRU(zsgku zH|+Ywatp=R?s`#Z1#$l~&2Lbn_)Xq{czkCa;(#nL!-vfICVrPxYq@6~PL#zi)}6mE zSvTCP;~$^;O!_b*jPWwVG_HvZG7y)iS|v=~By)B20LY>xzd{m<2W9^`tt{S&6ku$1 z0n>hjXWSKmE4z^3%WOBG>{i?#_xBO^DkZj9_v~`gY{u97s@XkW(#DNAbMhh5eyyg) z{g*Pkex3eQ@T_>P21#omC4HrO{f|;+2FehkAp%6QO8PaH_eVk9mpO&7gO3?78~AjVg?>?hq5VAQD-Z?Nn~V1!Q2Gg*vt z_iL`!gc=D%Q7ZAp-Ms52_5lUOWb?%ny;8z*qXzWr%OJ+mcEeF@5{W1R1~9MTXGAVx zg!P=tjlP&GY(J(?btwoK*J1W*f+Axo_-hE0g_Jgc@mDX>c2&==5zz9WK!d=V5Ab`uu+h z-K}ho)yS=jcR0HIL-}B_SQZNE=c@$Ys`sDCVvzFoSczcZDnxhIEEOC2tlf@syzjlf zLr*V`yb1X-{x4>DS)tDC6E_&5jtpqgdwKQLoMKk=se5LbON=b1U2Rx`t(zjwQ-eb3 z!+L#QUQ$zB1CU>G6VAF!j7JXxS)v!>-681F28X18df7glLMY~yuyH8c4-`zPtCLx} z05jQtMZao3s$Ls~=*k|uLjn&71R*Oa@UZd1Hz+r>bYvQ9aS*9B{z>hO^|zVvMz$vq zgZ7bPUf#X8;u7Q_)t!>j*%jsJ?nUpi;7|$oPA0mZBFtBNFg|7MCT|CJbjgX!xbJQZ z%zhH+7s-n?$AQ{TUJ;D5bbWZAsHe}!l5htV6O3`GH823MKv4|{D=b7@mW;tIVP&Xi zHRyKk<{!qH0h+rcQWxkmFNkF%t&x3lri_?3Ryl`JC7jfrtSv~{J+?T!#6HX4mrl5? z^RlziZbdk}=Z(c&?~S6*idYuXZZf0Ym@5vw`{zfx;@*a;3)QnFCQ%nTMOPqxA8~O$ zZ{sR*`Elf`vw5A)zh5{{S=&uH+IoX&(aX3#F{dRz|}btOzK@SMn8??r;FIjKn6CqG99A z7Ixb&Srvxy0Tv8oR=*GuyV+ns*n1|MQpI5Q$?CC1_F!epHq@`K8Wr&8ZzCf&o=sTH zwFnVPS{aeXLm6Kq$G<5A&4q^`Mn|WT4x|)my8zHoH&EsyDVy~ zr~Q)(p(xTTGluc^^sfOG%2{}`y#{{!1-gk}`#=enq^SR_zH%&oIoCra0AI{Wab4jd z6OHXds{|!rBsGlb_!dhpd6cxoF!1jD=RS+Dc_ScZ(4_E`H;;t`ay|M}NdE1>_86{K zkx?nvEQr!TCYAhS{Z!D`kNEdu^p|U|E|D44OBfe;TJ9DKTxwUs&&n+Im11-St3``O z9p5{F1p8l+C-?*z%)=!VRI>dZ$@M_;N5Owb87=*A<~UFG9mG#B%1%&t@IcQ03QdR- zFOXgsziR)m9IbvDFxL=C|J0(6PKhlg34KMv8ntc+AM6%)Zi|HF2>^6W*cZ(J-(f|~ zfr900qkf=dwpWm>m(Z^YjYKaqQ);)p#tMXv7jM*_JeqaLOBQ+PR2x91qFo4=q*(EYY$REPnd^|m_T$B(#Iz|(H z=w(b5i~m!!-2X|!&jWyNe5yyk<5ER9X^5j<6FO=pi|r0Z#Rbql;1dfP*j_N$=KwFy zuV^0S;lW&uYbia)cd@iRCnk?Prh%^g6gG`rO}g3T0*@50OF_58^eS4Yz^z#o$kZz> zk0wZkeT*NEovoTGe7AsRh&k$4hB-?I6!G4AakX{I>%J(ZD)Ifc+^*VR?d*B72jQH| z(|Z>*&YQTqqTl8mUKs8w_0m1#e2YgJECY)rO{stcD6E&5610>$+Q(MuG^ompf4_d& zeCcvh0y-^UOPzwFB~76REq#*H%j>VzYJ3ZXVv+2CMDFm_boB3vw!oAPiJ69UDF5pf z)9G6t;KXnt2POQE6QiERwr%EaNL!97z}*t9ViGXGk*)j}%=ZV0tm2ERVo;}s598W&FAl(lPLJ;q6wwGWBXB%qk2-a!QPVW}Edq&g^51b(iY z*8|#Cf_q>{?T{#cvs(c#tzD;Qn$Vbv!iF(_!Yuq2!gc8@& zo{e2TXL(DRE5{#G)b4%T^LGIyxB{3LzbiiK|2b)p%~unC7~dG%mF?94O8@uB1(^CDmZ67#u#7@30o)WZ z?5npMnrcZ?1?=#%VC!9vUGbLfIdNQ!9zlouu(deZC(74-N37*)!Vq-?>QW#*L<4lqJRN!Lvvm%$g+yVuGZn!Wmb2`nO^BSbu3r1 zFKIjW)rb-_Uo%NPGHr&qxF-K`}&npjWIKt=e<4A zfz%Q=JA{VmwB;;>xSSD4(U)uH21QpJ8M_9;@IkL1p1 z=B_IZ=oy1}b-YA3a3Dv2!sQ)nJ^wS+S0%SjVboUy5reE@u&#E^>`(azSXi02%0WQA zxY=?HvGB|n+E9_W6LjaG9FbE4wz&h-PJo_%o>-LyMNnBA2F_^G06TJV6nlzkj?_`}RefmF`PLw|*Skc6>zbT9kC>qc z#73IYNN_AEuj8L{-@bLitV}&q@tqrzliuV~D+ROG5y$Agy(zQt@il4}Z;y~&^bPVc@uwjcL1M42DuGf^Y=hMTEj@Po7#Xi~N}DoAxu4f# zDo&!MNoMZs5cGYe^xqZh-qZ#s8jwY&#s~+VQ!cG$>r@)&9oKp-Wag>|NY>5?0rdK) zC4-`ii1-YU36_fO|Fn9VcWybTr!?FVFa_v?f70`jrEKeoceB5?1)M8B&e^=%dWdif z<6hw{R&Q5it=Pgm+9>dp9zrTaZyB$F)+ww*~E zLo%=zcwSc7j133!|IFlZN%OtglejzWk+OnCd938J*?)7O`MGe-rOM>)$jWE}G1XUX4~SO1yb?5B4AO&4EgPNLh>xoDY33}uywUp{)oJ9BMB_n zx!bzuhpdV>rab)?gbz7eoYTm@vt#TKKoCu8NE-}yZTn4=BNC$4TTkftRQZVI_G>ot zk8vaG9*1pltx+v`m(^>G3|;)r`aRUJ0-?HXu<(oh`eOtl0sY{$7kapo!fjK0(*3Qf zDGCrvhYHchD%aQ|nl5!$TtK?iZRYps3wKuREpAd663!=2Qg1y;5pN%u2jbnL3AnWI zk1iL2cq{@QNO{su0wR-{^0L652aZh-T*@Z{7=rg-uWSu|bnqg5dGWtKNH9nRLGVcw zFJS-IXI1Eh2m>O$#wjzNQvJcCwpWsL+`3j8tEc@pe(fo4{tTfIF!^ixn~Z}CgFx16 z_fBEoOLjAhWb$fRr33=?v~>n7=^DzZ=K7N!hL!uA!3B>Y9n_vnAd(Q?hZn9hcLNOOW!Z{IM#r*mX{hTkVM1$C0cT$TkV_ZHByN5V2-wq<3{oZ$ zj#}$RX5Gw+_nJ9Yinw3{Glk=$O%3oT0Dvo>;aj>y10t$g2v7Yl9RwP_UUhPwvuk=*y{gl<4prSX9@o76Y3t zR?|c9sN=PCs>fL~{szeuQdabK7znO3CeXX-RjWps9o z%}qT|yb6>=Eh~Zsu-i050=4S*wZsp9=qCZXC=DRx@YAocsp4tG31$-uk5^2Ld+rQI zs4WsNurx!x^wMur4vu)8RYfAXwwbLmp}4+T_SGD9RCDSA>LK^GM@a0KdHM;M1{ zrvqN1mMFA;`8k%vv{_iEJ>@Q#fLOm>)sQcj4uobeBFEp{{Oj0|t=#rPKo!Ab(US(> zUCzvuDHpI`9ZbnorQ4|QlUXoqxJ^lp3J9?F)2#(5U;p@1?bJ7{nbglp=Z_Y14t?Q;Q>Dyw3f3!f}HLP8$Ly7N^& zK2>M%$}xHs5Zm2l`5e|9a{3)-d}oWfjdC}e8LYldL%fB)Z0|1DC1x$hm?<3L+-{x@ zhonMM+_5+}#I@Y8ndhde!y@{5_1`+A^4KQ0p9`!`i*v`^`c^wZ@dm#Kf!9Rx8z3Rg zAyw#PrM31%?WaEKmG`L4~Bals7C1%GNkq|hK{>h|4~oj&o9Kv@64p+^c>`20iat#oy>oFmxM+IwZ060sa?^YrDC zri^{BNdKT*C`(AKE^$00Ikpf+C>FK~0ZaC! zvbP^`gWDZFfUQ`5XSHtZHP+EW?Uc)tIZ8q6rI!w3`vyh(ilrCis0a)Sy?Z!1EXv+< zBdtvLV4m>BZ*@PTN>m~X_K31Y@I_?Q&-s5idk<(hqqS`~iG)aWNf2H1DA9ZGUGx$q zYKRg<9R@)}??g8`QKI*fXi-P+Oc=d0%;^5dIp=*}=UwajzyJJeS=Ov4$bOz@?|a|- zy081X(o)oOxTQ7eiAw_R1mX`5@Dw4`cSEzEI5BW7$loQ1;-!!cI=Sd&t7y|(sf)L^ zmbuZ-0)PgUvmBk?*sdZ|{|E2Y7+ewu;>Aov$>o4x9E5&^mUxDy&kxaJM5{neu;gi* zl;rnwMDPHGOL-7-Z!CjP(lZ3_RfHMPqBzarPf|Y^M^K?ii$<+rJ~ZEdG5sB}gh82m zygBv}{uiKMb#o>Xn%gM6Rw}nt6m;3tDY@OWmzH!)snIdbqBU63Zm?jr zqc!{QsagRgfCuQ1Ig`f46VV8eVkLP@$|$rFt0fNqbYOIGg^r!@U_)oVp@xL^ANN69 zV?;nmeq(!O_^|?EkLEAy4j7kJM#TpV2e3DhlmtVLNLiH4j^#D6&E&*8_c`Q>SD-C} zs#svEPqa<7c_OSuPHne_mFaQb%JpCU<~qA4#Lg+MtF9dMc;%Y;)VYtnBU^pL(Q!!> znaWxcyQ4rxAjhy8_0WjPn@fZ*aw}+FHTg)ezn;T_78ipJP)2&r@=h8>4hj5WEN#Bh5uKOuMC(@GQhwD2$-&(3ZMT@3PsdVr#RM7XtaPC4$8pJGx`<+?>E&CU9 z0kXEf?DY;^p7dch0BaOf_fQS>Z!ThKf3-=oj|6$felhApOGr}Bw<7_#{Bb`H)4oX$ ztf$1UuBSY9EQg9!-4q$rJUQ82&@6t>npf6rjvCsHXEQP|lbA}~GVs7`a9+j#2mcu<$>7 zU(57!WnCO-wR4?o1YiNz*#^Y=Hpp-8=F6GOvFKP~jLa{!6)cah3BP5F5LO%*bqx+i zOrdq@4&PS9($zSD3-=#tMEdkfTE1X$c>^1njvKXET^h5mSM-d3vfeT= z(6eIA=bYevppO1KU!GGd(>9TA8#22%_hS))LoT%W0g?ediu@31LucZ`1p7=w;(*em zfqE+L^uC9m^L*YetxK7C7~(Prj(U~6Ghf@1H6Ug8{r9~pa%s){RCnvmwt-fiZuAkP z#EwWzTR?BFNR{sL%k4XmM@-GmY^2exVcw?(#y}7jJhb0{X&=<8!>TE7jixaCfXB$#6iI zQ4em7luy>?)mGHSxgebV~%^I<%^4ch~8a@;b*e0*_b%2f~4xCuAlw%TYtr?cK zI6Gw4AwL_}T{RQQTnrkYQ?mnw{Grz`WgP-<)9>4u@3`@EiQC9iXZ6$02K7`N^mG?h zPu6D64#a&+8fq82&XZ_U@EE6W>Kk8%gA1c*7&QP9N(1uYk+05Io!0%Xub2Sc3Z6s) zHoY%*qCPDSkbe^d3n@Ojk9Y6CzU1j2QpGFM6ET0CNaJzGw}&<8u1}Zi%Jmh8uf~}H z;(KQ1xeP91d1}<7c9BR#JY~z}3e5aU!Tc)z1Y9tcC0ii>NbrFvb; z9ouHr%af+_+$(>RrSz*O8&$-~YRt|!`;hHdNE&a)u)|+Z`%caN`hlH}I7Hq>knqHY zzDCpLQqJI&s(!BO_>xQsOs|XRkr!|I_O>o0kRK-6RZw3^ErU|IZH45es2tOjG?!#Nro%czIde0|9A`RQS!v0OrM!^iRl+46p| zwk!W!zD#{cbG*|#j+MmEF^ND+S5{0%@a6}Wz3*?FDvfdJ3}Qj$;@92)YS0-yQ-x0# z9M7b=mG?M%P&5ViyKKr1cF3bKtJCfBwvCYlh0$Go)$UY+;8++o9-`abwFY|@;M1aZeH_SzZQ zp^iJ4J~-DZs=iMP3SrhL$$g<#K~-Tm#YR1r{2Mx`n3hS`4I@{UQUb4GgY8YpUa2JK zc6tY&z*o0RpF57Cp{KK4(T(&=krJ%s5E-#D`6sgMd|_690wei-=P zPE4PsUfkvdEqvEN5F&?P=2D2FC_sd$LE@DIf3FT@^7W)VYcC1*){RV1Xa4VyOLiIg zYtXu4tUk2mmFTrJTZG_fQ{xpsCl^Mas=aiSwV+rys%CFq__>%s ze~~qHZ#7Y^_wO5LdOPcYx{XVk9T9>U6ith~Ino-(S~pKs-HHGj>bk z{vY7O&(9FZt{doFS361HeCAFUKop5W6wHHZXK=S1f;7Bo%*H=>tvb1Mt;n&9iM58Q zhxS)MTL&(b(Xv!XCpC0|LQa?8HY34j-d~{R@BRvJoQD36c8aB*s?Kcv`pjf1x2^Om*@#KmcCpv{ zqB|21(A3bluc_(U=P{RYO2yBIopGL@1t6VKR~T5i4Tu*T{Hn5a;8}mB_+@J=5}tk` z3>Rfyk703Uld_UU#~bUHl!%6j!ox}t#b&#(VJKU+6SG}cU&45U3!l#)1wDBDSJW*q z?$r|SD2(H`?EDd(SMLPbNg%`4$=45yetOO zWI!?YIT@B@%Kp(zYZDj~Vqjxy-yO^r{g&CV@hUlSW!fGoRuzr|C3fuzv*#fnU)l(g z1XoQ>5zr}ZZ(N-tvzhyZ5+3sXW((Q2)Gzf?5%gYpd{`CG6|MVvpUdHk!0V>Db+cS= z6}S%rTEFl0S#ZW{14$EId`G#yxTxSJdT^$1{*oP|l#M3Umb)Yx;&N-;UYv;i+H9|K z5}GFRqh<3DXU|aYF-j5M?L=?+lVDm$nLP>if!XK3v;e9SOab4ok#6-0WA1}|b|q1U)g z-5$Dw+vK(<4=6}gGt>yPr<%j~W5#iSD0sAk$q+RGv+yG0Bbgc13hZhv+qpSps%TX%CRQjsga4B zyPr!21zt~Gd3yjxiZZDw>OV8#qG)+#{Qu~WOqB734cH30`@QIYSX@6a&zZk3#Qp5R^ z;fDX;;qCT=AEr2CklS4l_uhl|kHs0EqpNa?94B!eHPHwDnkyR19RAqxfXjN(Sqc5+ zIG^dawN2e;}r%bNB+3)5YXbf(=m0UsZo!-cFJUarn!lb+QN z=nTp787AO2#y7q1ptooF?UV){&JguM2h3bzaWzY%mF)ihUZhIHq5)dEr1SdYbD*L( zH2IcW19ms_q1tx6g?XG7hHVZa75J0aElr!nKgufQLaOX*F$MUuOel-BHXqG|_iA7n z@N*SV4k}9tFuW%#APmrcQF(RQ5A~{DRnNehe%+49>)+b$nsPZQPWZfAGv&7iKnW~@%Ql8U*Mz?$3UT95k}S4^CL*yi)zYx7)9k@;IhxA_{+0f~{y#~hHtqF-YB z9X(G$g7IXer=ZX~sNLtU#Q>FTl=lOgBV`uIc+gX#4d1@@A^ZBh#%)HQ#?KBp!4cM> z&m{*(?V#;IM1*;}WIPceI^bQWt>xOs7$atQ`M%TeH7s8>y=hDHn??1;c+z(oDAL|g zGXVW$mtWEWC>AOo<}L`0fLAfI5|19YQv;C2zJ3>7eMtq2Khk%nCc@)|&;9<0j6>7W-c-qsDe;8(2542?8iZhGwccHuO7X zLmxePK6$n}kQ@MLfZMaDTvi9D14=gWKYwVf7EkKY$6NDyem(T8U?4yINm$^_vB+UR z`}rlWS^Ut+#uDm_-b@(hekoA>(`*xM*?C)Hk5~-P7yFAP!`QX$hNiCRw>K`A?24SF z{N%Z&4A_IxMZ)QKK<;7_MWh>HvQcCbbczq@DWa>_JjY9o;sjlvc+kYb^YOZEcBl^r z=$?842#jRa;`bl!b1k?jvMGQxX2IK6Y@lZNL%c&leW42R*)aiVpI?(luCK+kxU%Ca2105w=Vf$}=~;B4Y93ZW6MZTzYAxf5Amwe>2K#1e9S-{+dpm4M z%9O_lvj{2iBR{piUVNuGQz>Hl{JqzB%~OUP8_*GUP?x?Vn?4!s$_nbsCG^Fl>tmk# z;ack5j7#ULk+0zojNQGG01X5mx)tni#oLo*c^RbS{LC_ANSc$Gd{j#ESi$I+8b))0 zoTiocaZSuQl>;bU&>aF$>pobiX6PUL(;h+9Kv7+FBmzEXWzQ79Gah;{?apmBkWiW* z=w%dmQY=lbHiZMxbH|8NZfIU`yKigG#PCjWwc2e|IGJ}zs5y|k~IDgF^aFAqB6v??@AM;R_W;Rq8$XV79vx@gXzIO3&C=&wSWp zTN4mYUPHkb|B9*_*7}N>f5Ur9qe|>PhE5cof0#WJFYx4l-qQ0kGojJuR z`V431c+oucRE*!PQ*f8dS`;#B*}rX%Y+h_J{7cHb{}<&CAO{;Ln#I~1Ui`5AFKOi> z-(iW9uOabsA!IverlkLqOp5>Qdvy;op5U(HB>O)lvwIIXXzu~aE$``LN>XaKqy?NU z3VuJzfRyyu;E+Rt!gTST7V&eJ7^WYIK0m!L5TYeg_;L-G4buEv{=wK& zLQ?LN?Ky@IIy(R%;1#czulW#J3LmJJE;p94rTO}FXf@^cW8V67QPIezuu<@$Dj^Lu3*JFC>Y$h zsADk~I#p|}RK_=WF622F%WcNkGxsm3Y4hSEa;Jg&!yomK7wbLp+kO?xwYv(Lcw1h* z;mfz`sxK#g=+k{)ezg~1Ov*1#QZg?k7fF$qSUP9Gb#fm0bhV8EEye^Z;ipr25vBL<x*UI}oMbcWnCM(f|zo*|7JJl7tzxiW#t>_4#Ke#TcV9$ z#u_{n0zE1&>`q|8IzJt^>1t@0n3|Pg+ab3e&Z5==G;F+1mx4DMaCpR)iYZ)S-BI5vAL3?A%$6`_>opS+9&7QXv#L?V}Rr|1u6nsU$ zo<+I28hgU{YQSNmum0MmFOGpPPbpO=bE+Ggw4|5#l-#0clJz_zA#mr!ZMC>=g6o{l zi{3=E@no{ZfZz4{vaWCXGY>Y2m;TBscUavD+|uUdzKdX$@2&B5A-JP^QfMqOVw*`N zgF+pm|9L48d*`PcO(TO!%Mo9UI0cI6eLsr?vKThF>Jdwm)*)N3|H{suQMb z6=?)f_+Gxf8dMc|w!y`-mYLV8&HAohF9}NXz2=9mK8!t^i7Ta5N4I;)q+aw}4leQO z%fVzJlY$lNcUAc+BFC6YM^!7?=ap=0JqkRsQ`4X<7XTQZLuhR zAmR}b)$#kB;MR%0E2!nqp3XG2mErw*1~qVQ4;+9TCYcvY)k6sf(oVj5;oywfEPRVv z1{fv@e_Ow_QGa68sb;LvrqCKIau0A{)R($C#~r^o%S>%Fht@m0PRj%;ToOUAY{px50~H1CBVxZ8J(@D<>B} zuY~i|Lb>TH3gE!`gDH}z->vWz*jA{( zZmi`qoD*d=Wljqwx{{Aj+2*sACThJ{9JBbMLs#+O4$I_|456R;R%mN8zrg=M-}n4K z04JH}tiN)TTHjZ>;lEBmg!B}8!W;47bXE`9C7H?!Hm?HeW*D~t`Q998LIEu@n(Ei_ z9+jBa+1cR&cXfR88HO?PG41LCq|f?QL#jigLiR@98T{40=__KKOR3<%PqIVCUCe9B zH?d+j;*3?_#PYHFr}Z+V{sGbmu#f#osp|+DOr6wx(FJ8Xi0XgN+-gdap6rxn6&Mdc zGN*N{YF7*oS@kGM%#n?NC}^^g)*ZPL;TpX|uM6w4wUWVipXulE-H9`gkv6)De?Qm! zD9M^}oi|?8-fKI>S+v=4y|y$quZG)~5y!Qq^iS<`v0YQ3#Kdj6_1RE$C^bhGry!Ot zbQ_Q*q|7pK-O|V3MJ18_ydmzrxVk;STwdZi3<jmw)iR{+v-E}E%g&8e*)xfzu! zwOrPdEdA5miIPlAuf|8is6Q48XD_q?a?iH2m)hwOHqS_yU?0HV)=*LH1V7~rv`efe z0iW}%)?E8-9YyYm_F~!U-;Xnv^uQKL$_)|V8#qcw#G`G2l#+Ci2q=g$gwR?fdJpDh zihdIbjDLM2nI@`=^-v?PQFTfHILDm6y9!rOmNgF27%12Zo2ZWALGgvXc!;GrC@fW4 zWzB*7`TN&1e^ni5P0ZF=w?Fo85F@09MDNtd@Z{BwzVw5q#maE07se_X+`mRnC)9)< zGsG_o0-`NJvJp6sCw8w+hu?SDoQLZ-77NumLa<0U5`mule`iS3~821>xWh zNLE+#JDa4^yT0rXdzuLfM%pKcm$H=`{IUa0s+tGwH%;|~S73FlL@&t4p@b4PX~p67 z!e1Q=J0;g#*SC)VuQ>gy>bSz++^1{a1MRD7shr6fe-7&#QuP>l+}^cIy&b+k_w`EO zLTJZk%%Pb*WY@mju#Qa&v1%Gq1N->f^3>(`dcJyG)nC-l z{dl}@u%5w+_@qNiT8Ng>M6mRevy5J)xN(G9v!cr-P|(m@&Q{;lr>!vEo2SAzFRXV! z_S51nEw>(PXp1;~5sB0kGjrUxOA~WbZYSj}h^pPKsH@_&XAg4`P`ha2AzfA_qiijY z-Mj+vyI3mzjnS{43)t2mXhZbGjb{BJgh`pcJV}4g3ZESt$)_VEq z#4Fb~pLnR;)UW?k3e(<`2K~C#4tGtjLUs4$c!MSb=$Xk}1&hmJtR>(!yU%~}v(54J zO!as2Ie7pAWSEdtJkpCRR=H&b7F*wD=%K;P<43Dow|TziM3?e$831bZWtx(B5yWcu zv5-b&CzBbzbz4ICrlN{x<&|H^egMsvJu|r&q=GwUY zFj=VT2ZLfN>tPi)tu{BR zubnC${#U6Gs2QH_6993v4YcdDcK1BO2B1Hy4>UN#I4xE@crJFCdJIgGnCT`wDhO8l zQ+p0QG-i7jBQtt$O3DbJS&NOX{of|cm(Y;8^2F*(=E4|ET8WMM zB}XBY2QND;do?USA8%?hjB)k-?5DVlXA(o7ML5sal2T8@KY@rjY`J%5TIS9xBH$(7 z9|ac5trXm!5YnquSh;2Ln5V2Z^fNhBU!3-^yXZLdMn&{YBQT-d6l~wQ%o72#+^;wk z&)6D8yT!5qpX4{+f@zoC^vje<%sTi$rcwdfT}qn^kg__F2)hd#jC?t+UWV~lYbsVv zL3O@JmlgT;S8+Cl-QCVo+Re6r&z)VGl{jh+Nt+%w!a+nY7KpH`p7_z0+s}7X!O=bM z>B72n4~6EsJxQOD;0CCpW>{76A!^D~yZ>6JFz)UdUpnxw&`?^ZKYH82k2efSQ3mt8t(Y*kuQ=yZ?Ps^DcMd%ct zJ)LD2Ln08n>DsSPDuDKP^(tLMTX-O=qFr%gdHdCh~nnBfD zo~8Mjlud6`l1#?y5|R5MaxW2jB~yc_G=O>f(Gf!f9Yq;{j-3y=S+_cQ z=6emr_Q*<&UyRf6j5ST6*1QxScgaSnF01e+k^!bH_8M^3WU?;dM|*^`H8!O!vUyq= z>jUjo6Kbg(uW)oj##BAgmQcjBg zu_=R~Ii24p5@gzC{;bLHe4~FIfs%A_0~ z#lHjl8S~bM{G}0FQqmuwJ{~_3SqJQL45BQPNqOzEH@U3!V9Gqg3lmhPBz{qHGT-Jm`X~+(V_vx54vSF6s=puXpUzWzvSAd?3%rvn$boRp)5$`EUhZrhUL)r}_zP_yLUVGinZp z<{9DPn=pAmv{zB;D=zMqivTssUB%djYy~2k0*3DMGXj)!ZwldL_|NyRK5QfH8zT^& z&zZHNZ2{{75geDaB29V?@M^*g#icjVN=W#khj}gC77t>jGPP}1+2;qV^&eeZ3c`@%}F@Tq=ppD7(D%+`u#FGz6^O?Rwc^pE!`r=<~2bmQ!>!s~CaH zkO@9#N*6-$^9anMYdT&62xn1~Xj&P+Z zGX+(2O&33ZHHdv&L!D#SGs;FK^{k%?|OX)#d%NN*OcR zToqc@dcNgR$^<$#?bufv%Fvz?OXn?Xx<1!BG@!R0`kG-adL~VJXw~=qEos}4h%enE zUM+r~HNviA3ILBVjJqs4wA<1x0vTA%5>cmZUcD+x>ja#!Hqk7P`%yg|#rulicvH28 z7K$kHAwK-(cm>rJvpwCE1rsj*B46j-p)Kbm-MdVpXkwJDr^`xLJIw$GEn%2BZck|k z`VHv9N(TBy%dn|-zLFg3i}lWVM^@aoNn}Z~K1<|RuusM_C*at!>Nu!N@H;!GWXUZt$ zV{Qcvh-0}mHZz^&YlBLJLNgL7j3bs)#_;s&%%Bn5w+MW$ z%tCACzZvut_QScdo&tp+{tAu3{tZ4mBU%li4rsJ?aD6uZeYMLrJt(3lic)v6f%79Y z5xE}SlJ=oIWNtmP{j|8$xb&i}WNU#x zmN@u=;sw2LPWbPJeb0kmc9>fs&>^8w-a69O@@sX95DE>+t5ffSwZ=Du;58rA!lN!E zh9W?}utn2Pz*q`@pZRMbVQvt$)q44MA%`jc?>az&9rgivi!|x0JL&kK)n`d(&%5Vo z!$|iuEEl~IgE2qR$)})?TpcS-H>f8l(lVY}J!rF6JNhfSuc^Hc6xrUj-8iMME+{T{ zI`^8B?3uTzQ}mwce@Z?7+72I(_lbtp_CSZr$5F-m6R0dPt+swkC(>92pWo@JFwQ~Y z7G#ap?9K1SDI}XR!TFrBS;Q~GbK#)!BsvqRYgJ^a7y6NV?h5xc$~l7|sf~I*YZ31G zt`C|bM==P&J<%oP(=L0bklpn;dDv3+BE55)I4u()oT!>tFZ!L;Z~)e3=EZS6$O0)> z@Je!I0xa1~XuJ79eCveU{2>!K9y>i(dnUHZu0PJ#oSVFY!=n!ES-K^_$yKO* zK-+tJCI7qY31DmHKBazk#MQg)m~h$JYpw~7JgiTJZJu37h{|nx1Z0C#b$rtu??gA}^ytQi1)Z%2nH$|Il(oU>Rxl1KWMP5n5=RngU3T_s4 zK4@?K7WbIn!r$oM!yy+yaOuLL<9u-Ev`|j~;47|l#@cC`l6ZdV%+!pa*8w7E2LoV0 z-1eJn6p2!SYUU)#&U`DK0@QDBmjz#_h=7>e<4vAIdZJ(M(UE-FiQw5tPv;`{G#WfXbh|q9V<*>vv3drfM!vknOneeaU?)Sn znKH7r-CyLUv!E`hQ_OIm=WjAnjq#8h1L>A96y&a*A<`^g5xe+QlS$BhM;s9UKT6=` zAUwlYgow!xy#87#ahKZD|L~UikeA6WBNO%lV(5VBf$4oS?PNPwo9-oc`K40q=nbm@ zzX_-}?0G>aBbSbAn+D<+PlHBzNgYB`=%Bmb-VP})V%Eu6ta0|piyp=`>p}_Qo&z?i z8izS!Yt1@U`>k<~pL@I}ooII+<2`>p0KHbbw?ANxtsY2`6UYzTM zlWr_J1UUnG;}?oUXj3bSt&Mw7?A>;e*X$fBxO-_s*QGa+bB>x{#vhf@uN-QyE&tnJ z_R|F>jwjvM)4S8@h6%WMRlvrl%KPxku!$SM*NkDABEbD6epTt-B zl!t6ZBwy4rF8}DRR$2_jv$W{FVADuzy*~eFJF~29JDXck?jlEX2P_0$h<_mUFkoIa zTVY;+q@bg#G-dX^z8o*|ne#Y|PA9hPYk0q@d~d2Vx!>2GeOBV~-J#3+(8yoFqyn(h zvv~3r@aj;8CNV3p8gNe;=Pwt0UU_l{qP+KD4_Ga$8PCn`KA%er;Nl5o^9H}=KgQzk z@v0piDdv5HW^}$yVA=OQ<{vx$>!&1l`#bJ-c3kpdUIPA^#~rw0>V4Z~Iar%x437hW z=_#SgLPGTAV{ACPG1iYrH$18DwIzTetGw9t>8m`Iq)3Nm>c3sk7mrf!z8R`$APqW?)35yB zuTx>-10%yU!qR$lXJk~Hw^uif2*)4ax%wZDH|-YmkbZRWZ-e-20?b?p)7=;FpM4fT zq5}4eqk7dsx;u+qoaT1GG3_~e|2`w{Mv4F1=L315(=w_ZkNnOp|82VsA~%yK-7|M* z-G4M>+S^;EgMAE6xjO}E@7)h=Sj%72T!|JjH=Xn>a|>@QP$cg^Uc1COFC zeEVC~zun${@|Ww6ZgJ58U*1>J+&(_MK<^Hx}5b*d(Z5-nolA9SZ+n<{d5P?Zi5J zkmYe_Z2q~o`B$!wkGjyEtN+o6@8jVC?;jJBsgiKbdO49F3&{ z1|0iWMk3+vgpIvz#)uuN`L>^OdA{u!&uOXHh@Abn+*C-LbhEbr+W0e)6HnbSs8yp# zH+6CW2ONnfNvk%2GZciND{M>zHBS!6fB;02Ie<6@s1g06J0>D5K$p!Qg+q~ztD2Ve z*=Fo*@@QV!JfTV_7sAVNIfZt)OLF88INK+yeONd@`8=Vm`*uOF=EOzR_^I?7)u$l5 zQ-A`hlFF1C*5p28(s;NV)zyszdVNk0 z;qi1XR>kTNZaM~2xof}>?!(A3U1x44YjoybGd!}N#0>l4JA#8TJ$B`lMISJ7DCA&_! zDY_(oii78v^x*%v+x>F}1yTdckeHwU;hm{>?{>}4*qh$`>ZrC~Hx>lWmFK%tu#?_= zc_&s8q@G)`ao&SGwItr&2d3sZo(Q6?^V_aFZnpt>m@B!k1@U0o3HpNje#EH7}+bmn%4%UoN=AMlx?M;vAHHUgZHkk`3Bo zi`y4~7S33g37bY}Jk!Qns$(STD4WSIsKF%j!oWIaX$=cMNDZ*!$_1z=8>b+*-K=UG z$I$%Khe}~!_Fw8bS$Gd?9TrqoQh281>Kqp3lQ}K1@VS1@ZMTr{R(S|Bn!g3%|BGBx zyC3P*bqf(LalZiSfV3y#`8UD@m?J?kj?)vh-$HCxCqy;-!*4MK zCVPrjhjtCer3pxucaq4(u8s*su9W>w3S&BoEJBu@A1mQ~lc!3lCE`GMME# z4q~(HsU5`ci5f*2hrsT~_9NIcQXf`g{J*SNe_91#iPL)Ot25nM;{Lgkw+r8yⅆ( z*Gj+9`+2E;n}9|4t^p4un9`eS<*-xqKrt@A+WtcI$4kdjk2s3@8_|z*{Z9flj6YC2 zR6lGEb>_93mS#1mne5tIY)Buadw<`&+RiM2LbAv-7I2u%1x%MeMo9u(dtA|PtS`Ae zejefQCsh_%FNRtw#c|`mXn79E6O_RNoy?0qElIxLPJI`ov4Q$+tV5=J?*UMtG~OS4 z$p5H^1`rs{>+BTUG~XxZ7e~B!NKalL8>qq3`bI`9>;-GS&wguamKm$uslw;=(Y*db z5OlH8EjeMmJRACaGFt#CecJarV)J}k3ZxmJ&KzPqZlO{7Vb<6zlO|9jsa)LI^V$ed z=eiqX{`cp*~o39?);wzlM=Zm-bFS=;}vSC4<5i*`FqqLEVSl_IWqs)^G zdXP~#KZ$=>M)tri*IV7wMJbb7WTMcRbvZ=h4Icj6w8ys(r(7ZJLEI_Mh~NZl%YpENtnC+gsWMuC$A=6XD~H`< zeH)4Ld}cQcs(y%~v1-sWHs`5dyl^UvaTR$6$@{MNZue9BGS zZ5qq;HO}Vd%Y!(O8;r8@SwB|npeSf2?tMh3>W4ih)vf7yw4uN3B7Duy`dzF?cN6IK zJ<=ASJ%#)A zcA|upYd%h8-?ESiYX22p^Z~sWyM@Mx$mSc-tCJ~MypXOcv))fCl?>qot06Bb&xyvQ zMdOBfj;qC%i;Wh*`fNpx@@*|f<%p)*>d0YcFXaIikJ~XLvrdV;`fOO+wW5&w4xj`? z>b~fSA#yw%;DkirWF%juyH9SItF8N_Xc#4hBKPLxE>QagCg28nT#-{MfNYpCny0k= zJ;NP*e8@Cg<3iK|6b3DO(~>g=lH)RbYW@1)_pXk|pr5CV+(!d&t)3OEPB zJ3fU+b%pc`-4fjZJEI?=D7*#d{=sVdIXA(BMbKxXzU|K`l0i>`RuQ*0mB?M8%>Y6M zKo?CE0z~a>j4I0Le7{ znEhB5Qz+42YFIZ$>3fM^&F-2q9>1U5EOB8G5U~%wJd|tkJpF^8(qV88xAX0k;JR zx6P<}3=Q;A0?FXlOi)}u!b@d9uu>1WYxYsJ;sxQO1F2&mn;3C$1%ps6N>h1?GEtUL zRX~XYvR$rbNk6u2JlB+bAEpZk2;^x}X69>JtB#x$l@);YT9r;jx{iHVofO*^IGWRMN_Xl|2JbmT;1UN(e3KP|`^d;%K-Oe$J z(E)efH`9EQX#A?y=K{xvMJ;3gwkuh%+ofu3$6DaJl27-;FILdl#&nsH`io?3$`sv` zF4d*mrpOi$awM2RZ2W2-NWxSAshc-|BP)Bp89YZ0Rxnnew#|ItBl2k&uf-V117@Wl zi>Ii}mK(01uG_6S6MP+lEap941O;ndD9A*Krlj{JuoG48FWW)Wh2%gmpCnc91IU)r z&pLq(NM(k~1mFu=J@d-ZX9ny7aat`18XzH2@yWEoGvJrZ(}WZJt|+gaNKq)-vv zZyLGs_}0GMD66yq@15zKSwUDoW%{{%6s{-BZ}Wn|)LJ~)Q-KVRd8c_g830Z5+(mhz zcI5ysM0(^Jw>y%w+Pp_~0=8GM&*oi_nQ*4%+Y5IrXkS0@fB9-p6aM$M<7~+Hf)xNb z-ik0#WbG91FRz!UMh{cuu@yP3l^Hc{0!VEq91vYH$`+bv>4utu#Ar@XT)YXQh}MgJ zUHCM{h7r%uIx2|ERylFkAEOYE{*8pXZ(ii?d@>Il&C84$LyGB6EuU84(l*qDk1z5s z*TIgOrga28Hr^QhXnwgqa2Dw0=%EvV3;n@yrvHSGxAw`=&+Hx*5OO#KcLJfbFz3y> z*Mbu5%Rg@^m^hJ77j&JTUlenD7a>x+jYRc#h1ojCq)yZ{2OOfY!7q`eU z(ht1{?;SRcQ@g6Po^C1mTqa`iY4T3wpI(3Va)--%tL@FU=o-~=or?tYEl-H%c#O@RIec`z=iDi`eAc4@I^RX(@F<-_#!IQ1>K zWX`v2;B@!v!byamm&G zF{OTiLmse zOrNVYxR^Px8OddQUj5;Ob&4Qj$t`~c^@gHD+(g~y!K5lF{Qi3l z>keE(EkoWyu6x-sLdXEVc@H$~mv6*L647Vlu1=S`l!~nPMUe}WZvX03DY18PuDz3;XHCav@vT+^%;lEJM9!}{y2^NbN*`(<+gBqgcr-}4W}^;y zN#mZgA2y3}yY@-b(deuG&~B;>_P@FqeafztOQ)2i5tVVD?ddwy^;vzzFIZed=BR#W zpSc7#ie9&Kn}-A1_9;;-^YTVB_cQ7X`R+ z@N^-!EJ0V>O%+YDS3vKgkgJ^a61ersL|6RqVK!<+o6 zh^};|E<_SK+Ia|13(W;O0Y*M)GvIt&)R%E*`c(uoMOMWeJ=QLGuBbHz^j!9=0CWUp zv~$$c*VCVsiOtsAr_e`s1Q3#B*o!S6ehFI+pBL7Q**zufj;IOuDt&Al6l6r$-KILl zED2k&IE#_INbyCrZ8WY}wPJSc=1jo3Z6O)XK?wH*b1;UM)2WJ6&T$xWLFvxh_9AJ_ph^r}B# z0{)Lv^LbMpC)8yAT|Z7pc|y4&-$Tg6>JrO6`VWm*R3+Hl59&7)=m#ac*HA%IJ&_|{ zs6ezYQa^7mYN`#q7o(r6vI2UKtpB)G_v_WTv!DF6mIA!kHdlfvgOuG)|6Wu1W66>6 z*kY-qaPnBZq@cul$^T}EpK-Rf=XVXj#_sY&!9x#D$a3h}FnLH6{`7qnIJg1;?Cr_j z>hFsbrQjApZl7+JmN^B+)pa4i_@$HN*PG3e0kq`o`da)Re}<#&Ghpzkaw93ecfKvE zFLn`7$2JC6Bqt8wQcuWi#QWy0bPvvMq49!XtjciZTR45SlnQSL%z(GdL~1JV`#m~F zeBS?ke7c)BQ-$1?WI{}+v~cfc%}%#jbNwYe=QwiVVkGFPSRu)vi~2<9>IY~m+UJ9V zw~d8?SazoIS4(zUZ(w5zI^rjLO&Ep16nZ(hqtA6bumkv^O`;pyrHSsHa5(+E0(5Y- z>OVLs6&{T$X(z;7rO#>-f)Zl>uDrD!U18$3{oI}v6CtdW#GEHAbd~-8aCX*lQMT*8 ze+^Jk6i_-vknV2j?iLu7ZloE80V$;s7`miWkdm4qq@=q`x*4e%n)9&M-k-yM*V*gr zbN;*4`8f03&mC9%uI~kRW~n+@K2qr@u=|VKeww>0oq)wCZ=@DqLH=ivc5%T}YUMB> z!r%Da(&~lo83Scei1R#rp?mwV^8FO z`tVmO35wtc=O@dWZ%cqSSBbWVpN_>pLG6xCG>#Kk2rx4G%HW}^I_OpmH~aXv6E6Kb z`mpV;)EETN+4&S#+cUEQv8-^c#< z2?uLGF++f|TbFgVuSnSX{ik+%-CRj71qKay2Vb10NJpzDa)Mv=SCF>p>MV()1luqn zc7fe~Go3(PMYRVVqV>d*WPLyU4+O@?6PbZ7%!e)Nyz46_((QjAzzw&n}fN(~1^M^2JQw0t+Ep={+IgpJ1h-)m~EHR&GY9yYWL z6t>FVg^HmCsFRl(k$w;M9(SW?wFyVg1Lo+H&dWkqdzc(8`RJpUITe)zY}(iv^(&;* zK!;jlqh;4`5K$xA+5^+!#EzMg3V$^(nLPCtvFJ}EiRF;wzML;>x|9J*d29MVPpx!a z=4@rK4x5ZMg;|Yq+wlcy%<4R*=kW7HF1wgYPAzq7t^h4ez_BDeC93u_RpZ+iY#iy* z8nqLm!H9P+d`c5lKovBQq`C}FMz*)YDoHgT8fguQ_}r_qJDXMx41?Xts+PXhz}Bmt zCyyjmTCxlo(F5fL=sZJ3F-1-OZqr97=B~RFL1Baj6ETNwz)8HCG!XBB-ox59>I6;` zK=+=I0pP2tqB@_h>3Lnj1&*X5J86?a@${$gJ@n~mvdH->8)okm)LN}i>ubf}wq?Le z7R1c0bff`bjd0c4r_7-@qEw@7;qC7eT6=t_7y7OYFO$m)ALi1qu_7O>N{7=)>(x2O z&G+q^j&~iQpL(T3>%Ih-LA*smdJo|ta2M9Xv;KGS)V)YFeFuhC_zOv`MXL8379i9-~mdV zan*15iL>fH5eL979`~(DkzMpV zMKNBxtPbhD=3cZ#`0NUCJpaIFH947V7|?F4TI`X@Wkben4@%?*`YQvL>P2s3TVDfF z{vp@Xm;=vY|4@9U?ZyDKlUVzFf&N2Z1#th{vPRnPmeYs^k3nU=xtv((!;7%Ig;+rd z`Gc_%Bi*LTAg^mt^q7EGsop!XSwLpd7W`I(svRQBFrN;jIYeRJJsfDL05`I^Ey zvz9(tz9x8GV$c-HNPiFdl$pL@;%hW${jADyVIHU-Psj7*^+B2aeBWB{;SaY=>e$>- z{;J(WAV_fy(FBWa62bEo=UlegO135=4=w`n@@dt!|MC-`; zI+6aJ9n?vT7_x>Gkup#9w#G)hjQgqKJ+5aX{88x2#ocDV%c>2PrTy@|-bn0C{-cet z&+*LqMM1BHemI1fYB6!SRuaA6O&%lg=o;AO97pG|yFirN_@lP_*3Hmz%~^rZA~U}J z40^n!8J34I*1Fq(mzzF;8Ol7HDD31-hI}ylL;tHbkd2Xr0YTx>j;I}hP!YMH$9aR1 z*zkO%bSb_orhb6@oKetN?^s8E6p^|g;`|)af$z&6#SFp-f@_2h9HpdIR z@dvl)%K1$*Hvn8=)U4MU+_&Px#HLUJ3f2|vMjK3~hQCiGb)@yan~G_V_b9%(ch69i zZvYQ^9Ng<;nulj9O3W)(_MmUOl_CphYW9NnF;Eq@Ygvxh{(Um!9z=m&CkwWR-T|qw zdS8p+>r!L*Tsa9Pehsv<9>EMZ71&*(?J2MR0E_Ujkdn3Uc9=(Xid> zXY~Mml12Mxt@COR#_?(@C7Q#$TW8RO^ISjy+_hNO>=v=g9~{vvAodi1teyRCx(4zd z7HjP{{fKFgVE;>LWK{UA|jbtZE2Y_{DwA>%}T7$V+| z&d{{iLchy(8{yz(&xfmLRmjP%gR5%a1&8|mjzJ_d_0~J}t1OH=i?1?VN+iSL?U{$7 zP8(mYriwykz_$7hA<1B0Zc|N7dUId*Mo|a<)noVlER3T9yyPG(P{%`*u&3dlBQA`i zJ>c7kxh#jiWVX7&fu_hdP8lmaCM)!`p!8C{U8D(K=Riyl&qZU0{{%pv!ZdG*#nCF! zl192D!^BHz#nx*arg3DYel1fGq*!#rIOrFwcumIU=f)R(ug@u0estbxT$s!wi-VPZd`qPPQ>y(co6y(*&TEfW`%+oLw%Oh0+2wbXI(IXMwYn z+b6xwr$byL{WN{%tB=vDbaC7c3l#I`A<>7D^Zj#oHKOd!B z1DU@9`5qPVpn2%T?vZOcU3F)qvVjLKCZx=*ipv>Fc;WTtsIj zq5|lhqXh;rF?`;b72uKd7{c?Kark8`eNcxjeI;P+Y9$h?+9M_SAd^1lbnY~$x zv_Q{P(d6IK^yVbGLf~j2-zx(~z1rZ?r;w|yg)&uc?XcMKVvZsP*w#of9pc2Knca{x z-qzgrN;#2V9&Gg3^Rxo}IW@lXIbdXr-4#6qGoQ*d91SDOanN0}s+-i39DK-Zz6931 z&lFx?lqA=mMUoF~CeR}_MZ7RelY*mlzkwuogi{tS%2^{6sgdS|edL51r1@!|fB)bKn3)2A?bODJOrZ9rS_>JSP|{qy z_?a&t7c&)30qMdLSnNpse0S=7&rS32cX`WhH4+S9!b&k){$~X5SNaHX3Al(6Zo9M5HdDfG$(wUO z5K!b^zM$Loc1~tTy8U$Z#C(ZoNv$t(@2gXee%_4RCLe!eERMo_z65zq!Yg~m*+7Pv z4!G|eJnWSGfjszNuVA5P;VlJCObq{GJg75o=m?_m4A_enT3^-PUq%}Klm6t;~1V6?JxQ~o)fUg!7Qm9Kx>lYiaqN(0Q7u1l<}tJ@lteFebO@oZ1|omA;37v?@! z;_R~G2?N<1PH@tgw65<#9~hVvieJ_bvG6r;=~Zz?bmEUZEC-@vKqbAoQL?}*_c+G5 zLgNgO4$I`@5!0yk(HRw)UGFziPu!Q`H^CaUE%Dv4~#fWF#tB9dz0mBs@X z226eC@j{Kd7+$rQWq_+&1HzYF=#FD#EEhKxP%=q%K!2phl{#wvWc~oP8$CyZX`eZq zG%+H2uYOY`*E3{k6o@I_5CDWpW1P%^95DPLrc!n%Ynq^%{i;9@h0>iipau1EZ-pb^ z8I~*oWggX*i?u8rktqF5fR!LF`T2oV1klCmYi|;#Kpwh+3$Q09Y^giIc8PtM)6o*# zyU^z5iK+=9hkfdp5Q@VbRmx#Q3_;Wo}j zjdt>6WggUD%FzjVxJ+U~Qz-%WH?t?c&3qM!E04bFw>Mb*pcD@}L&VaFWnp!PJ$-8dTHL6i%ezL%(eZWS%oh2H8pgTJmPH@?Irv_mr=gCCa9FM3Q_%U>H;mN0?-qQ2 z{{=l=6%f4c1D{<+EgWTKQ17z`(j2Vz+3+ZC2I}9N}qhZo&_N78$u+@04wK%Qo#%R*K52Sj2qQv?+ zV;OvZ&TXAs+-b@56~XSDBErnOdjP)IZK{zefngMeUq^+3bAzdWTDwTU!a{Cq32#T?|1qYK1*9>woyDI7Z>PU$h;DRNPu%?RN5AU} z6_>PJHjMNU4;#$DD{KcExGy6xyo^UW+~_BsM?D}{?ZwvoP!7=V>klJ)Fm3yc?d4N= z9bYyi{!Zl#g_O!U8QG`LrCEY&KL1DCfhD5# zk)-z+H0f7ibfjTA#h`%7S0fun;SJzx1)k?em*}~kVsY{C`end$BT0CG?_ihMq1CbX zrFaMW5rHH3do;bsjFPR+VU-D% z&6%s%K%fl9U-iU&k_e)di6D}pRm2n)NOI#9$>S%+OYQ&za1;%RA0Pi}>20-!?p<-NMtcLV& z2GC$X_K}Ss{)YJ2C7Pcb%P#!CFJfGZpe^G8KzUs3oDmxJL6I1&t+%|9FmR1^RU zUq9%}_JQ#6%{{DD;B?tKbr1PLV zY8W}dcWeg}E}Qz?i(--~dcc=}I#I*FlpGRvrZpuIMqx z|08Zu@vcQ;YsB%58Xk%fCbq!U{aowYmnC+eCJOd$^>Y9D@k&fEo52KB!g=<*`4MWA}GS4D4_bfAf#i5v(Ko?&6YvDKs zZDy6!wrO<$^vF|ufotRqBuy~2mp9Il1|3a@>wTGY&iBLbsb+{~-)F^loxO8;rk1O~ zc0YF+n`5bJz=X9jBn5NCV#&n@wX+_I`Q-ZMnil-WHEr{=b|MbfTf9usx;HVgq)Vxbb$j8+>btS z27BB8Q6u-AiC4a%Jel=dEPNLiv6=W05tmVtm?fA`T-|iN_#m0v8tr&pm59`saEl4WkxcnzYk6_i%X5 z+#|pbi<_yuZl6&B2ek!d$s16E|JOepSbbUCSkLnP&mk{qV43Q??vjBnez6Fo9Vc z4cqT;Kn{Ihe5HjF^JxmNupFRJgAp8bVQYH7z!_@AsOcyGOscL1UG_{EuSuNK;CJ!M zFFc5yik<@Z-M6Ah*vnV4o+Ev5OA;o&b#T5OQ5ofgp-aR;Nv?=Ol0>0&SYsaHG+b$| z{(A0U^=QY1#UCq|_4qB#!?|8@JYQ65-H{E@7qV!2pB}9Xpnh%~!L4NBLu!pL@4WQY z2v1a<^$1N6&q8%GVMp0V`2sR-VqHT!#UCIeazPqHy^t2vWPQIaz_)u_O{p!W9DO$K zd!0KSHLks+U#Fif!&AQ%%0zG3pV~dLSUT5Wz45(9OAvRYhlH5f5GzxQ-!WXpEzI4q zC`h9(nKv_%oYm){WCA`wYjL6E{z?fsDn1S_%|cEdA3JnQCSDr#61KEMW*W4&+b~!{az3>z3rcz{tkei|(8lDYr9$j~Cpo@SOP!*pC zLv6l2;!W79;=e8D`+G?9KQiq*Jfp<7QdPfWfK2;VOlG4QhTGP70%~tgb*jQLJM*5` zeOQC#0LbfG!|Kaih1Jj_&$Q$_PBBYAZ$!zB=ak_ERva*&q#zQ!NHPJ5ngxCFv?i2| zEC38*aC5in)LX+lWG&~nMOs%?#Lr=h+6}F$hh~lJAZNW@*W)+9iGM{ZdzPWPF$akVB zbPw85rx^o}2>LwcBHq308ULoW`o1St3SdtbsH9CddLM;so^H-K5@@CiUwsP9i+1B* zHciVHwULzJkh?UwS_hIBes+Gxr4sd0w1R~8IeY9y==})q#NG6gXd~7DUb@7(2ZiTQ)jWOo5S;-2^ zpq>6WQph1c6m4KKx3?owDhi2K4F4@d&lA>8GyRudho{p6^+=E=HMy$8^hyVj_Q5eD z%;sc8^9Z1u|AU7X``UXWFO(x>+hYFw>YTD*s&b&%xfLAG!9)m<5j9Gr9oDF8wF11@$MEFvaDZvE@U`+- z*$2}Djfc5;N=Y(^Gre@MKbVwzyKEqPD&;j{=IA~kd5bqdswwYHHnQ;8S{$cXP|5BR zPhPQFbU3*lY#q19p0F`53LTQWz-o#MVRBpOGB`})vfMZ$byP>U?#3QZzvxey znt3HV@=HBiJH6@p?2vZywKZFvQh#t!IiS|!?`QmnV1Q@|x5pv1wCUm8S09lrRQ+*~ z?JL{iPoDrsJStBqcOF{8CdUSJondU2AmXxmWm`wkPo%wM6;RxUqX1Y$>ckR?0xJIg zzCY}8KNr8fo1890n7T%GqUC;`BO@|-*=jT~b6JwbWZA00>PAuR8G2$V4(C!) z3m-7Q)_)UhOxV8~53aF3O7zgyi6%1H;_S=jmBM~)$GIFya{M%cNI=8L`QQx=NVv*# zb4*w83sKE<0B|dC?C&5$HkEfKLTGXe3*}?q zw0Hsqf_nG=Y1Zm-wxM5urb{!HTv_5fGEaK1`M+~H`h~lLi)svZ z9<@zRKoo*({lVfxS%C0PNHZHJ{mwDP%T&GgLh;YfDD#B&>J8>dcwHZAg9|D&xrdP^ zfF=?Lo1Al!Ezr4wv-Oe6$8Jgi>Q)?Swh<}d=H6%Pwoi*-FQ3PlUnr`DAmL@Ckiy23H$xlxjedE}1y~Xd<~J<2RHd2ll*GaNU@*pGe=5tTSwq!i5YdXKZP0 zR;nOC3@+h(F(LCQ1Bxmih>ekA7r|)@P}@4*6R~~`F^zeLimQKjfh*Zj#rsTon%^iZ zNv8T>hTErt)YPVo18<;tq8V6CmGMg)@at)acRX>z9@Lj4;k~@kvF>`;S!)UBA~xgZKnKrJw`l$82~;;jgmt z!8?U)iD<#f(sDcn`4qqcUg>zF2hW`SF!BuxG56bu_3Chdz$@`|6WFw)xo>Z5US}ql z^SPbPyhGGYWZs5{LcMF}C*=L5+4Add(p0I3z`m`tU>Wln#UuEHM=Zky1hfl*{!3~B zn5ri{`fbwGHs#2n6yeJ!msKVhU0f{l=d4W^&c=e9+W`3K3`Np1PMy|#birwfvebHw zJ8zGob50I0}X3ft;cw~+=^c{dx9p}+Q*QaHWw&R~#Any~Wb zaBl7oHeuD(?sy%y^F^Rn-sQ9Wg2Pk+m%-+FBeIAKRyWCBHgd5KzS1P8D1%7tlg)hm z(VR`Lt$Eur_neLhma<1{y6x?g#l{))3e43mK>Xxv$E6BjY}0%QzrTRUs3L zbU+*v;C*^dr;>Na@yWs1E5PiPQUWM|D`NK3Pw)b?9^JTk)78)r=MQ528P`DrT=BDE z4>Dn!aWcXmkN3u69T2M)h3Qzsdu;Ml6_8hCgtcn@-_L3qPF6nQUWC}Sky^ubdyDcLB>d@s9H z4;J6JtBM|0&E9!zk}s!>Ep!gj0(pMUrww~rpdqZv#qUOM z@_J6Z=LttMmzlQm9}1iuZ%P7{V>BH7yMyJiQBkots(4AMKZz-~fm_#FkAA{&(I<_2 zQIbnRso(U}u73AlW&Kll4x|m4elOSfoa&@CA`n;Hc1CwJnWr()GS`{o>O^M#HKrrzd%&j%v_4_gz6A)Ul}CUgukS zS}>@{?PdHqm5UOz(la{QG?yjQKgE(uOLTEBiK2NLKKWL}cU?~ypx>YYgCMDKWf7IP z=FOB9mB3#HG?)0iUfDx`uZ%+fgOcs*Z_ zi9#X|bu3E5@LJ^QX(=4{>X%i32u~jJ-ovG8x6=b;;n-t)tw!x4t-SY+Pu16lt^hu2 zR28d`$6n4Fyva_h=}y`#q$)$7uAR9U$R zGw^0e-Ym^|CkJyDi9q*(5)0at78V0P0vYw2_tw~q4fDx{Pk4%m7F^@f?mxEG&*ICY-OiMQ3Jj`7jpApb}ztHP$1dF8wzcY>X9G6 zcmLnI57syCgSAK~>h{^#c4K^Sv=j3seY@CwMsm65X$+NOce|Pe3wCz20pgR9)Y8V` z1D%{Q+A*iDnB38dW0hnol_Z`i#u<`C}n`?olwamxZJjCetq%|lv0`M9fdD@Hd;BS!DZ zN!#RtT|TSqH00#hd)kc<;H(=}33Ec z8~$ep-OxB6A$qmKb#8uqjvVtnF(M53(V?7ma< zsCDmg$SiHY3spGg8JW~54H?P~ygYwekbUup7BoB%7Q0{07Jx$sAeg^~*r~CKr*}yk zLl>WK1j?kHocB=k83+$g-VY2{?G)cP~}oa<~R^N=u-u z|I?_cjpKyF7G<|}^j5bsRSKRLVL65_@L7kg~tSfh1wplkrG1NxoZ z$K}Y}c6Q&hszTT+S^vnTb#WPE*r}}80JRV9%ltq&6eGFut`vYw8PPwDR7rmh#SR+;;V)B3;~%ymd4)qVZQHuS-WeSYZh~mtyn&XYLDyDfEN-G!@zN^G};0zbK5+ z`M{YX;*isHeze0UuOe%vQOCOBSOCj`IGT9N+ znEqGE`=y5u-DW!u<#In4on`&fQ~nj5HeRZ~-mrFL*UevcE(LA!v@lvFmtV(K%p4uc zmff6Me|sWKDNmd0gLiaDyIy`IVY3%a<9JTOHbFB9)V5wzHh1@(%QD?;Zm_dAWW`D> zIwkzv|Mc<6nqde~$tR#&w)FE5%16RcfaFy zLjTLr%`tE$`Ud;953-a990N|X0_r;UyJr#H2M8zEonO-8p=L>6zNCSyhG5t~L#6d4 zc4AWI)fyA81@PH)xUJUMepo94kVg%)@6{%xJBBt)P^`yz8oA*6H3PSL)~dz0j4+_= zS%=RCJuHUb_u@Tt*O2r@4-ZE#Nn5)97M~T3nDt8V;!*Rx@r0gLv=47v@dq^2nS@sk zWXzgawzRv@l)*Twsh$=EbZ%S;JF?KL-cl{nT{tU3m#B*6Ij?pn*FcRAua?3vy1&bZ zr|u!v9c+4J&fi+SwD$>Vnx*Xd~-ZX^2k zcC2Wtc(I<6Hikf_4QV-DhZ+@`sc zuiFV{%Tg}390>X3Oo&HXKQ&34b{f5q+EOy8wi9PD;C6m{vb$%-;r z<>fZN=U`uumT0_zI`;D^8T&cySk7G%gOls3p{(WZXq* z=fUS-UqaT4WJDlTYMZYVy{*~+^nSSbZT6dy!U(gGLyy?M>|*otN1IT~5WgLf{gD9+ zjM>;9<3;)bh!?!5!!tshp=FGH_hpOuSd+HkCdVcpA80`s;T^ZWtud&HEemyQwZ+xC zi&j;8SOUG{OB43Sz@TEhDqRKZvMzdtacb?CSIg#4lSQGgvB`ud9WQY>^cxaDS*|tiZu+9(wOM^eRVRrcBeEuJS!*e5WlE_|qM`79Updu^&J-|gH0l@u^VI>NWWs_pV| zM|!UT2||59(});W*LUw-qC3=r4pv(&SPr`;pLL!-DrvvBjpL9{4DW?Ze=u&NSI%%y zO4C2F-)oYZ;DYQ$o2FsJEVV+uFr_{^l-Ip&RYQ;l9Iv=6&Uz z5iHrgJ%u10E7qL=Rjn^9wKxEQn$3upk{8nq#0r$vR#))s2cm2cO+><@*RXMl+lRZnP*4_oQBTG~eM)LzH1aJaCcHE_!;Fjpv&c z<}A1>c@uA;Y{{m~oe-MiFUJVkG7vmUpMYJ577;8v4IiA|M@UKOAiU6zChu*o12DBM z2KPl&mqSPTsC@eJBe)~6z#4g{psvLo+W@a9h?ht7McU6!N?!?}`1?--Epna5cJv7R z=h9({&&MTk;`+g0kW}Sb{nUfagQNX?e6d3ac;KUeE=mkLy+D?&Y4p=@o^o;aIXR-l z7qli~RFDAuqQcZUgoJk<1{}W(++P;`10LX6{!Ml(PjS2G<{x8?yFiRoLxL>}`}WBg zA_M4`q$ElCh$%(9!g((%rNQwGh6Pls+>L7#d~xCLrQ56=`lhR0BeB1&6z>spS%mX@ zaA3{UT5~-vQUIhILDVVOrU5mfCv}`1HZ6KjERju`*SXIEbVa|v5j2(A>RapZR z!)3y}moXQELuLMleFKR{lBZEuhM2M80Mz8KM(aZhLGKfG0OGAXrXBdZ{E2)W0oeX? zugFj7arUVm!y-#2hH;GduvC;d=v-?!BtH(QTrXl)62FAkRTi3c;ca;_UrTWzyGQ^8 zi5EdVWe!srd{{lzVO)DH{hPs$7|)@Y>Qb#ujgZDDj-IGB?}2p07FehgQZ50H#Ah(1 z&+(8dgN#^hC2Co>KmJ}42l(@v+EznF)p+$C!=zsvQasU9ZAiy_pa!1_@Y=>iE!O+f zs$3=<1l!K9@B7HcLw!y5>S9dbJzNYRXY|gLWas3J&b&9ckiBVLroy88Jjj!!@7K9M zQ+Hgp+cN;c!`xgJl z*hv*wz_mzfEu346pNio|3;;V)=Y7}~*$jkL<&-nPD(3ut2+Udc8J6;%3%qqRK9#1n z_Uu~r`=Tag(ungT&GOf($#3`;Tul>S)MDQyd=&g8TO*FKTQbx$*%LX(LFl;j!Ueykfnn*R7@gZqfI)=(1o{{yWrQzy313_$c_cV&PvNuq)(7cUV}XrhCh=@l#X* z9$PVz48#A8*8OB4@}*sT#x8R5`yPx=)gf*oqxd?@eRq09=jv^(=TNe!y+Xy64QKdB zHVJh|j2>K(?7id!?Si*tz&u$IT;9lIj)xn%1(RZUmXybQcd73;4&(|q>X~V0WOf-L zn>x0)WiNd_nxe;3(7c-HoBX5u-wA^D+H>9#YP5<1KqTy=wUpL^JyW-h(gP0$WF}`6 zt6^<0$YDa9`c1C@DC@|&?D5hZ|Zu1k>%^aN0zgB)qi z=j0G560j+0MJXrb<t-g)O*U+>qb=^l%gTo$!uH z|9t8$y)g3709@e$nd74JZi>^}Zv5ZJZ)R_%!Jz>Y|2CN}_yM5WYnNpE`jqN-dUIDS z-RAF>0GOyGW_KJ{{`r$HcmMY}62ss(TJcN}NU)F#x(4GpuY6@rVo}ZNs_IeO)spzV zQ|IFPr&(MOvd0V@^n~fJH07JJlWyOx|HpAcI`GO8(ghT5H;9mvx$#J3(t?Wqoh0F3 zB#5^kImZTgu-vUZ+2z~Vnx8jd0J4{}{|*uLFIm?A!rSw^;T&msAKkOLy%KTqZ#LIR zqyv6 zck|xG>Aqt9AK%LV=r3{Z05%l$$3XhqpDs)QesEz8a>)7C7X0^9`nP}eTe+c0u8bKM zu-sxgG}FEYdKl7TMlSug6Zn6f>gHF#4;J>9jfWU-&jiuU=8&O^#{U2R!T-c660`po2Akac&2FB*{5<{ty^3=vZgz8eqm1J1 z-7E}{J-hF5)7<&n3i_Y_(U0b4H{-pLjlMm9U7vyJ>B=ut)4Vl3|MUM8_@v+H4M}oz z?%GP^)?~gR0-hr1PTnlcq6r)QWZ1_Y9l@Z93a!Ln^twe-NM(G9=BfxM~o2qBe`RLZACiUYbm zu}kB@Kns-oY6I1w(f~_2F{%A|6)?ekQjP9d%^CS4tvof%%AULZaWyK1V>EI8H?&Wqj2nJdOjVJNgc($oHO4d^SyXvXDKs)y&wTO1#?UG zmsoFxiRShx-dUFyLEUA9DY226IlYZ5-M(i-8Fu=TYUDAYiNFI4rdA&jhH2|EN$0CM zmxHItHB0Q$2X@K3hTjix+)`>ajgn_7xmV3IwG|$&dvXnhqVbySg>eT{2O#n9#r2av zZtpvbqr|k07rqvae0zVV$mhKM(od6ouGCir+;&(k!n`Hmb|%&xf9kC83$N*tK=p~_ zE~3a4r1~H#BYUgHJl`P#t3$JzP2Wgjw}nlgR_zO=A8eGaEMi8+_}T|nVFpiaq);0# zO9T1R)SykvAb4foF_QIq2tvg1l*8FxnW>F;ggxQHbyI*74nL-%R;%0mbHHB!_7j%zEGIN2Oe$_Y)y%9?}<24=1i7pf^dDYJF0c`^hM z(QZ~Pczya(?>>m^lI^l;E%7Xx?Iro#Th`~fmgGUHVuv5B9A$v_q@h3=DqAKX_Ljmo zM_Dt0ET^wQm2Qg*zej+6*henZUFL1g3J!pfvQfZSP}YZMeCnIdu4m%Ns`6s@l(A|-EB9>gi0F@yP3J+($T+h$@gehCtV6uLhEv{KRnm;Ch8IR z#dM-Lq*CS<*69k0P}eL{D_RZycGUA}8TNA5{HX7%bC(Ujn z_hg%$XR3aj`*XMDUdWPy$#XetXcn5S{D?koc0LNRCyxgyp*6oM_hcgxb|LK3vw#Y} z*Wp9GNMmY5d^91uA6csx9?)#)bQ?H++q?kAR6I9iYn~&{>4&$RqTfltU64LR+Y@+0 z*<%&KEUFjNE5FNAfVQfz>I|*!+z3*?E~;-l^Az=}lIS;6cdUscll9n5##TA3!*Y>2+<>wL{f;8lf z7TQ=TB^W#cTXzu}4S-ORVwT*#EfT{IHcsc2K>!F&oxR0S^>iWao(hFvG6A2+B3Gj_ z(Rb?m^(EYYl72cZXoUy-J=F2{86v(hX^7njif%LVKWqY)ncHR(oq336p07k$ymx-l zV8c0+7Q63NPKipJIUnSZ@-1SO#=QT70CcM6Nv^3Xj>TkudY+FV1jc40!-JFq=(rci zvKjZ#tBZC1TvdpGc9Tep(w5LdIr0e9(_ls?SKWePLT>E@&5_j)MgwkDR` zZff{bM7WeIrzyN~+3D+f}Xw6^ELCo#{4zM1)R$&+UREB%Gsto%(fEc&$z7 z_q?Pf1KU8k?u&joK3IR7OX789UQZPzqG}{v7JfBfan23xIZXG~_F4CD^IygIJ(}>=K57?f^&m8Vj$%e zb{Sf~5D$58XIU+M*YR3d!q-7m-uD_3T47cjUTPtV5ALge+~@{g;}||3P&S^|V{{pG zS^*TNZkHGNk~F5`&)8k?wBIZoe$D4@ArXxg5CGw?$3~y{*NlM$st(6*CA%l zN|)>B(6Wb-)c#|Vo&8+s@U5iKD%D;~>9zmG*;~i8*=^gx6sj076txP6Ji?Lwk+l{9kR zc71!_@Y)kVCIO;iLWz{ZCe_!uRXT@m*in zTtDtR>hqL}yCp9-g!4720XtwHI-^R0Xhn41kbhN7HOt~ng@8)dPj~C;8yv?wVG#@G z+;*1=OPmwXGls@jr1C~bug%xL?%J;0yrbVwdJ{#S(=Ny`?NRqn;r88^VR%4b5T@du zQgQbZ6nzUkN8>vIWiY!}5a*N~Q@gbuReCux(&Nn&6Rfc}hpDaia_OFBP?cpz`NZ)@ z@Q@Yb!%i+bST=A!?kA`sgVIHdQ_jvrE|qIYeKawDx>3x((mT{%v}Qko%Z;llja}E9 zmj$BVJX@8N3ut)h^kg6E018;?k|? zq{xu7@1J{$h@I?Dy*kf6(d;8}Q>(9!{9GttAuCi`6Yapp5bmo`5mfRIhOFi47r;?Nbk)$Zh%Dw9@USPZZ?Q`lX2GbRAECw1Or}6vuXGg*s}y zK-+t1b-F-a=ss()6HM**D%Yb~(JbhRX_^*8Zj0Ip9PAz-ABkV(TJ`Q{ULRuTvq*%K z$W7ETj4(v8*0F+a8H}h&kW%-lpiK=l>4nx}BxS_$F;HIOnXF}4Ywaku`5F*mQU+UU z@&2yjQxHAe=JI2%3#>P4|2|ph#2=!}Q-0rqN{9}#{V2IY@rhBw#7I5Q_bI+Of#1H| zaa`G-LxRc>3j|q9BjsY-Ja;X87iq|xMIfv+3b}fL2%E3QPimaaHTqp=xQlUx#d&pBXGnl+cK$3`cwjUb z{?>c?giQa1p*+IAmGag3b>9zs|-AuH<~qCbt&vifqTshU2!mM{>`zi_OK8e za@sTa(yO8zA@2(Iv7c!5Xi(vw$PC;#eviO6kzH~?+73s5db^&AJBo-?CWj%C`c|%0 z#&h1QtmS%C51pTwT{IBw)v-K7Uq4p)fhOBftUBvlqrhaM+T6&r_4sBGob+`^bEHu^ z6*3X{buYt9LD{0SZ3$i)4DGp=#Y;y$OS^Zn;2=vOG@!;;fb``9VXnB|G^V*-a%Ld` zWto1k0Tt9Lv2CYeo~Y7dUcTLr!%Yp-uc&zh{E}mqk2JQK#eKC2!Bdy$U*)SCJ&^l6h3kAzAGsFkv?y5<;nrP;h z%yOSp_1&dH`>2?XK+L#FR>)hY*>LeWi}obn2k_(h?6m-_ZNKw08S1N(LW(Yz{L?p6 z&=m#=9zzGlj32*ihzChl^%!6-MTxD`-!Ex9FJI%26@c=xkMDl6%IK$n|EO2mZe)hZ26$8mE+$x(*ymMZim8Mt zfA@s>r)PzCx2ZF$^;F}#2hQNF2&1!iC$$ew^BVe=Dj>JW_W#-ZD>MEw2RIKbr_pZY)np zz_0haU*%H3sU^gHaxY9U5vI#LtjhjP{o3Hy2Ps2ZFf+X)7`7C^e0BEX`(tG+l>lrz zZn~_YPpX?}(1*Zv{AYi*of15?>seHAR>*KPiJW=ZeEg)gT*C_{a;%gH{|@)`p3A~+ zA+hR21PFhyyFR2gB9&qrol0q3%^O^KiP6B3l8KuEarbGZEa|Hf{K#OQ?(*`1W zLNS6I1W1#SOj>M`#eOePn0C~#OJ=qDLy_;d%%LFFO@Zsm{pd8i{WC6+ zJn{f?u&(?Mg49?~l8rJM0;vEF4^w}A1w`xMtMv)iIp*Z=o6VI5SxK_EYyr(8_sX8;iV-so=3cR zM5nz6VPg`sqgKBSriu^|GX^kM0@an!*!uQ_OzJ!)XYDX(F}7rE(^upE9Swt=tIbH? zU9rN_o{yeMbizvnP0yFmVe@60CKk1DdJ9Sz$0Wo}y&o)el)!i(0IkQy`3fT7_Vw|R zjaG;IX~egJ&H9ShB*sc=VF#zehq?tWZqPu1iQy1BH{1}949L}Wv-hN02bJy4Hc1AZ z?8^|r8j?^FANQJg{)@I;*gj-q&5Q~jR^28AE>^p-s|SeWylE8lKPx(`I{Hbtzanu3 zXQj#y6`C&@Wq(qTgqMq-@L_%JtwZ`t+r{wJ`*z4^VHo_}{_7k(Eyc^4p5#d)sOy7r zV=zY0x9M3v3%UQKDHkO~3v;ICXwN?!pazd#K#a9MMQrYmOf# z#VpiVKDf{m|7`xZ_bI$;uN4}MV%+U=vTedmaacmb?4`icw=$`Acq!k=m!xqSwN5d4 zIyqi$)ShIKMnpO531_o7MGn@bIqkeEmCmW$26-92IXPjsY{i@}ojmuuy23BtO3iT$i{u@(4_n+fx~G=*$zk zEVq&D&0AaJGa<1j`THHsxQ0W2;EZGZ{9})(>2C_@` z+kP~yg|8UM%Uzjv*sXb>6cux&b6<5-9>t?g=hsf#+RjFlne=^yB{aeJ3bG=^HJ$o5 zRq4r0L3<+iWA`6*8g|e`y<;9}=T&Itd?#Yd!cJ4KFhj%7zAnr`>SV2J(TIz&{2NSj z_4xmtuol<6O>b%ataV^PQ8-DK;| zD}20LzwuqLeYCy5K>(=?xgrR5>R+-Esn&EIzkyccfip-2%6_Eqt1r$$$Qv z^1P`AO{+%Z9escjE%kZuB1k#Q&{OJ(lnQZ5@D+~vlYZWa0QAcQzBk!uS7L-{{`V*G zu2v7X&8$o@bg$H^5!}h-wfJ1d#r?SKLxufXUU-y_PaMX^ly4fRUb4ibl(?Hw3@@lO zG-<6yp+F;IknHFWT7SWJ#zd|fdYDAbNnMJgEp(dU>fK&>5>x}^@+NQ8&Pads?5~+QaOsVI zi6M8aSc{QLlVB1&$$6P{%5@`<<)|!R2RNw)D@{zGdhsqp~b5}^tsdZ`81GIA-HgK^lU;P>TKCh~gARtzA z59Gt5WijCvmOC5e$&!vhrFsx)fX6HAeRz)!-MI?>SpOY3h|j>@5?dpJ=4$J>^aqo| z_`%LZc4J3tajgMd!i2YFN|DP?%5`5OAqMf?DJSAsRv;ce`B8rP&)92_YMV3CmgR`J zMRc_q{eBEuyS~|++|c2#_fBg}amCxeOj0zh$lDH&FDw+AqzIt4lXAS~aRJCRdnBv1 z=u~vr8=$r)I>$@lQ2nBH_m5fDVUt}#>ugkeMr@D4OV+RjDs6Y@iACI5IkkU)uvsC? z;l5aZ?4KmA#0g-i{MhZY!6GT`s9cVb>xEl z6j>5VYi;&+#?G9DF7saOr5xQJ^MPFFp5?`PG>h;V-5$5hoG;at#pJlX6hdSYgy_Xh>7duo)LuDY(t$Dd3o4(f` zHsjJ?0D(TJZ@WYpw#{<+DaiZH-YxZ;Wql}r?(Fd41A0VTdN6g4lAu&wFYMU}WQJYs z%7;4JEr4&MQ%uxuA4s7&Y#-~$2HHB2QVWiECMfDM28Y@lxrz4mBBQUO$iUv2#XoaQ zAFhWKd?`Nr_ztrn9Da5|OJbaNm?X+?MV2~Cca?zpEJvRx#+3=2J-v@gD)c1VfCwqT<1%Z<-~PLETco-A>uycHFo3ddFZjymi}7yonG z^yAfIap7mIwNX#wda&v02J>IeI@UL~qYv{muk&>S_|pgSus&XeGzR9f@j31{CBIS= zbZD6%#?6aXN!l&nsy>6dpADt0zSCSfep- z?1WHG6lKLl1DSv)I$uCBUW~X9kUNg}Zd!DpfV4N|^|Rhy`mXP)8Yh&a-t{$89)x{7 z8~$pz>e$u|*8?K%7bQ=n2O3-c4NI{;q6AKd$5nvT^m4W);#xdJ=^*8cq>ekHiN6{6AgJpl94_I5 z&UlRX*~$6`yxeswW~v(taVzVbVA5TqK2#aFEUc3!BI`{j%@{nH4v_rq(F^&J>4*uP z6+vNwH=-PvaQ+ZcKGA-9HD{?V4M`ZXX5%^EF+dJ)uHqYONzMTEuo57_J;xOd!liCV z5m_L8y}^^l9bN9pOTHQXdKN2?3g^R5+g~WYumknjc6-!bM)JWr#%Nc z^YM_$@ED<-;vN6I5X^i%=7n%ev#2Gbk$`Q2aytY4g~ED^O;__WJy{&9#gKTuDw%bM zdD31U|8aiiZkH5M66f5o#w^*{$CouP10ek~Qr@pDjWXAm!}u}fbMjrWYD)IA)*gM| zru)LK{e>6p$`msFZ8RbUQR*HXUl3EUoP_;K)|bIZD0) zNAAn3J$qTKoe7@ar0S%NN>vDLB$pXgKY6c@rTm^0%L|%P)Y9%xDo%f1lP=mTGB76b zqOv{Hcu3jPRVvpt1LHw90_|H&2vCO%Y+-C=5FO`muWw5QN^PyrPrfmNS@dw0bckAV z89VM)GX8Te{QQ;&a(+T=^5PC!VD$v(KWhJktX&Y3)SW$g&As}|XgX4gaU=IFlRwSU zFXv$8iEb@kSlB9-Qzaa%PMF>?#^n5F)hR{0@_#^+E z$Z%tMoeKl8Qs263E4Ji&Se#Fs@&QM}`&Rn!38PvDLQ5VmBLWeY6F^5YVM?d*jTVZ4 z!&U;)W?+8YsV1%6heqNk;2r%-_JBNrCJ;ydu++>hpuIqM%hsvA#i;lb7U+z3a~uz4A(PZLE~Z|Nj!3d-S4%L&MaJ&DwHf+Ai$x_Y4P2IfcmWQrj$*&x8fOdEE z0M9Ly!yojlq+O`HIZo9r8zFv~{u!ftS>FES#l((>3phnz&i- z<;015;P=DmgsK=Hnz>Gv7nR<^JY4Pe$ZQGUZ~EXN+*P3+-4T0lUC*8>+q}2l65mD3O6_$o8mBb8d%>SM9P2QD(8m6W_pQ;(7$& zcH2^E%AJtO-#KW6x{sb^3jR`083|opKg*jSlxUE%e9WAkr${bHAe`~(RYEj-L@6O5 zAh*gr?rm_Gv3Ks;jpB>jQ#Lixze^|Hey)dj;%qU2dtAyx$OVVxz%SRB0sm~iV#-l6 z4&i%mRC@8cCS0c=*mi${E4L}>6_rvA9pSOLg82MX&Mo%G%hlHH#wPEq!X@!cKU#OP zsbR`}@bu8&gzgID4$QG|7eY`8GfEfUglW62)+1h>=ZtB7*Gj%II&gc<-;H$>+jf~L z*X1`0V$e1|uh6gB`T8Xfj>niL-UiNZ|6T~qPP?9R&gNX>Gq>YOdW4h4ghtLC5UnJ zd4*@+sDb5IE5kWO3*$Yc&WR=7jWZ)YqG}ypLWOW0NhO)P16Jxg0zvDURyEHWG5DAE z`Ll-sR+2Ve@1(%huSu#?>tEB~DgOnhVhbQXDsAn$myz6!pq~R4+jLLze&x*>!qVAv z#)e)Q-TbW4HRIeu2hE9tDqaozL0Vgkjy&E zJ2!aAEJ(`E^vs~H0DZN|?^O@)arLJB?_G1jcIFVZbZri&acgP4WUF%#lL#Pt4JpaU z=!Zs+iJS+AqGgEo&3$^+P*L~QXjw(;aYZvkK6BP;)((LPiobjNAZam)SREv-EEfZ`U*Ba!(Z>ZKkMSp}$#{ zpv8e(zYNZsA+6O0B^Et-^D*OX97jEiH1@}{jBSFxPfmgL8@`Li4AVL2R?WR4=#R}( zgLq^WX<@W#!g{9;YxXau>fmhezkW~a*~DsjI1509fOV1^3r7R4*B31tJpbty<4_hm@V6URj-Bk_AvR9g{VAJocrkeRPxl7G@Wx;Yc3YJFfsbI(sS%e6gVp*S}oF_fh|@4*T(YjDys4CBTafsnvYQX=4DdV zJnz!e518ANo@K5Z2$2Y@rn>P1qo-T zdNDe46ejzZ6^laD>>v8$Um8i1FeO5?pwleq{k@mdr%5T?t_(UF`Qrk%<~yfsgd_7C zjr=jkKW!DoRn`l)m3ZBam|Kpw7Y|?KHcRZ+T<5g3k#hVc!KDc(GVwJntubNx$5Wg4 z?h;0RPzs>ePlW8hHw`g$ZXR51+1fDJqt-W(mnq`rdICsr-1=KXSahmK^>C|Yzx%>ft0~wGIAYNjY z637qx5$Y6abLnR0>}^B2Mtr?#aj`)NuQ$IR@XdzP_ghbeQL!c)H0(cix3HmZ-`#D{$w`3gISCb!?W~P1FHXzv}B+ z=od;G6C%_St=wtLw7vRfY{}O~BRBV{*R8cH9NgN+h4WLe=!MmhtdlGB2B{z&7d z2#n7L^kt&OBm4r%7@}qlUrD<-Rh^^jIiomoGHf`$+5l~JBUh**{Es=`owp1_08t^c?dgp)=nX>bUCoD(u4 zB?Jh-T+uuDxy5+b%Os`9Ag&;?sHVg~e%5C~ef}d}-7z0Rj>)+}?bk0c7^}ng0$fWC zMK;PHb1mLq8Ed>)fE*T+G>UE>1U;cyPw!7xGr#8;53EP1SzQC$GS)UapOR7y2mK3S z*&?*6oKwNNmu!xsr>@BICFRxgO(9r+&8L%P$mAFNKgOgfmZlhVhttL|U1u8q;|sv~ zH1T;VGy+kyTsuZZxA6qh+yCq3vOcm@e-VMLxKpt`QZFxfY2G*wXjXDlGDff@3%Y_v z2C+|iHZ;QA=A9BQq|1eUm#1*D@<3=ELaU1>Nz?VwD|;dl0GAajNREE0I7@Y_ntA9y zFS)m#&>2UvRd_zLqQ3k4bKxvi(y#)2MbPV6j%dVp@_`OL~xO)A&BI5Ln{ip2=ZGoRV%ux5qq2a`? zf8x#F!PHOK{bWTC#NM(y=}_2Xk)tSoSzs z^o83HRprnu-AzaTQ$8u(?3Mw(eJujNi|M7_W|t!FE%TnIG>c`sW|8DZQTyYh6edo! zL4y2R$m+1%`M^KkFS%+n5Qcg`lbVR{&Rb^i)THrb+9?+~v}WOd5#?UWag#E-eJ=iLY;I)eR~Pdr$w=so4=UCgNV3KsZ@i4KgRl=J+IA zor#=w(pqHjx*bju>O9v9IP>eU83?gxSw_qW{$db1$RRZbeE1%HJw81tvb|Bw9qne} zw@~B4E6;Y5cUiW7;cT;DpjNqz{pi(dprPlmJl=DeWPH>H(#_k{w6ZnJo?Wih(y6Nr zYvyn#N>OQ3Grfvo7ukGsBJ11TJJ`JVSeL}%6aO{wdgVAQc!>qH*Pba>A&!!f zi*>bLJ1Ukl3yu2CFcFv0z7{FMGIaI=Z+;XYfA&}}>-WBIWj+)P73{}7@muM_?fLME z8`tq_!Ox;}I;|SyH}=pveQ%{EIP+BE;4IP)aG^pbuTh%>_gq)%#L^zHne6hfj>-$$ z0+~-;%Ytjtkx(b_M}^>rFndw%f5n0pyJbO-XBlDM2{}JJ2N;n!A7iXzMiU^-VtgE%-GlWD|!*8 zj7*n04!6JZ&Xuk;FG7m?J=anJCiLxsAxARC)q*WR7zsAieyJhF5t zKkyq2rAc}F#e+&xF(niAWM4<}z-X8UuqewVU{Pq}!fO2_^{?xTRa^7y6xp|?S5i3$ zt$XBQd!qArsd|7(dHJ7))J~};sV5ai1gMQw(PGI*`Y)0W$IHE=Ncz4x=Ye06Kr=y5e3H#>J$oKY zO%ylxqu8urPNxK3tE2ow=dYQY(^X{3HC(ljPRDCR%`(;F0j#)U@#1Q71*P$c7%*?9 zxbN}B`e#~dQ~iu*@LJw+iW;_Y3gymc?g?j3nm0Wz$4Hl;LvqO>?W6F{r9Nlz*6C*0;$ z)ouQtwmhGK|6;Ci%{6W)sy(r?b6w%|7Z87!Bp^-p#|7jM+~6A8t1%^w3oYgXj%=xo z`wnoI4r*=Xd}_{fF4KvRTKl7{GzwMpVYYnFVYY8bD=*0&1>J01h;qN60vhj17RG~( zJs`2~t3@x|7$Bzdxe2HU7phB;s^c*#Vb3~ZZ7A0GDBybTo64cu=vI>&P9C2`nO;A3 z`_ZC(1q-Y+*~RhO zApUD_QxC99tofCrk%wHHUSfxvWMOQD%vcVDIj+*rW!AZq*d=vx4jUdXRNm{Kz{pNm(Udqhz(y}dX6|w zF-7jg_v2(f9sg=iPqCoZjj6sFjovi>j#pfvYgpHnH7pjnC z9b{HF#=y7DsKsRbcap@dzvYuWqeT%;joRXtr>`Rc0mCNBsO1lb0pl<89J-t!K-y_k z8nd%JCo;N_$>7PSQvHSmmh@7MC_hZs@zO$VWlez%wPaRiZ;m|H{9 z1zA0~ir^qGh&%Oms|he*;#o2EpQVLt=2DIvJnWtxV$^3Dj^`!y75_XIYv_HQcA<(j z?WNGvumz7=e^<>>XwH0F!{sL~`9l>*H7&knjS|XfhwRiY`!kq4H##VG%24eAr#VZP z;Lm)ae^o!683ViG^XXGSPkKv@-4Po6*p{%oU3hpwbJvGeZ9Fv?$WV87v(@1qh+noJ zUJCbSeffQX)DkmpYC=XN)|a50f`!A*9s%8pBs41fMVO0LAmhqouW#n+_}EU~N==X* z=+<9X16^Q=dt8O9n+PI`RH1Vbt|Q3+d73WZbU4H`7}bdpQFAR(%^d^o=N|1A! z+?}a^;lKp3-jso4RIq!q`_c%Y;M0X%?d#?Li}kFFa0t;E>y;sr=r;V6XjjW+rDbar z92v=2>YY06FI@=!j&+?;UP*Q)-4Gn;5(%ThkC9)xuJn=CCg&B@zj5T@@%?z@s(REN zwWA)9KE?}2nryX(#$7r+$!l)1ybgTo?_pRE%xA3Ql8CQEWM}x}=gAQWp2`;hl{Rb+ zK5+@$#768r${k(L#pAfj_xQ7A3AQMEXeFza? zytyG0Y%#pO{++V7Zwy;dkCtn@_nb3~@KmRBjPudJivDzoqc4A03w1_v&=00>;}`;@ z|4opKbW6h9l^fiz0(f~pv6sm{+l%r4(?#^|OFB=0NK2PSI{5U?>GNHf0tlHYz6W`l z-Bdd8p_8zyjjr_ZSY#T;v-x|fc@;KAo8=#02zL+U%nv`>{6m;n&&0W!`iSo$2~Dh} zHuwjTnJ@j)+b4kYVz*P|I1fQ9ctOMiM|_|j|LA(Gm4#uLP=JWm1K}FFHY^p@96|b7 zB53|h7L7|&u4fL$HPGzv+OO5Y8gx-ON~I*aT6f-koTuY4J_*4K+UbR!APSEWioHCx z+1+t@R_Aq&#!w|^lK}&@+8u07EtXgHr$Zvm4$K$O^mfs|SQKW_Aak+TE!wqogwIVd zSPg0v9t{n*C1#CM*Dj`Murx(p{hFFXu?E~7gsb0A9j`20ls zG)9}$k>Y6tS(6-bLQ4>Moec5pjN2V^-S?-k2R_>_uTjuCU~7J)#yr#(HfUL~(8MG{ zWD+zygo_3c^kNz3?@jW)K839sivvmukQyYgm>L}Q{vDQCb~b!bLUfGmp)MlX(LCzh zIDd(kR!YcVDL#Jb*iRt}sX$Kuku zBh#w+XZxC9n2dUq5m=rUn7NzUTFb}BUtsQFpp_(^`56pOZ7TuDt-JJBXhfSzp+*nO zyNF%#QgQSYGd1qOlQo8Qnu8TbtQG)JcaPQ^*{^~)MKcD0h9I(r`R#>Bf|1i^;^T>M zCyH;jSmR#Zlru=96EIkW3w%N~_%!2(=XmkDJx1Xqbn+BlU)D!C#n(-%|AD^mVMl) zkc!oZ51m8ENQq;HOCuTzg2D3v4#b>NAxCAx%ONwNA2eTM?{{;|ghcSZNCyF6?qsD_ z{8SEA@8RRZI;Q@sr&uzoTLQd8JQU@O_F%CS{6U(*hAD-3_)%6=Q)=v8t{B>qz_pWbtMy z-gD+H6IDJC;oHPC>uoG-Tou4qxD0L zj@3k>r?`X$hP(RoE0kMAg(^m>m&gM|V*qfeX|tI^8w{ir zbwhk6Qb-a4s)CkTu_8E&0U7xu-CuL|R}R#i;qwMno=Boc-m;z8UkldhoFJ z>N(QLU5X*z>d7N&HVgiAp53-KcBFi={3Y9}{8RQXL|8+Po6hr7V~wBm^F(R`+3i1N zNVklF`TW-HiGI*ilI*qSCN2?cuO#r9DM4Fg?GQpiyhlnIGj5-5?3b;#j;&^qLUK?}pt7KD64p+NjnZ!49 zA?i*fFPVR$$@czQKW@LYP*|F;xpHyLhg%Xsz1KZ&!J6?#t)nU$i0`lE=ca>eMeX!7 zDG@E0Ap*??JE4kNFZf1dNzVBgyI2aNKkk*(Z0H+ zJrU*;f#VM{?aXh%jCn=T(yYQ1541XMi=EVlbuA@8rV9&(&n|gA|K`I*X&Zd1F?YlB zyefjZXNpCOCke8uV-g*$7pI~(_dm) z<>gY=+e;XY#VJ*$m3qLJ9^^q_AGJhXcxTKINyZy{T0#AQuLGZy#VYms8Jc%UH1M_^ z=0Tfv4Y!=pVW||&dz8lni1Vahdm=Uh0ERV}_qexgELfZu=n6?wq6Hgp3yO>1@fbB& zN-^%1SglvsTEtF`#i$E>$arQC$af>9Wxtf1$m$gP{4kmeQGR^2&U5l@BA}jq3&Upv zerTbwGll|&A{1+$r9xa~lZH?Bkc35`P_{%ax1eXFqXP%W%yE2v)5SDk`pTkgsV1uH z4?j-WbKn$W!>@9-E!!MvE#`u^H0oSGExkADsjI&aMcG_mC+YiyPQP~ewiyl`{JqJu z8NNCuCVN!38EuUsps>Y>-(%bNJPAr=mmisf82Q4&7&W!FvwTqi^^DS29_XZM?G6=APh>DU zJu8;`7v9vOWWKYj+P&vHIi~IK!Ex)9KcjSbATwcuM*W}Qb5V|T|3y7KXu$4vxJrZ4G$ILZS( zrDkO?2MQ(~Jl4Cqvo#il zUt1qhs2CAikLCO2Q{5}!*TP>Ac!b``dVLfNo!B~R%+layk5{TUh0AshAwNQ+_TD7r zkMg`oAXkNE@T%HATX~9oK83(ey8v(~x)Abl?EV##x+#PWLjCz+YkadeaZfY^XXQXW z=FyPFMf)cRmI^>Bs7zj2XHoq2Wy~BLP0`@wJ6*NRMhvij&uuc(QfIOj$KGjzK=5Q4 z_$q#H-FgPE1cD0MsMZcI$L|MiYq2jObK(zmt50nd0YxSyXqb3AnBjDxrq=>+TkVoP*+c^;XA@P^}X*xPHiQXi@5w%x7M$n(R6E-ouRf+=d3QU-51u z|66pU&l7b!7evdy-s6GLUZzXnD0n(vOG=4CQ=|LB3C_XVL zX(GuaZQ|qeHs=@f+xy+?qphQ)2V1(CV7}*n>tB(Ir1-9lK^q@n z*n9`LB-15I5ZND+_KaJMa1O7Fz*YKD=aPI<+DeA&{9a@mN# zcxCjIa$huLs>QlC$Aiw99H zJEx{%s|jg6if#47iLm!8NZUr1@pKIBA5NM7qT)sN3&71X#I>K?;i=tw@B$cIz4w&; z`#rR?$Zxwif-G2nM0G?MSQK+wQZW8MEdcUdLDx9eRPzFyxE!0YV|`6yRQrqs`+B~# zE8s)z1HSFv7mmDIbo*p9Ao663ChEVL$^V_(0y7Y$@rwGqx*N%D-HJ$V2{EPxlD*VG zIWdSDC?(ArPM&_6onJ?>7m9hDBepKeb~Kgg9i8LP_a!d~WlNGL`|WCekU6$KAYXFt zz9EVMLr1z}1CPg3h(s-!gF*6_Vvkft&)F*NGl)kbZ7#~J$?i%a{mak!TLddx?{)*~ z;ROE!R8s|Hu;@;m7LER=ll})qt-n5Y|F#tLN`z$dE(C<~2pIH8jr%F_@4qn_TE+m{ zKZM-Buz=K>@&nY`e;EBIw&cHlN%rj)+5HgtJ^wCBg>oy*^Z+LFlk(r6u9YyblwwLn z=y$jt_fUcX8H1;v?*GHs|7!Lgx2b?X^0FOwMXtUj-#?~Qw+1YmJ$E<8baUt&JGJK-H_ zV8S~E>JiWX-Ie&|0Q^L`4x==Zcl9|J^nk5)j!&C2B%_|I*Udm)c?1^?e(VGDq{m}@b{BYBs(xIh7H2@Mpl=KqyB{KFsr?!W!cwvWgl zcg*(t?_1WD_Be(O>OXj)yWi=*ef9@_B!85D^f3$HfWa5PMw&jIKYBFsOYq^-9IZy& zsKQ-r<4IDkqwxwMNn;|>hc)d`<^5Sy$&K21+d$68C%#$UnB}GB`I8Zr?*c*!WmO!H zgJV^Gn=W={=!R z5Jw3X4j`YO;P(0$KJAlAL5Ymzs~lz$kUaXGSGyr+L68Hh3EYl$^wjM>&R9Y1 z>%JTC-bbiGh^)EE1T|_eD3OXjP4Eq8VuL|*FDHAM23xVKzR&B+U>lz4aAW3dyulFA zbG|%jHZ+Wh?!aAV2#x;XakkHyXv)#fnUZVxY)#{IlBV}nogj)ZWr${g^XH31VkyeU zZ{YDB&aPKtE{xY*q9d8_4#u!+&PJvv(O7XX1Rj3-Fh6KqCNK?Pc9Uv7dP2`b(9y2{ z3dCY3^_7ynUiM-&Uqxf4KYuX#K^rH)Q|rQ}5fiZyudwmsGkPw0{dkxa6?n?bL;9KA3d&9Njkx!wc~l}fr$ zTnwGZj^oIxcqyQkNNxHLss|ahq03ZY=Eg1;b~ln<%v)U8r?v9PDVX6W4i=L`4amce zc`m3v`#%<s_0Z4~$Cvt9@2&I$b?==K;CwH_DOB?P*I8H4I|(s6KcCF_}LzV-1gD^1mc5~CPz z{3aJ*Ch7H?mC3IHuXKhz|hHp0J7vem$@cY#5@8?Vivg^_)YVw!AbtgCoY z6`GRiGFD#4{k_AezjzZQXKQPv8(QvH$*@H>GmGh3Dytc9lnJiP}IVY;AzuT2D8 z%Vt>SueUcVMrz`CvoGQU-56`%Muc&NxTJshp^sHl7P12M#5aj260F6*9>#E)FZ!y< zBuKkDt>!J)RmqDV!ws2{Ci z##dQY6+_}|hVqPv!Q05Po;~Q1*moGGS}9#JiHmEcd~jBwRH|uTKxI@8dx3OvZ8HDW zoUSxid=Q;_Y#^O0UBn?xWqhVtRxiT{7D)D4dzJ$7;`{Jp@`&6T{jclMgVdB2_^AAD zpRmw3z&x<_TnVLqOdw*#P-%JkKnOS}f19AzR6PLx%_UrE<3M?_8a0&B|K-P@dYO>S z1CD7ols_CfX9BoLH3t>i_j-TsJFnM%6eTa@=J57L7cagfYs;J!iBMM()l{HFYE7tx zJp_U(36WI%f9=ogB=%D4 z_*|;{^`y30{M9y^q*;P?slZG3nd_z(7lwx9BBCQ%dvD6JisaOFdch}b<<5#EhV>-J zd^}M?AI(lJ=M;tA_h^6Z4_wI9P& zPMchRS!_R$WMqnn<11lgL{j>=&3^d}{@Imn8~@l&XWeLPfTBs~s>FI{?7>v~DyxrF zpRYc9Mb8LxBZ5mkMvE7*VekCBb8qx!b@;;EYlN3MuqN@P2%n7KXCIa7E8A(zn#)VB8-w}) zI4v`K1o!)pUvI1Uc%(o9kEFV*(64GseNTw#Q=@@NokHE{m_>eUDO9sED02*AK z&lFNN!3XQ<(&&z`~#k*NY%7VEFPrqRE3Z`#RG5&Jc|T7|46}z=cGu&)6x-L(v!^A3XuV_8?tX-WxbNTNe};3K>V z!KQR(X`L={lAszLS4l~}_Ys?9DTh>y~{CYk8dytP(!Ouf|)o5&*`n&3V@{w|%- z4*Uz&^-C=h!wd@r$jnxctCs98qSu%^@V94>DF@zw+Q~_!b%?$+cY`tt5G8aK_AATh z@pnuYcPOVs;3ET~!&lDaJTtm{8hD)}jQV7KB%*fWMFuvvA5nkt48#a<>&~1X4wie< zPbVX5F3#uF6Zq9KywSX@Y;%rRzfK|YV-Mh>@v}&!fTMB=z=fF zrah){EnfR|nW4=UpdB49(o<3`8WxbmB4qYJqmQ3=ZS$Oc-Q4{%9iwLS$3`IR($V0J zRRp&nGM~>SRaE}eb()E&<9xcC9viFJW$U3q`8xLhu=dt*acte1aGVgF1PJcIArRap zxI2UZ!6is=Y22L% zFm#PqzEdFCw|@WkJdE*@;9im2z27~=&$=U9>Gg)zYwy6PpKd_xMYBr3bzG#wDH6XW z3Pf~%t>vey=dgh{bq(^*o?HsOz{i5i1hbCSP_i(xLlmt^p<7^8_P^at`6P`xb=j)M(-uwg#Fo zbv@sG`s<5q#Bz~lo8>>{uO2=5+i)lK#RH)AE&WQJ2RHVR5-7PEm)P&_Bub6gpPkO+ z)!#Gh>EirialP0ZO+z2XK=PC`Xc=Y+4BodxqhR}4%J6;Luj~V%)`nlbF1Yc18lht+ zStsL64&^+a?+#xZFq>4m=%j zii`{-?c#X2$KSu;u#R0M7ntf5`&mn2rqiwKqOR*U8AcoKR==~RKcq!QtgcWPBq-j- zlh!RY74#wZ7Jq?s1#vS$vz*d4pOUF}ID%6Gwu-cqP>6gEL{oS@t|Am2lW*=v8YH&XQTHQ%213E6=7>f}drL`xn z>+Z^VNo(XmL&m@-!DYfb*Mt6iXSPurZb<)wwGsCoSXm&yoVp8w6ibLw-W8}!^ZDC& zMcxH*ZcLc*P)CTpTRx*GY9SGzoLpU z`TDH0YFnrowsHY~g;T8Q-FxA@)@!I_b@q_xO*5(;y57XoIcO-lEShb}5JaHs2tV{^ z^10B}vBp1cBJffMNdG)HZX{9v6Iwlfz;KW9UJ@Ixrg(CUgq94v*Wa_fqOm_D^v%=Z z3&of#o(Gxkwc&)h=k0$QJxW8ne!spGY~GbvZ-xbWk7=+@X;DZYV6u^lzZoncg4}eb z5U(JGqaU0$JKZTd99)F2fC}2hc&g6q`nq+W>WXSt@zZ1s(Co3iCcS)Obe1{w9`*=R zukgOTBRh-!LgUh>QBYcxj7Zb9Y`HXvE@1l*&9caGnjZLwCD`5-G7)4ujX&hWHm%m$ zG)Zh~=1i~9nYV62hl+3XV{vvG=_iYYqbPOO%6C>KPOgv>&Jv_cx;0ui{&d1hQ>EH1 z#&Ev($VvLl3Qa}oPv$``uN~_8ZLj8T&z9lXfF{+6c1wVFiz&d0rIGq@X7>)9aH3ha zj3q>N8xGxL6ibgm%=%dxm5up*yd^_|+)bB|jM^$CXzX){kg zIkS(JKU%_>`b(}#yihk|M*I?$kq@zH`I$W$_3izma-8|T&a#HFA2P6q6|ZyF>R|KF zLiDGV{a@7q54vk(OTO9*UwJIsWW}^hHqe?mn@}JUmQ@gz@a?*$vAQvCIJ#=nlBbkL z5dHR>D)qVrszvlu`4zQANgQOaQRR!AFN3K5|; zTeNiSJKr9?M)X%Kc7AtvWo*cdYmefS7r4+md($op2KT|bB5`D!yf!MbH0zG6lQ>3b zkbN3BOvuD3vwNQv!Jp9sjjBK@<+zX}QEpyr16_ z2$~_R8K!6;8kK#q^;vo_CTVz>D2r1RiNS_uGM8CM$l@w=s@}^OHF0#)tkT_rG6&TEE?>vcj>|O?U*pY4z8Xk;;8w0y}MV}c&exCNT zY8YBMw!MXsvy(a3W-DLL5$*}eA=XhwxELr|e)Q*}f3f?)Hq^oD9T{@G@=eut(6Xx7 z`h1i`^C)Y3@0^#UnOK5SEHVG|%ucd+3LOt-4P6?zCw6QTmk=M%4(XD_>)AFH_UY%j zs2Z{4&4ZEXvaZ?V!Dz=NGmttj#2JKE^d-`JL=8K<|IFO}as1Nl!E<>Kk_P_i zxjY_{{lv@a1|z4~^w}-wr9TsCV&pD_K!0f%I+M`s(PvQ^|jF3J}#KLr=t2H_)(@*JDtY!mwK+9sO z;ssFPgCLHTnO~0Qx9;f~K#xzeK%Yt%i(-LeDp8OVzeh>~h*S8E2XqW>Bcw664ti;=<$H2JBfo|k zkt>S7*wL?<-repO<7z|K33H2?GNHm}T72)4IKDO2H`5ig9m&#DB14^(l8&z(A|AJ( zt$#kxuRV#9*1mKyj__&9ITXWY3P`9a#YQNy2V~^S{D^1v{jojrf$oI!y0eva-%eVq zE~+t+w?#)z26en)XF7KeumOU${qE2e65*D>|&vsNYCkpxkOjRw#R zDg}}JEF`Jg!I~7wd}R-LU3Bl;HS?jTs!Lxkk}wT(>Ybs6mWUf^AKHcnZ(U4T_Q$ZK zMWYfWSB;~u?CtEY(u(N?m4&G*b>W=3p>pKmH84BVKGd8>1pJMMv* zGzM`eRZ;D$drO;+Lw9?afx`0I)a(5kYxxQx^&fY*+$u=lZFT8z81oNzWaqL}^?EE* zb!dbZ=gF_C!9M(8h#Ln%!5~UQ+yMCP*wk1k&zN~C% z@dGn5nNH*7SaSQx`S^3sy_fQbp=V#GlaH-=LEmo^It38PdMszhU{9ZSmal(X{+Huu zeZiUq#Lm;7KlCKdNns=KyUNgCx7raRqt!@JQ?5>wcN2#3YTxy#t@fTM0PGara+#w;akg ziSxufSE1b^e&wB$`<@SdSE})K>cMyAu780{Y8qcrTFy^^&7;T-!gl zZ&UDnC>IM&u_cY|EKUq-9eS9?nhOgPpa46+hD4U2t4|TWA;A~(tQowjyz4i;hntuWtOT+K2^AW<4iub1}PGagXR}U+HI~}}yLBQ_lgbhq*yK0UH0>dmpKJta( z>R&X3maX{>!Mo)U4bg$&@KKX;MUZ&Epax9$V$8Vpt6v6FYOe>Dy`JCnBXvS@>G_~} z`X@hbR~U9eW3YXHe*s3Nvtt`t5mo0mEnl6bb`#^IE)0YxpW0C(W>E$PhG33tkrw(PlRQfFHGmw&5Z^*r%uG7C@!uJ!<3@l$i zC(O41^Xty7_w=f;s@by>4|jUkqmYkybE6aW33yC%Y*LK$j5$b3u2{2K*itDdm0!Qyv6(G|0A2+Avfc0|mP7|7 z#iT5XYwL-CnZQUypAgiTHgPk4@bs*j@jJ%y=o47O+Z-1i*3Row=$@uUKzl`8t!%!5 zsweI7?Twp%m&W;eL@~Uufn}$EA)6u&4vNn1iDry*17fJNCrG*NUo(TaKGM@HRP_C8 z0^pGmfbRr%8uM%OxfXgn=p1cjWcz$obmlA(n#>QK8XKY~DsrZGj{35C-bpvQYdabv zhZ7m@rb!4)G#mg+Gmt9&aJLO9>DwPj;yAUI-w*;IA3rVWX-q#zvkf-qNPaa*tY6ZQ z7Zp&W;9e*)ubg*XI6$B7wu{jlcd@ivFWJDoP2@d(`$-ip=IUavH-Q|X<(EK9od*7u zNJ&4WL|pj&)>Vsrkt2^^qCs)MgfyRsrJ{H1WH{oQQE+myfyLLi(_Rp_v&ZJs4Vr9g zzc~~`szZA<1u4>K$p!veaSzx;kQWaQ6D5*oSq%pt!ysq`F*3MKy-lKfl#=kSY81{PIHZp8p6%+7oq?iu&zI`F%rj#h^4Uye~5YlTp4jDoJ_I-5F{wra5! zI;podVu2E_`;WDj!#(EKt)t%xy<&3fw;b|BKPw_yuaWT^8Jry?-O zh<9Ty=PNc+@QDM@Hx<;=DjwZA$xBhAjea@LlrFJ*a)fvIke3Tu7ROSXXJ1Am$~f36 zUK8evfc3r_5oI-)A^eB{!uxgkpZZ_wjg^V^c~qnjrk*NnGh3{rkrkwd&6I9!C}S^j zB8@YEF(YXq6TvzAf<5_>qCJnaOj3D&1f}Y)BVp@BH|b7Pmo~!$cp~1G8imrH4UYW!>_&IPOt}pa04afiny9K^`8qT}5 z*?1xP;gA%VN~Cca$Lr(rv``MUvPtgK_XQv!X3)yfxJD*YQa+~C_WoHm z(&}~^9bJPnvhm>i#3_-mf`wOfIfm1_X0xZ+jiFS{Mcbls?MzIkaW}(d%$hh-x4lGb zRy-R*#oD%<$qSH1CPKPQa{(^#o+2V?BIvvBG!xm|gghzgb=g_v8wW+9#R`41wQww* z=rQK2llvO$`^6g>gJ|b4sUXxcx=kogsA=ng@M@g=G`So;EQnkOub$udc+0@5m)4_w zfgXIqDvMP@AF8?KeA~9K7L_;yze(=T15&|b@`+iPQd}p_R0t6XwXxVm{cOBnxptuz zWcJH}R0dJit0{)P^Nvdeuwu|RJ4e|Xul_6%aF9fYPvG=Z#lca1+FLK5kg?HgGN7Pg za#NqoSAN?HnWmr~l%-{gVc@ew0F2p1dyMa`=kEm~p0XUSTYeRnIwj#eyNJ;A*h5P_ zL;3l{&92|k3=8L#1ZQg&=X-TRPis=Z->9p@JoNRl{csMf za!rD}nzCiRIH@gQlejE7F^7;> zQ}z#Q*ufJ&1WBo_cEXw6snp-D3BNhT7W_LsYB--MK15wO16>b+R)KG zuq)9w^yaL$3H5kjPIdUB)t%a&iWp3j?Mt@Kv5mi9u-kPRS-N$NgYB6(`zyyR;grJV?R%R3BSD($B*dyQ7XU5#luA~seg>J} zpVeM(w;m=VO^@3YGx8CJHsIc?EC9=?E03y;N|pV@!a9M|#%)ZN;yAzfYUmA5>kV>hDG+r70&Vc>(N)py(ly=MFpXgmoNSdrs0ZylO3(R@|xX2I%{h z0J{g;=&?G<@VKkT>^{_;Zc8-%Xo}r5b%Mo{aDQ#=K=QfLA7(zw&Jrlr1x6nN${a#9wI;RSu}> zO+oM_glTlEpYlkcvOgqfBHj?%WvzaJx+wR)TONz_Hz!NiXof7$Ax=6^-a1m;jL z3;(s0LaO+>_g-5creS$%@e)ZaPFAc1bzmSkxu@OadNV#vZmDL?(`w7AOf}zHZDaCNX zme}C|wi*qj9rUkzq#A7dL%&snUZ1H3?3Sw#IWMHW$hUxw$nYsqR%R>UOS6{`FDXTcs+f)rvN{p~_$9Fo zziL`~N2@kvLFy}Sfl(gZ1(KT4=AgO@A)GZGcPr#dtDUK*X(0t4KsCH&s!z{H>&8mr z|HBqj`2)X+X6E@lv%UKcfF6zZbyFH0P(oV(9%T6iU-dj^+qB!>F)j!hjb(!?`N143 zI4bZ)6+Rogr@$vwb&cK=icY5oS7_O5xdw=B48BSNa4YEz$&f7qwavWZ)!JUto4Rs>)D_L>Xrai(ad2amM- z^?<-0(ENriAh)uQ+ip+pxs|pz#j4DE3>jka!Rz!Z3;e>;AVGWRK4{bUd5CK!1jrpS z3Va9Df)yp>MOKg{oMIkhC;9+;jr4!%hM zvHO3rX`ekXJ)fjvy2bggtmxNyc6}r77@$fPL|gO>)ZXK9+)FpfX)v^aHpny8)#(z~ z8{9N5CtH6o1NOiOx9B673E-~EoqME?8_>uRwvy|ieHyt%>uu!7a zUNJE=0II_Hq6)K5mwYlZ`aA(sD){JQgJkd{*{~|gV+NT}!@zCTw zUW?%2&3UFn)&WH0H1HciL@aw>Y5cdS9u@{7z4-4mdCmD;9Wdv=2{p_s>Ra4z2=|_! zp2w+RYiT^=YNmL=yrsTxe&d4`GFj5V9Q%SH2a>a+vyq>BM}ZQRlRGV`jOwkP1E7$k z+6ci5VKR!8m#^rgLX6-Qcw=|(a;FfAz^#jzbrmU}p-rC&bO+#>R7`dxycAiOlP$Qj zWAuyRrS?74p>M(WwDaFY_C(C;an}E~9Gw`>EaUfF`OreH$5flcrv%YsJgQ@SwQ6`l zq0M>|lXx3Fl#Xu8+Jv++&Llht$*Tf_LW2IqUn1`1`ACL+ z`p?7>W!5iOlMGHfv6Oj~K5DF=vSI}?me*XSJG-W)G;%N;qtfWu^a<0*kE)pUmit{J zZ%3}A?{GMW*ugTm>= zEVKINMrcLbZ5y6(ZRXx6dbby#T1k%S)ElYk;1qIs3nLh4T<^mIGXFmR}%7Jkw)g?|DnpJjap8+)0?G!VWrZ zT#Msa4!eB$|U1k=Bp_dXA;RYHogT@?R0>Ak$54}@%{lksRP z5243=EN8gnC(&}ekO1CqIoiTzvra>8n>x%@x?_!W^Ukf6j-3~FzHj~DYfR3erD6?e z4c8jGv=tei$+yvZmU))8HT_&YJmD6FGSo(PnD+`|h2`$BQOi6m(B*psd4Huur_!k74WPX~q}GtF5mgbhP|Lh8vOdcp!gV1M)o12~ z;#j|S0b7tai3}k2us*KK(XDvpcfc?NOd>YNUpMOSB)KWCsP z<~lr{w`|1RTIi)iu6L&dW0rgbO1}8p+;|2SEcSbW5NnL*G~Po%%~UZ`nYO`$lbd6x zn(wy`gD5BI{udggs#X~vOG>L~Tfi2QuS2)%f@i~4_?$(_P!6`I*$R>79rUanZ0yuY zKKzvy?oC=1!9H-x6|O1S)C%?@IA13dtc}~Odrtq3BlJy!q;ghCgIUkvLo)f?8g&Fn zvt~dx!z@g&6be?2fYkp0Z)ScK88O<0hZMrCO5`LTZfFDHJw?S|OU85ag?##iBeQ_x z3YOypZk-bO>5amOvEvwqNuYGzSfbG|?Z{Jx!RM)d+Jjm>tv>A@m9v_%a9FR*C_n2B z^;7y+u?LPd9L#XJ0JFIU&BAerGbcAq@EY7n!9Rjjh}ZUfk)^zh4tymvBK2@g2`^+UWP_bW7B_N0(kp)~aLU>!Fr0=A%1tX?;+L#>4SeSPv}EvveS zfkW4b!w<`%VRhcQ&cv95l&a4!$)Il!CEvOmuxLA~6`q3$V6Vna*05Whb=m94!LAMq z9tpL5`VDuuNLin*6{xeq#@cztR)Pts^Butitbnpklq7Q(9L1h#ZO_ zQ|c@f_4G3a3(e|CO&9tn*0M$nYa`Ddt|&X=3TLa@N|Pt^PNM#49!8rkYKt5Fz;J zg1R%p|3k;K)K=b3EM4fRQ(5E~2ilb?ucpCU$%(nXqmKow4;!T>ae07YwY1Gv17*;9m|v4Ks!>GW?Lm6Jt41(=h8qOY<5d>gUUE6b*L@(mOvs zohQk+?{<%9dxsrVin+Y8`&nF@6xmRg=rR4>n&6Hyvc2C~MokFkB2o0!LMF}e?rD7! zFELT~slL06tn9|+RCLjPTNDu5*u=HN>YVVzTbQ^yqnElhD7KwH+;w062Y%A>`Gv(M z;JDG>Z=1P(SfhDudY3@3n+=r0uUswKZwBgG{juC3TL%|ffE_GA>sJ+}$_tThI_JPL zu@F50Vr;a^6FVoaw7aPW5732f#`j0I}4;Wq)C_&cGGm+|JXI9ky-Msvk(DFZQTi{gUC#6hBAsy z#K}^)P2FAl4bDaA`Q~x#t~MfEc2N{+&F~(T@S6t#N?j>)aCO4A#YgB(3Gdf88=Q;2(HD~cR>9kF2ki#gbQ$}8J`ovSS1_(ajlcek22P4VzB zBk+n+X>in_FwedAo?wc$SFm+4) zs&R})w2;v7rupC()OEWfAI*EjhA}bqg$p|>0puLq8C~T>QV%X(g3$^!@XM?^<}^%f zzZfZBjJZ>Q%aGnKq2N%wT^acPJtH6xE0}7^XP}u-#Ad!AB8VB%P)x%XjKX%+hy(3Q zYQX(UKP6v58~QrmSP9}oJVOil&R+f<<C9T7gx&*I*+oLXo!37MCR!OV2fvqs#|vzr~V$J-o-%e%Yj${Ck+ z`JK0Jd^zbxM&F88Y)$smHlMAO2l>}DYJ8^C;iU|#X`)H?v?ZSb<$@u`XtpE#DnR63 zE}J4AG$rEK-*cMXmvOgXshat#ob6QIYS<%%rD4T3Z{>XM7 zFj|V7XOra~m+{PJ7nQO_`yVfgt~et3HEe_Qr$lC?;lIAOhLuKjt|X>Z2VIC3@_F-p zmrSX(l73u$-l+B%N*?8tW7czH`#Y1B+-f_=B+^29xwFAy@b} zea=1B^8(1jqo=xiIS*Nw2!1RgXFf9o*OXHrghF19Igo{oE?efwj(!vQWsd=Jk3M<~ z^4e*Gq=HkW;X|xft?p<0)?8!z&tL=Xr(}rH0`ua!#XY3jD!CESw`6acB09BjeLsHZ zw(lU?J*T_|pvI?T_2TiT=|L+xOE=x(teGh_tL=mx8uZ&-JHCcIPySdj|A>+PXRzcY zxb(1t-O=t#u0J!7N8~^Zb@LLM@MEC2Hz!9doTEW1k0L#tJnud9I|>zpZvM~xN!{5< zS8X?MZVHm0ivSJT#P}0(7S%$6Fiy%RQq(+J-=3gPgj5|S`g1fw$SUpcwR|+Gv;5>b zA4HPg7*sh^2gUta4;QA-Q(50cR8BMTOU* zTcg%8c>9Bl%pbArrnlzDNnR9Imj%8*H*&abc;Jov)UmLzr2u2L-Zuu)U!R{66wy7E zJmsH{n`$`h57ZjY+s8}b0NsF&54l)Ro=PXGN4?yYVNX)gYw4$92e!Gy?~2`aDkCj? z<%%C`xzhY~Biho!@5LRal*?}Izh}%^#1#6G0lO!{S5EoX7dJENm>TCi0s}{t zhKO7xw&+XWlege7|Bi91A0gg>xU0C8gPqlyh9no?us*<0oC=Bcoia1_$;P*gp_`nD zt|?0Z52cOHA@y1FiIz{MmV$;c)<9QPPlw?2q1^jw9*j&W1z|!>CyK&zd-@+2eR>n+ zB4<4fPdED6#r8LnUFUn6uV822S&GVc$J+?J!8UR*+ zo*WmeG$esB<^{%ncb~Q#xXa0``YB1Uv=7z3;J~RUDAt)GC@?cL(&Z6fw1Df;&TW-t zr!9EB#AA^}EB}Pr2%D(8QJ5lf@M>;k`qrneL~&bMyT(>G6CQQ&(qEOEYzG&np=)wKwOELk22Vp~%$fz2Fx*A#eg zY9~Qk&=LRT4*c?XLwjOTzM(Nt?Iu~A;dQI&K$qjh{(u_l26)6MgI-6|)hRgJ#y63R zXUwXeW#|n?i)G`A*~uO<{!?KL3|T0Y5}=ZXWKxjbN^dw~B(fvpfB&fNR0-Bh-t413 z{b`2%lotanI7q^ucw%!TdM{kqhg>L{dtjO+nk!h`e~|;{?Ta%doE7a&fEZ)81h(uj zQ=Z}xiBwLvOy6o$BWNrA)|>!jCfk}sDA`}!Gy!5fGy0`G zKN_Wjmih<=zLcP}B2dh(D%7@4NAW7`i!KpDdf}t&;kgkjy|*yBWB>) zQ}euFLUPEqanpifX_iy~kY^i`hAnGqeZG>W(a3!7HCiTtJS!Sxe1M<=^_YKPpHvIz zF3O+Dv!|;BY`1t0x4}B9{GmppacH9`@>q-V+g~h{Z_CK5Msy>RW8@0%DuSp~bc|>^ zb$>7T9~hz$<Q8(c+xR8z`kp0F_!R2KCn^Un#2>SSLM2krJv4+JQHiuf6` z=H>b?L+#@J?btGoqNu0Q`y1Ho2rd_`^JNl0j*V|le-TR5EmS!Li}WY?SW%oy8>jIgl87EDQh)K`~6 z07`utjf0iJb=ecAS?{}HuiA{M3@c&kZMbnoEm7MC$eBq`##kUPeCM|v6Xn#t%coz< zLh>gJL^W;K`W&!`3*JawOkiHg=@$0Ai*JQ?5KXCQTijCFbWb&2rh((BJ^rEK zFL(qrq0>6=yyE(^&HCu+!+QOw@kKN}J$?G^HJXB$@jFqDko0%K!C-Xa&LZ$f;B~CkU<3Q$&NlmbB{UC{lg;eA&*3z7)LN^)7!j5 z9t|19va~UF8lM$B)dNczWRlT~qlhu?Tk3T*WSwYNv=M%)@Xecq`PalAomVK-@ z&hlc%2;+F8gHPN!saTvaQT&-b&tJ`V^DN@Hu9iw!Ps|&`nyVO3L&)33i+rdl4+?td zz+mN%PG;93nuf|!k|Q3@M5^O&u`S#_N+jFVfIH2PM_j?13Cm+i{wKwsb4CVx&UkL( zaM&nWoz{Y-Hdka5TzOTC=+ed<=u7c>gy}F!Z{r<-4(OfXV+MSMR=llHz@#%}79u=V zZEUI*arU*J$u^2xG5;89=NK5VcdGw9xY8}rCUhy@Lmhk={)w{AQF~&{Jt~7mvZ>lu zx7^MXqy6+IT~+^M%!1SHI4Z$9wA?bw=qLAdyOH$9UB+vqQaAUKZEE*gh-i{d$98CG z#k*n?@yp-re=kbS;L0Ari(2mbX|2qwRaR8SEi^Z4yPBNBtd> z^93wnC@cBDj%8Q@bjMmM?lTyY0g}-eSWx=1wfW(f5EjOakb)`>3(^xI`6YyyC?%^w9r-_1SN@L&#ScmjSW=vSR`^l}9=c-()ZBk#eQvVV z?pv_Iobr{UWA7>o7ix06LyJ`wrh}=)plkDlleRx{?!3dyNLs-Vb&-G1UUOtZ4(bNj z=P%9$1R@x>2rQkezL)fRHgYGv^GMKZ`!*fV%<8HTf>gHr{R^d+BYGiQpqsDYJ_(6| zrm-5X`_-zR!A+FXL;s=}s!TEAb6&L&c@It*PYZ_BuhY>(dHZBEVSGfvrJx7$dwY+? zYOalMbe>b})F;)01nR!rco0pt-b?Lqwg@F)3IB zDg^Xuri)JHMD^+@S7+nk^R!h!=oY&744Se3_T~TkN$)xdpe$Oau}l5okC?)%7hvhx z;$oo!(L*lj6KF*<%ze~JQ#tfY#y%N#d>;CEDt|xjaMi@Ceig3WD@;HFsrnI^_}Pv0 z(jFtJ^-1kC4rIFM za`G6Eog><;?W^q+&y)4g)p=kUMo*#4uWk*0O}OF{==$4W?SDNf_%BtbBsAEc+Bai1 z00Vpc_=VSsub|iKE(Yq1y&Kz*NK|i5+TXkR+!cY23d!~u91pJ%J*Hx{qp(4VmYP>e zc2$HXtdWcCB0%tggaOtPGitC%Ko>jrp5XX`f%L~)|6lLpf1DZ9d-&<)9`WHnd!h3m z1Tn_LzCHVIeaC+Wr~l(yfiuq@^pc*WVSh*XV_JH9J&3RTRzqR=CzXi5E%g6&k{y`A z<;~v2|8~h2X+TRyi|FF{Kd4Xq+lT&~VUM18O9A#GScJQV^hX_&gyexnfIjXe^S4j= z-#RH6^T3S7#6bO%il6&ni21#hk>&pYApRKB|MF|kUIRnylUMa!{m&uBqXLFFoLt=J z|L~~)@UJBd;EQ=%0;2y6rgmNdCwHV|Gm`${)BirIe_q`DApP;BL}=~bq(43Z_O4pw zjKTH)KG`n9fS8V7Bsl2L8#aH)iI^J{rOW(z5C2!kfFA**cgoZh-~BPdk5d5opB@ux zj6Zi(_}{;!|MJrQ`cs7txVL;PCbBgl0J}-@AIJY`bT3sNM%cWcMEAe{85nr`;)b#(rAZ4qab^S_<1RL1+%$9J%1;k!fSRhS-Q(DG>fv>h zPNi#O4F2xbjd{_tVYI-#l^sXnH`;EW!scOuvkIRa!5(*1kyS=-lpS^AqZ;4YqycO) zMDL&DUIT-LMuF5H>cMsPL77%mY_GN&$n2Vh9vLW#@?pJ+8Pyu`wA*sBBB+!_T`0>< zS839|HXqlcrGNJdjk0c|A*8mq=l;ZW^A3}|j9*)yx>zyD`t^i99Sxrpv8`@bHHQ6i z-tbrz&5mU2>iN#07Lw^`F|o^T2Hv3s7iuWmEL2i|q)3+M3P(BnQ96m-1z_)iD7=g6+^0 zG<>gtB@-n;xuFt>q zW8_A_NdZN^4!gCX?<8?v&lR=v3VV`nc6H0cQdIdW;GKMzgMf4eM?nP*C4U%E6Ro7t zY}Wde=JmzAG8aly4Ri8PzPI$|*qD&_kxpTD&nLodz*iEgU{L*GaDqjofDIMpN!9z@ z)STdaI+};uQ=g*8#KbCVXi!&LjwZOm_k?ni+uj>xZ9~9HfiXX4xBeO5n{v z%~DOKGj6hoX|u!^ri_w6pu=Z7^OO^KgQWl<1vt#qT#R8GK=<1GL3apJo!*oW&P)#AjH^WcLo{*U0e(4Y-ChP~*Dn z)sPPBsWKT^G_d@id|;v;yNQ>}mvHLbbndyB_L zofVQ%hW{-8-O_=EibGg4sO_#$X@{p{#8ZHW)1OhV8O?{myExKvt+w81b|;ow(`kXj z9Vhs=Sc{g(C&f*6l%i|5MzVk(jp7!}>4YvK$~z#r&oPn&U%br}j}=2A3gyczr~|*q zr-oUZ9&BF61AP%w`$+;-eE4$Qn;VkJQb^UOU z^dnnkw|u(*PI{_lH0-o{vbC5jF7fdI^W&PwLa?Ge`QZFh*)z;OtXfk`_V9#wt2fJu zWqD(M0**`c7pst)dC~Z@jTT>`c{B*0VTK|Y2(K(Hz9+?$ysJiN0hJ}tmz0e9DD+Hz z#324RQ+W{BE6SJqs;A>|OL(<==vf5qr}MR{31HmwfVkKR-cRVhk!$cLjzjQv!jbpN z$8oeRp8itKj+j+U5N+HbQ)Fcz);`W)7Lnr>r&PqOHM)~Y{`&p{1L^#rlt@!?h6KO76!zX}0itZP6s?cJ z>^v8p{>k{mYqN=o90+~ZPAUCk8}MgMg>S4Ok|NYbTCX3xDnLP>sjD8MzGE?`&C+<`+GF#khR% z00E&c4O+8{%O{csoYV4`9(;ttG|(0Cl|L76?*|AYSV9Uvtv22#KHK2=5dPZ0z+kJY zc~aSSlwzh5gx1J&`kGtm3sAc-pUN1^S3PQAJs5<1b1dt8b9s{PMp>z|Z*yi)WXQ}I z6WERUANhJscy|bsKR`0X8paVIW?zBchfsu=|4s=jOgkD~bUC}21%-{dZKHH4!eZ~^ zrN6RoDjVG8ukSY&V26%v;H08Uaha3GSl63Q8L#$!ZYBXMp`j0QF^1?gaX4SWdcOV^ z00G@FXz-b&{CWdLOXKDJ(B_#wu^8F&&68POqAWn*=VY?|Lm0ibiyW_m1x8x=radn5 z3kgBTE@~teo3S-6QbvQ zA~Tb6nCsS^TDOvqWcLjtP@8MuMDC4n-j2B>+iUC6`3&b#nE4g%8$Tts(SQOvEbGO< zxzirGEAsuIC_a}1E`a#JU27@Q)5DNenTWI%jTN}<_Y=rZl-~iWc9gP{t4elKT&=o? z?VkrNmk1HlRF=jO{}O=ykI95C2t;nLQhSaR)dC{pKO(?P@XBNpSrb+?=(B{4H-zxm z&C`!7v?eRICK8L)>z}yR8Y}bjU0hj_N6H@~)c|5)Xj?a~3ANjH@X^3UEuF2w4VLt3aO90gcN3YPO0y8>M+l)FA<6*G7~}Uj_bct`LfxeR7a{=OYy%-B)B7T z82g$VF$y@d1B&j*&@JH93%?o@JYHnMD}#9#QwC2x{FX>l2U8$~Ny4%Ad8M;?`ESU+ zW2S-qRPIWfft7pO{Q~wyj^?(5IBb8OvMPV8bWs!l{IxH8Is>2Yf|!X7jySphH`31L zx(tgh_p4%Qa-3Pk7 z(%x7JgX&|Q$R24n(nMBhM?TbK(nzbA;OvJq=}@OP3eKkPcT(A_2%^`hKZn!{%1x;} z`96He3B$zmf2%9E=crG9RXr7JCOKmrEWTW+%;U3S^txnGq>XIXB z2MI)zVc{o~R4P`5B)TViTy7s?juH3{yqP8*HY z%08_twPSV2T?@VcT8%x#&zQpX*;b6cqRc2e;xc(~y5(jiUc6+~K*FQsjwxlrnl5uA zSZ{3Kk9x;O>KAuI&w9#i#LzsG7SRq$Z9XE(UewYV!7=Z7Z-CoOT}fQTs%mMqZBFb?ocRNthfX%UL+JNo>EFF#i2+D6t@D!OL2FCdvJHR65QQ`_vV~)-}^q# z{gjVMe*a{4c6N4lCRBEO-Mtu!}9#(Hs4wjCNZe)F@rar z<{RX7*}Amp;_CGek}qR>*8s`Uwc7W345&_RKLG#4iB-)$Zhp30isQqAuC~hE-Lg3` zFO7*h8_Fn0lRVzI^FG30Kc6fWiVH)%T;)riFS#rJ!G_Q$K)^}@(V*L!u=^xaDG}9` zxiIJvw+EdlqB{K=%w?L8_;9?!6QwrT@8NmLQeT3MnJ>bKwuX#FZdFd#jU{aW(ij^g zDMq%$Lp^@j^g)iOrA{2lHuDO9_>tuLx?cue{@xGJecFuJj74a8+hLakOBvV>9&PI~ zK$|nj>=5gt4UQ^7{J71Gp{|si_7xiYff_I?k@9{ZNKgZP%l0e zT$Ca6byc?MxLPDmP{x|ih zbm~cSRJ&D1{hK#3Xk6fFgw0*MT%SIrM>iX@dwC*v5xwX-YL3HV!kdp|ch>Cx#0k8+ z+ewguk|QI?7V@Ne2OH+TYpUAqIzCKHy*o%B)|hGZYav4mL~6Uo>@-v^*zVhA$D^dU zx6)3_e2IpcxEtSmKRDRni>xLraCiza*oQ%@-9tIIy_+1BQkLIkV?ml!f zv{h-il7lJSc3O9(Gg8)xj%P&^4?nuk1c@36Z@|r4w*8NzVR_Hjgm>VJz=f8Dx|p4j z*92udw~ja~gzkIo2;m146|s+%eW!IJG~;0=%2EE32t_YbH4|aL_NcX`SMr4_IPSJ; zbEnVO*^@zAHfAYNbKP$4oX$9XAvvJH<(*arrs~~sE+-;+FGL+bE7XL2dnb%!#o^%< zOQg&-i`FSz$*4Ak8q?6>t(+Bl(s6wNaD~*hKjJ5VTU7kch0n%pVyEzE^ zHQ3e+`6;25_p%KRqm28;J~!{XyBO^ z)E5F`mZH7;5Oom?;DnanyY8B)BM%T}F#DhtpJKroWDDWD9V&hGEsiXZxFS#+&9R<>N8~hyEO4Z;ing!*h7>30y+RL1pLE1wrw!Gy^>Sr1|13 ztU~F|U*`5UPHICY?!?{dt@nRHbObWygn16PIiOMhr9Qyple8tm%DQ)aSB`U(;OfhL zjX1SX%7H?l4l?e|Rh-vGVI8Q}nPb2{fp5YnOfE%Fxdai(IeyGFdGgT8oODAh1W3;1 z=Q7WQN)5$0N+a3o>e3Ve6_W%!$$&+0MkS9+omLa8)+9jiM}^rWzbKB9qoe0`-F zDZillJhca~VR}9}(#iP#+__GGm3*^4&g=E;X=6J9KESV~JfG)@c~)H_eLkaLfrNUy zhh?Q+IDRv31=O$+E~Oy0i2M>yQGO%l;Z(#GfC?BbY^-+bZYupF{< zyRNzMq(Axe(Zkq>a_#1^c`(Iu@3>)gkIv)@Gky48FWvndb``yJ-2EqaDQBoTN%JeI zBVDv5+=I^La#w9cbqVqbI{9kboqC=8YOF?nvF{ss!G)kvrRPHDiMGYR^kcoV6G@71FKeGoHgOXXI620N8}adgW0oYwn_n=HE$p5cMDXTN62rZX=*CZ zp=V&ru=PKWdpQ21obrNjeg_a(TTPK_MGKyTouV&3uYR{ZJeR&7g4#^1Z}nbVD+5$H z2Q-eNuygT0ym6t$&0}2xzkdP$7WSBg)AQMq&KDc#-|{s+Day+u&0MJ!nG#t07Oqz< z@rmsujxuqy)9vf2Kuw#a=Kq@!C`z1%92g}Zdkwrmo!oHV%G$uRfN*ND}e|vAJwk}T~MucK7R=xRnNaA&7XN>=tM)17GF;B<-}wOCEr`0m%cW- z5I5bUq%9PYa<2!YKJUwokB_Z;&3Kgw{I~WWz#&S&^Huv-b365I?|*b$i9NpxEqFh~ zV^%+mO6`{V#H46PdZHI022hqun3#~hsQU|;o+PqOt_PnR|2SYst4z_aJRN@T@o&JK zx&FS~*4zSGMaR2(+DCg={%a6@#=Zh_P13#-^`tnw4@lF(Vb7$Vw+q}gGpW)yBUY$r)pYLGj1~k#S+Ug z#K8q_*^}orD=A${RLDo)1r&?IxBNH$?>8Qo8w;`Wj;b;E>M1c8SeEm6KKXUH6ru?g zwCTg`tFO1I{%;wUHY-u$(3TyJbYxZ%f2_b1IHW};_W^%!)(hv1`3yUYX-avK)3!VQ zEs=lTR_Jk_2Iw7nU8`ZPFG4X;3$T=9dt4zIYADSpm^FvB1u^I9w&{W0>sQ|Gj6I$4 zYutOcow*zr>NcG<+67k@EJ70*0W-b5za~;{{hBRf_M5qBq0*l~`|9TJx4BQwOymM> zxTakhS@)g>Kb$c;9E#wz>q=i*af?W&0J76s2l#akBzw;j5SaHOSxC=YO$jcw`m%3J zt-8NtN`J{&$fcsCbsnu!P*iLq3>gOJ;z_F9ytJHOPY+ox-qt3`@&-)CRA@=Ra%A+i z(m_}E`dOJt8ZoMPx%b)R#`n4$RWs_t9;K7Jd!O_YoFsMinSZmSoP4M(nR^nZTtFbx z051Aa0Q$}|LTsmXwWY0+zr!U{mby>ygQ$^l!fG3UxcfHK@!ukcD^DbPo|>V zqFI4x{)VCB1ns5Q0c7V9BE8}AKI$~41%$P>z4K}6u%}1~g$T1#cI-lIl6n8GU?-FR z(vj+i^x>Qzxl$a4PuF!=jqs@0PbsxO!ashhO*=9DdP5EFL!`;OB$Xz;(kW*@GXAsR z@adf)wn9oiS!^F+_M{nb_YIaY-O8A0NXt)Iac?PHJ%x(*{)~h#2vT!@Kw7Xb|XpzYEGT@P$v?=K|ga)zR> z%qe@iTy>ctrhj8w7^Us8@fi}E>K9TH7P5373AxokGEO@SMGSVG#}TgPpKM)#rHw@4 zee+!=7(x7Nrh849aKbaOc%lT6$S%Tg9X9k`Fl>|S<@d4gfM$YshT=Ovn2&w1DO${1 z8{On9H;wWGRs^w4%J2xkIajocVW!Ys z{8com?&ta6KBs-zmA`0xedYr5?7gM8G7mg=?x4g|NZCL0>6LTaZ6XagbH+%i*jRmp zCTahjd$*&yiK(DVv^}?GvkeOG;L5_k1TWRw=!Wg&j0pMKFq_u8G#6h^3zv}qYbi6k z6~;8008#O@L5%1xN4S$uj^GiXwBvRqP^N3LJs;c5RE-9ZOHB@|?qulr{UCr+u-aOQ%rv-a}-v^c;SRwQZE*au}RI^B9tEc&iX`4awz=QtUsSvrid1 zeNs5oaqU93M$mgW;2zP=NYOxdid=r( z(&o~rJP=Yz8H~Br^J4IXBU}eK+M_M1(F{JoVPQg?C%Z|YSxeac68u$hZ|MN*e2P*F z$VugJ^_>=`l)~`%qR_tZXj@eRHrU@|YJ zMtul?Qax!W@p6eQuP%4Fu8syZ_&c)&7S(?*flBRUdGZR0bJ(3+YH(NhR*eR`?^cYr z&90kPV>qkDb~N?0K#x4@ei3yQD0$R~7#H>8bi4v!Tximp$i}>hvR#o;5xB~NgJ$s% z(D-~UMyZy&Dkx=e^buM*T;mti*#~c{WmRqLlvGPfk4wF=xNh-ySEXIRN~Fkn8|f@@ zq7#rC17ENnbpLW4}{~ z%0|2l8^O$3D4n!<7WekYnX=MwL6VG1LYAYWrx-5fvmC-UD=nbf>0AjDiyPhRue@bO zeQ6zQ;A?Hau0L-vGc67ImB{)Nt5JIUWSTIqh#GeLXTx2^>RkeU%-XvyeMK0dFurX8 z)E6lo&C`$ta&9|9Ub;!T7SK$G1Jolet#}M^-t2aV zAQZqgPRD0jf>lgZ=zbz}PwMR*9+uZCL1?~#exe*woPSQ7Ei zF~F2HzO;)yb~9*wJp8^uQ1aQBGv~U(Ug9*;vs6X-t7${%G?V7@dQybKEJAUG@NW0XZB^|+!(s?aexjOQO zA~ryj2Yr|`N6Zw68z>e;z9v(~5!6NDG4O2GFoLEUm1oc8G#=Ow62%#8?)`V+UGH52N z>{Bl#SA8W_CV(5C%hmlNGBDHfWct~=d3D5H&Ao#mwRG|ZQ>u>zeN#IAr}$o!wZam{ zj6GfY`L1UCS5(orHj1%2xf$8I@etszE7mz%UK-AD9c_Q*zb`DL14L~&2>jijJ52C@ zV`ApAqX}Uot9^Uqr*Z;XwcCh)dPg6G=PkpJ@*NP14vddo$s}yzk0!X!jwuafQcE&J z!z7WRG$;!_=p(#JB(~zCa3hGV$~o-u0YP0BiO9Fhucfy0FK$RJC2p>l5B^B^%;O9m zogKcimhl(MS>RWbIC-bjCpK4j|013`iI3xoKrKaD#*5gP!Tt6%e{B3{T;*m76xi!W z@vHj@3@s~}C}`PML-BQ5IVxr+%NK&=*r1hyeb1%Pp32duA=laX*b-_NZZ+e8_`+4{ z76Itad_-qS?rGwE`a<^^!7I=8;-Q|C%m*ecenvOtHM7H?^vS+G-ciuxHtq-a%_Mui zV0WyHik<%Szs?Dw%$W<#p;Bso=FfxJPpu|IiBJ@M(Xme|L`Rt&nu0$IK{-?wIh@v! zb(8Ov0~<#hx#!M#D$1?m8sN>X2vDDH`K|Vk2aib($w6v8x8#95;F=Y5Tv*pE1o zF007!>U4I>izW0Eh-^*Z*OSyR1uX>Y*S(iG5$pD{^jb}dMgO89A+eSKY z?uA9p9*C-EKH1aiv|K#Erci4^o&CBxgK%%xgVjdXtL(UwHOUM<&DWn25#~&; z0{5N@ZF#R3D9NZur0>U@T=nKQ1ux%fL^bj5-m>HUTTvx9Q&7wBwBTj$?Vqg`43yAs zG&l6k|6LuuR^90XdS3TyjV?MyUyQ@vE1{W9Y`#SwBd+393xdZrH}>9}8%e3z+@bPZ zr{8-4L%n#nN37K!?ZS18zl<3-<3CeqC*nuP`|kyn*&5si5_bLmR%S z?^X{|2RW^@-t%pHT$5_M@fWm#!re8(;`j(_h&0&Llbn8ycD|KOfib)~{W_X`HR^}@ z>g6p4km{$qSsp&t58w85a1*`ZZkWRQ;1&GrZ$%$H;`>>Lx^cy8*?kbk`WHn`tU_dg zP3dOdenX)TKs?!uW?}B*Ky{N*32N;5Dt^=I0&#Ao2Ph=CE5BoP&YLH_Bdptuc3fM# z+KtV5;yQTdG&|Pr#jLmM8|Dn~#;vhGG zFHOq{o+10S&f#nIKlYxB+18& zHN7uqvofP%vSz?uhDka8TUmXaBkjx|5!P6SzUAJf;|C{H*WXiAe@SxK?QffTJt^nG zvd6=Wfcvxdk$c^c1F_BkcRaVC*q(@OG5uv&I!u|x+sVBzkbP6kRePF&SORqqCJ+DJ zyXQkhn;}-|2$;QpQ3`@qUx9$<-?vOWJkTM>JAPxvbD5S}+&=fzj-HpWTcOK$1_L6Y zmE;b9^DRW<*F1Xvlli`FX`C=+?}4zM!TPM|~=VP_>jV(kNsyou>iWiD7r0q%$^X1>!iAt4=L0nA|HBNqv$F zf|Z6btr-ZFd(Re#ts*B!6u=j95I9~W()1nh+ykOWtnGf7Rk2b0TXBGS-|W>s9qZA| z1zEb)p9qlMp~$#iQ@7~#3tvlm+*X0y41P=ew5}Cud5I02OpLJ)4hLA23}zX(3HYh^ z**BDK=7CLEro;(+p;DdW`9_>XN?#VMsw~Hetmr=b4Xk~Y&7r;sxOBf-bd#&(t>rUBIL{>L>ysp?GcjMHnH{UEuOkPT0%dzYx5B4zlN@L7_a@DADjYGfca0+JQZ;k$YGaIu&(p6@c9I6~!_-`>|y#Bpn6RyK6 zkE~xDw~X@}eC(tgsxWW$6xMzxP2og*|L>we&G)`T7)L?=;kN(QFAhSkBD4=rDDbYK z3I*DJYDm+ot45`97l9vWz=rO-CuhPWO$BPhzu@AF4Og`!(B6~DQS{W>&tp$r8YwO~ zqua>LvW0_G5l~5-)WJBX9QXC#z`Tx)@yEjZeupF-J88k*_pylKjXZ$}60@>e5NU42 zVD(?8Nv@_;m@2c!o&}tuWh|eA{F7GHgER3atrk0bY|#%`dCF=_lzhF_;M_@Ce^P(* zS7d)W*2DUj=kW`&cHLoBT+{xqp77%vWKa0I^BtG@@*)_8$E0Dt=pGVquXAK1OnPM4 z;7AN#1Y4{lYzkcSctaTCZ_*q@A!g}u6=V|OfSu6Os`_*z6pm6iUdS(@Se}W`YmN@$ zK5ey~-wpT&_*xOvzoPrD(+>@M0#W_uF8rgQ7T^B=b-^C^I1~| z_n9HuzXyck6HMNqsGO7dC4Z$Dzw!&oKxx=ypKU+=18xYhu0XvVZOCiT0RD7h@Oo1H zyS?#j*XeNY)6?XY->dI?pHwq)Es-9+c>nW>(`7riI@Q^GSKq;gNrvL`4+jgU&Y6JB zftzWYT48>f6T=)sfvgj5mo%>>S7#Q9D-?q)?PxTaPiYjT{g{eHQ_nTs0@wIQEhdUT z_`S|TDaG*Q1r$qzJqF8uYW`B#D}=nT>kvL{abNj!@p<0Yq~|xNl;e3xlf-BBucw;m zTAXqL9J{xiq2X7YmqYzeM{)7FuebO@o)N{;b`AR7t*UZU+>QW~&qw;)n*d@cv;*l_ zHSRnPX;}{uo{Y~ZwvH|R&3J5<^0Su>=(a^cZn$4pz)M)?jS1{C6KScDADx=3Q{!N}&x?DSVW50iPpGq2 zR0Q8nAW<2oS5a0$^P>WuH5@SFcXU>=ASnn|omMjb=s=%D5WdJKKb5~h8A^;hPMA^Y z=7{}U%Xy{}fA?^7!0b~{K4BFvIz70(@WQTqjTDNC=E103MZ()+-~ppjPgC#GyAZit zHrRi{F#ghZd%G~I1@R%>f@y+@cMuA|82IcYl7pj;F7+v-bDVLEQr%UHE zDTTz~wV`gySgA=!36lyjHLFyKt_s1@o4-b26-PeWbgGqyL{if%}W^ zrn*G3an;e^%ZGBnp{nGicuA5dpDovV@Y;;p@?Nvb0#BQbgpLaBwVfM_*H*iE16P$e8N@_=*zmVuEiV_#DpKFt9nIR zd|nh9ZdGXo#UB-5z4WegJ*l2)EL=&_Q!L8vEi2t|2-mTKg_J@5r#41~T1mvOSY&v* zWwj1%yN^kStdSSkyB*ACU5p+oMbZf8CRcx&xKJ=Jm?@cM-G`iq^-Tgz2~;0O3up~V zB!bC9iK+zLYEiTe6*-Rv(8HMw=;pqE`}$6}LKa5hO*LW`Tay&e$u(Wn>#|%S+9blk7LFSLd;CCjx(4qMdc0SmeBLGy z5*%&5-Qok)x(|G5>+le=lBNaqJ~sOGlp561Lg7@~!v^*LA~mxYq{_mlA@oGAuyy=;Dr3 zv$3`EQy$lN*?2SJHZ=*7Jl>c6p|_Xk)_Z+^(k*9E0b&<^?>Mjcc0svxQnF%ayMQke ziE!4Gk#gqwK>wAeS5j__xWKK~f#UH|i|kFUpdG+Sd>=FJkP4gs>cs*dH$83#krFl7 zp`aVw=w~tD?pu~A{x|wh8&}7X6NT70MoKg4`*Z0-a(O5q-34Hdp@GYndCKq4WXlxD-?rJPYHWi4+uW%;K(rj< z;JCvEOW&_rE3-Xcs#Qe|y@Rz97RTpIktNA?W1912HJfGZkd*-}_M2|teeeH%B#>`) z+FcW)!q|&5K^84K*jRSCVrKHXcgnI?|1)=yXdS0sx#>!FB=_wn{95VCsxUu-zRjW#TXDTJe-#6NHwo487rq8}T znyrLTgS|!r*C|@v0~rmeA8_=y<}RxKPZj`<^7yhyDkkyYgl@V^9$LJ{-}oU+OYT6I z#lL@%;^ivbx{VetyimfC5CI_`#BWOPrUr8;`*Wyu3{Bl13WyjZiF5B7rXtU~daFx9 z^&*7!+|LN9zVw}~ls`A;Shd;ljBN_O0sT2SeI=RkKof&x+pfc&^Ik zjF3`dAj|PR^wVuTVCTda>ekZ4JV|>c3R0lQT80!|@GJk#K911#Hfd!# z$5Kc!Yfo?RL0-ebS~SkSVCrfs+>k|Q6V6bM&%Rt_QHpn!qK2;T5zoZc?yBiO0SaVq z4Yn29q=U7$=6p+Y-@kPqdEb)W9`NSI?vBRVJP>?%R#VBsDT=3&ZR04@@gbgNFjX!- zj<~H5wI%F3qS^8Gx&P^9#8)zW+o`3O2lRYlr_u|>?F|!J@AN0E2*(`L12n(q{q=!1 zxv*ZJYaHmQP_>pJ zWFV=2`9-i;O_C@yvrYYr`DR(hFnED1?@gw0#@=qr<@@kC+K^!|XcsNb{TJ+)CNj{p zjDtLcAxeT^Ttovi7*L%kja^NFoi0ymnFb3U`-@cgxpU++a)-!|UfMd>T|9_EUL1AT zvi6y2GsKrWIKh)G*Sd`=pTVFeCkILNUcJ1!01BFz?K6H5os%1 z|L2+8Zf_hdd+g_@cfTML?-lYVy2Wid)3O`4t8Y1VUUF&6i{>c;angMZ$~oh)3JprU zV?mQ7{?K6xTVzeBvn>Pt@0!pki)6>nejlE;oW>Ni#sjg-^W+_l>fm*xk7ps=HBM(3 z))QvUC+m^1lx@PKod@C_#vNjBt-nqpE0}zRcd5cXTf$!V;}=Q)D)PbRA5@QK$BH0X z#?!%OCIir`bw}sv?*yFTtP+Gk%`>C$M6tHF#u}6|H}w}&wir6={lfjsWa~I_oNzgN zi99+^Oaq?z1<$14LaR%k?%XC}H3IbM6<(gbKZ`VJp#E=llrUt@Z&7p+Nz?T5&;C^D zUAQz)yO<0Qxj+)6!*VzhM!8Nf5Y^+|E4HzLgA!J0RxeDqk1)OlSYTeXG3<;#_ZHW z&ay1VnL?hKbttK}&AE}L8q2&S3?tpL3*MUoPLmcTa{+V2Fie%(*;U&Ci=sv+_B8M?GrqOApC~-lQQ|eIY88x+(&^ z5GJb)8do0w7t0-@f42;)=KPhWw##y+o;TXAxUmc63KH-hvOhi^yqL=!QY_7}$giz&7BSW)O;GsLwVb)8!ZY zOx+lT$s7=e`PQlG>Vn3f^2KmURQ{CPWQ{R&igwBg+#GHx5rr_@zWa(wyzC0{wFNas z|N2&Lnx>J8iRCmalN>M!q6}nEB1ftqxmwNu0@QXj;9NG1-r&^jTlZk>yFZutq>BP6 z*}ADtL~e&!<(cZoURLQF8nedkkCG{?3Pg*#JCxV0xu0W^Ku-1#kmH}sME$Nbu!bdu z>yYh!9|f4EhjVBt<^`;mDlQhgj7`x@o!#e-_AN?==u?efU6CGoJUn7tuCteqM`p-| zG01>=m!YqwQM2R&jhlr(q=z2}8(t^OZ$LzHwX@WdNUvL2x0#5~l}#UCQEc4$eKiE} zOt~#T29~0QlCthXVxstNKN`Di{6M}hhuTrceWcC8 zA|gRN494EOFX?`RYsb<$uD^Fff7*%fu!Oa+?ar=4{1nkd86k+N>e6}K6S%^(}e&7+sV3uL*MwzqWkqjv;cR3vJJRQu=5&sRqn=a;Ju0L&dR! z=;FC+{I2)8T{85;C0b9&ln44sexh};Yekyu4);tFw3W=MB*h=3wY@B3l)+k#(`*=m zM|d7E@9|SwXsXOvlRyihi~YZ;*pdeyf(4W3T4SP>WYyaiO3GN;u^i@zSQB9tRCCCg z%*TvCFI%5ntc>Tvx^d-j;9QXFEymV}iWjVP_j=&o#4P8OA=@XV(7oJ4&`AKtS&2O4 zI($+2Wi8+NtSARQ@70x;Yr+E(f72CmCNcX(2hHB@=Rp4hUkVd9h#=gyNq<(dJGvOz z|MSFp-5y0dUbo!O&g^Eh(A>~`8Ajsofh+I?cn&(m*pgFiBij+)Z~S#_CX_9D;%CUb zNE?_w;hGvnns}Tfkc0f6X4w0taz?H+7Yd1;hsz_BO!nc51WJLFtUQW zCRcUlI9z2GQN8Y>lD3I%RFnPZIQ+)5_dXxGN`c>-*GdKO8eA+q+}KFit1cx`C>F)G ze^Te2LXM(0m+R~Pf3kxj5?HI`b-%E9KkQLno)uWyoUOmqq@@umR0bb5ksdX4eG}Sj z?Mt<;hr0^bI?S=1EH*fgu2T*J4yP+y0e&MK2B$E2{5MwyoIJ@jFE_zrJc{8+{v}YR2JB`oT9}C z+4sUE=@9hXftfvazrFj`sixNVc}D)x2X%J&0g0((Zi*RzZnK&v$(ISYm{4XylgUov z<*B3aMfPgS@&eNPW=zCh1t}Y}*BqTwNI|bfdGF+1OR!DvI(0z7WPIZRNyvT^y)z%4 zs$H%2E3Z8-6>k&4+Ly=PnL{y6R#l6Igft+lqgDFH9h`xB8+f?usiyFjwPrbpATQ@S zto-x5r`E4?qVioIm14(&QE*9FUG~&`3a)wT_{PF}p=;mrB&3sD`Oqk?EJFXSM+!J- zhr+PcP9I8we*zbYNvQjtr0U2OBMa{bwF`i!vzM(!^c)>;_?(Rm zGlKeR>^B}$L62!8EZR*E>!rR0OBGupAVE)CcG)+lxL( zJSj>1D=bHDs##i7e2R%l%FKu zUjj<7>1)A9Ul$a;kRZ#7LC$Wj&D}2zvEyDWvMVk?DN1TpqQ~?>jayGpxwy~GsHtSl z!nc?kErTSonM zgmZGh%?!K_);)HoxkRhA2b<0L5)UXAfWx|P)7hY&?&Z@UTBE6N7*fof+{42AO+U-= zk~d&ST9h()ZNjDuav93V5$K}`ke#Q7K17t1GKa}p*Xc7f^9H<;Fx}d39L;nub2P)7 z2I49k_Z*11+UsPY6;o8qBm5~gXg2&u`hoWg@;O{OPCI@7Hb55+=HsIryZn79Y(#9u zBcD#1oXdm>PD(C#r>hP%1 z&CP4`T-TDntB)tEqswG*o-B`U_sE>{AJ(fT$NrxDf6qD92%f!*=Mq1RDB09V5t&%5 zJt+L5XP|u>hdj1E7mV*7-Ue=`*M-`D%DP%FF5um1^ySMf_zUazj31SUM)8G|dWh_o zU+Y3YRS|W;9Y*r%+g#z!$UBOxeEbb;)tP&JGQ2s&8%dKSbuS`O2sJbCJ~(KFd%m)H z18V2acU& zvNaqXV*KgcbsCUZzN8W&ctX3@*>zA!WLxZ?k@^f>z-aBG?wh|KeGmGKf4X3j;cIEH zJT!(c|NX|vT|D!lr?4sTk+rjWuKq=CQf8#^cTvPm0uKp8T_;s>%)=!TGBYp65g-pi z70Te1;-N5SK1dEg;NuYcetnp?Fwg%4&97A3_xG7X4IFzX)pl-Pr<0r2tTh5t@72ie zWTP|q!2&WZ5k|m4=ZP z>a#1zsg+JN$DB9yKQb|JkxEo!+NBa?`Ii}}kljkE6>Usc7vRn~4YkN(4NTb|As;Nd=d714w=T9PI{k5)kVb}n0itK8qC%bb)R z%o`(~9oEuz48?^u@sVsu|EVvPA6uDU95DVnWltQ%Lkqx ziSj=n*@7DmL#@ANkj16s%`cuQ9`8b2T0`lR6I{ZZ{VhCPKMjr8-s13)ja6@Te3fHN zyb(CV3qJQ?Xi!Vs9VD&M{YgOfLD9jELFy+lbDG< zIH2D#x36*8+;#m2bjnQv=;WF0b=l(LF9r)D_SO(4wnK|ewJiv<5TB}F43L$`6L#4v z3F8{C?>xrL9^@ZeJlEXk(Z+++MPm&~31i#LlbtentTAAghCk6(P~tBgKSKlPND4@M ziJyOFQ{-&x!s+*0J;j$-`n^Y^4BR)XTfJj5c2=ZA&FA*eX}HFjW(arVW9jYJe=h^ zEjO+2g9(kMIdF`yoHE4VlseC zLp7w+#jws5=^z{sJ!DSHRd>giXWXp;&Av!Ipx5W8Wb9<()xU-&f;)RPyI2lofgyJ7DIT&waTS!c75$vy0K zh>}nJkFpol7I%=MUW#1&|0LUBtV#OTw*EoiXiAR1R`23*)Kb*I>O;afM|t-wcxhD1 zKrUW-vOJm9qP5Hv(;U3LmfLE$gvCETlgHX=!u#^AW(gFKjSKZZyRm&w2tg2xI)_OZTtC zuIPRkuqj$J&AB*H=JN17Lh-K=k~+mtFLX8$7*!!2+wL zH2z~~)xW*Eao2H=>qC;mhviQZ@;SJ%*1rzGhn&{3eg&zRymp%R_9+Xv4TtdDA79nx zNybio+n8@snx|WB;40;52ov(YZO)EDTZ7y55E9ePNolNY*mPOQ(*}}HPKU=)@@Je) zqDtpjL>z2I4`(v6H=}MeYRAi`%Lfz);;ko*INgI5tr`VHUU&`OFj-l zHGbYvTO`gweT?Nqr8a0hsXUOQuCF_~4zsUC>TQ5QW-Uc-YSov9=aYt+tcdA2o8 zr$gUK|TpLyhh@2pI46pTj0eT<`~Zp3#mr zP>t%lnAC%Zg>vUJtGZ0`JYH+R)C(z8aX6+{&Vb9DL`ZT*RY);c!sJd-atO;^O#z!* z9L!PGH2Zb(PljW7^JVHAi{Vb3lcD;Wn6Fc2wd8}N&C|KhcA6N1f1=2@!f(bgmt%=0 zv42SN;XZ|ZSCGIpkl;NAd49>DH5q!d<5sSm#F+90V>%_D!zV%(d3Zct|HULl7m)BM zj5RKs44p)JO+W|eDR6%cO>n*1C>OrRC4`PqXq#>UKv?@-K4-^Of4_MbI){M^U*ANE z3wb}h-pRzrw8t_M8#~E75P0e}2AQ z3V`|UgJ8ebBp<_r?5Fk;t}VDHEEkM~I!mNRrv&a!HR1^^!{-MC?%Vi7CV}^F=H$2e zpw6*TIVUS0tQx_FHUwFHoRJF``c<+)Q6&O2eaklGzJq|&GcIGMNd45yk_fuX7xofR zmR}jCX?v{+UEfU@l(GqKsgO$opr}I=eyAnfA2l~!*_GDKe9(pGCMJqF+Vx<5$V~5; z5*rDws!*Ef0Lgh`T z1LOvp8=yuN%yG`R1Uj}=J?F7_MuzSl&~MYFa#{T4_R2(SEtzWKO)&Z9T*+5+P8&;n zCUqS^4rPYOF9x~9kUzIZw}@-i)GIy$KPF4jU0kleD{)pmCWg?{I6C+RK0@l`Y6(J8+%p=|4 zQ0vJHXgMOdGu|d>1JDKjmRu2;SabgjZAbQn?lOMA`xayeBEA-)Gt*|U4$e3qBf5C7$DVV#}*n{BNwc$IN-MJ{Pc)o{Fx=rzEMr1UP?3q=^Tge$h3Fn-dAZC5qOPUU5v6& zOr9P247#>3uuEjDH_cfJ4*%!;!fB?QV$dOYXrpl*Qw;KNi7haByiPsWEDvRA8PFWz zo*HMc+e|pQ($sH#`k4TI`F+_tqwRu&d8(De*ylX${tN@jcRM!}ZXsh|Lu%{YF~@-1n-q5l6EDK->^zc65x9KEl4z-3kk!93-7fX?SDJBU9-82i*ox4t!C zyLQQhv?3S-O=@F=Wu*U@f1N)%MD4QYFt_9gZfv*ot=DQMTMqQFPe2le=l#u@b7BYn z=F}ih!5&Y|OvGXGb;L5nKHhurWFW=imdPX-!my=)_=fOr1|m8+B}=Se&zSKd3)nxy%M&OWXb4W-wQzG-L7Ck$JS6n zPRQ|rmLAtm(hPxgbAFED>PKwclurMHl&?<0D(%mR(d&sV09)eM@)$jgSY*qm({EAV z8TlKaU?q|nlrO)rrqh`_(?Ds;iGxgVHP2XW9u zE7xUzBC%N;IYP3P1Ai+%o<YqfE3m_WBKm z-9q(0)4E}WeNw4DsZKTF7rFkfW*F3MEu9SCUl4-O<$qzQ$^M;d`ZBl%HGdDv=*XTj zGd3uAM_8Jo-5Kz*|Nmj_D}dtKx^#mD0tA-?hXl9at_ctzxH|+15-d24OV9ub5L|-0 zJ2Vh9L4rFpPU8*@jlILY|3CNU*1V~CGp~v&=u^}=r}tia?X}nXzHbdpM<^~PKm}*^ z=F6BwToj6PO4{IQtRRs+rs&pam>A;_hQ`QTjsGtBD%T%p{s8JqD{1;7G5YhPm0`GB$tF=YEn}7xsmo&uD4S(s zM}D|(X0d2Er|!10OnK;i)$hBxSN?SvgRf4p_;vhioMSp5CQcRoY#p2iqtf>4Olp+Z z|6X0(%$qajJ|FRpQ9G5gJK%MB#YDw^$ha(;(zfe&nho$ZpxGtNdJ*P%$tZWjZ$)Tu zTTtwPy1aJ9ZY6P%hw~yoa^qzjb?3$i=w_ClJoN_|j^>M=pAB(07{v_18(U{c0T#-Z zgwYY&Vt0JrJmB-0w{VLNCK>V9F9TXd%c6HUu!l^-E-t_}LZ%3)g7;-fu!mtolTG0B zj~)OHtq%+II>gy=3yG0*s>0@T-52^xO9WaEx+ws7R?*XTGocDi5i!fxRWB4jB03?T zeC+;7{LnKt(&+`m9_#*2BTP}L2HNFqJyJ{0tvHO*?$b`{ROpz7KE4x4*KJUnb4>SQ zj|tLSLG+lvEs5*hZI_dY(msAA8pSJq_5#(YGQVLAcUPNwX=jFLjPiFVW@ZZ30 zZ6f6{#2=WVjS**^Vh6`UZ^1EJh4I`iZga0V94Fg=bX^vr*E7lYTUd22`Ul>_qCeO9 z4=1TJd`)i}F21*=F-hvIC9jf>x|Zq(lJXSCJ6W2yL~>!pi1tn6BO$%U?nKDEp>{y% zpW)+BVDGzg&j84a47TAda%!P}>>Ra+(e7mvhQW&srL479`ae|+1M3Zf5)yqDo6HHJ zqFCf>W;n}>hmLo{OD1LSNYPkKqQg0YyoYL)UbJPm`*j(-qpm zAervxK7WPXCKPRW)j@bJ2n7fg=2OxTJC^fB71&!9D3z5lK>FCth(SGH=S!dGH!(&J zf3cdxevUU5P`MR(48Rb6RxCorN|2hlBLO8whDE)}QHE1v8A#wgTR7)M<;d&I*xe<)vm-5V}ymba!v zFzgH^Q=bonJclf4-)UMf@PXLb$Na}U-gqX-aZIK5I24(I8U28|uRg5mLwCvuw}&9D zep~{Z<*Z;VM;lOdZTcYJe*)w|HV)pVdz{Z|M$yX;rvsv8V$^>7y*p9B+0!SwqZ+p2 z`~BuULHQCUF*G-yQhg}tWq-or?m!CUB6ry>m5V9K?HZ0@OWa*0e_65}@NAoYDREHl zPPt4d4g8IwUMa2112rxq;}(V}n(<+4AtJg+iI`laq30P=5~U+${5eyUh+JzVI8&Hm zNe7P_DEobtXgECdbRJQ8oW?cG=rlPZg{`3U{=Nb-{g^Y+BM*?8N{y<** zO2&wcV&XSN{l%Z(7IcN;RXo9>bV!=fy)aY*I_q2Cn%TiHe*}!pzcbD&jzoKb-y&0# zY~MV#(Q^3ZK>0*88p4#=d?>0Ft~#4SJvm7HMCeJF-xmUc>RTamo^YoCM!83L%jKR1 z;?}?!H`*kT06;Q7WR0--H&yiy#%e|PJ+a2Ty1%@2%T*U#q%82d9#2kn%E;-F?O0*9 zPbx7i`CI(2T2NDXA9ZC*$Wja$eECp5P6i0HXb#!8GA+-B`W`IL_SH64I%tA{y`2?q^tQVLGaP_^Y1t_W9!teg*=`CCVo+ z50P(T?XX&*rH!(Y&N+n|gM<&h2!n|>)?QHUn&hl$C6~e5pQkv~*-hE94`eC%x|SBN z3bgK6YxLz0$eSMQ<-`iWAV(n~+NMNR?G4E6ox%Xg<#Ov_j2VsAu|?}C^jH|vJ0_xb z^GQ0ChYnV$x_j%T9LKiFOQWbVBmp_D_*w!D6cRC#bkL?8{$}Wi_3(4+b3~S6$t3Oi z$9vr-IVI)*0tC^FwuAxYeGQ5X6xr&QlL}n8V?c@aYDx>__?!g$#r)yFXh(1UnC3;RHYI8ie`X z{K&-~dd;A^1GgsmE+qEM5NJ@rmVNKtnOiW*mA{n5!XWa@4|5e(0rp5-;blxRlzMOE zA?7UKcehEPd0#^QH$tE}dV;;jXYYBT+5dx&nfZoXLEm?7Hr3F@SNEnnnyVu z+@x5ryI%VKz|f{aD-8iXc>q*}c7c`4bV2dnVS9vmG=D@`FK$^zZKV-;K~xx{iCt3v zvd$se{yBru3pQKyTndKHP-+xLqr3CthuM*V{D`rCfB+}B zwnjckjQh}e9N1pXf6yw4MEYZ{S#vdM(GYVeg3Ie_O0Efhlq4$^K!IY*bftH5oQ{z) z;x9D*yaxB`t%`*>nO6eN0HLr;6YnRck>dI~*?}auLE;tvjwG8qSGSAegE#ld8?O`w zW*~V}A-q*^U(;i+sjvC?g8O2LZ)5oID3}8G$a|XypBr5me96f7>Kk^}ZujUJ(gq=e zya&=`k0Z=}&s*Qxr{9y65X)1V(1rDw>1B3{wU=wwW&aamA82o5CitEL@@L8h-quhK zoEk=G z(okaSw!wzY@*6Wn(&tmwYU&o?gHr|j$6C)ip`+4z_eAC~`OQ6JuQkl>8NF+T?%Ji;2z_{_c)NpUKCziOP!rl4c z=PxC(Z%@O3WRgD%19IwOF56=tCqukvm~yY8D^TVh)_#f4iAkt4+fS(XdC2~Net)M< zXEYZd6GuM8a7f0eLMtaF!;RGn&RDg1e7#T1saWy${27MON(_-V!uKK!Y%e?W=i~?z>}=IvffeH|IvV%nC$I(I<9GM^b}akB0O_X@Z4NLiE95^uT&GfE~Nc*e_|-%`J*S# z7E=`M;y59}BhvuW<45Rnrm)z!Ug#5ZVr5f zw)G4&JvZ%YRdRXz!Xw_)$|&e*jPxvHdH_6qCMaq)o7a$5a{Q9&vwxGnIiou{JgPTK zRG-XKjdXjQW`=pOE6HZYrk4etf=$qZ@aCk`wNxUZI@4XDb;!IO_TkO;cdh zzDYO;k_^4Z^n6VmA{xW-eEnTho?b-ywS)i5W-3y(ZNJjxlEY| zpb9tTJy~T)d>0{6uF-O%3i39!urDsNMD&I|5qgwQcy4y${#O5qQ;frpC?3m0ltJ(8 zx0ts(u~MfQMq1BDlo~bTB$rG0#Mzy6l%n(#UtzD3TLwoo4!$GP#}`Ml_O^}LTirb>Q?5JE!<@zGMDW9Z|@(a(wmI|AiY zpO0mcxns_Mfb}p6Li3TpGQL2qvR936!~2Ng zjB^agq&oYO2{Op}iaVIS0KK@5ff+*_j&EzXdSXfPhab%{m1kte43qeH;n_dEAUEF? z`7!C+UHl8FiA@WLUT8qAR;bJ@4gl-0 zTGvgrDMR~`)?r%O%lKWy5=g*j9|!s=Byy@KPPN983nB(+AIy*N2D?h7vsV$v_hPgZ z01KMJpvOxZ;M;dV@<~A$Eil`ula!M$cm5ASiG&|oK1Qy2%NhVI`z!!<-O2#s$m!-g ze-6U+UO(|wwG?Ryam=iq(IY0{4+xT(k-t&S101#Yl?j?>|k)VJ!#7x0pfom3%Td6jg@N zuU? z!^eiV&Xorgm{2@7XL~uZ3Yy7m6hq>n9sOuFvB@+p6}xd97lW8vOjSEU$Q#Er{MY0F z@b6x#=Fk1fDyc08oKyynL_Y%ZT1kCBVQwat4roVW?U03t{!V#UZD{W{8IXBW#8B{vf1!oX_yXh57by5zX&Gmamg zu3U7KP8>epUx<_^SVdr%9yCm7Vy>-B*I|>vnujXG;^0egy_hFODCoZh`|fofd`R8T zeED;M;1&Dl{*!FtQMGI$0Zt-H$b_gmc~CD&*QRwrq&H23S6BzVCTlUz?AR@PVdXW3_q)&kI00Q@;7B-o*zGbV=Qq(u}rpy(!C%JwY>pZ&4lbUBe8%Wj$lr{kd zV23c$7>OjR6H!@qQ*EQt&cmIt7v^hBcH^R9YN9>+rye_2vn@O<1BU6WI+_E5i*HL5 zO@^o|kijC9`wz*5n$qR9>~dsNd-2i_QmQ(v=6d1_m@uH)p&<{js`3VOWaOtwO`o08 z4w<-EBWLqXmM!iw`25nTiP}D@AFWeakF;{)I0&}2bsRx8*@R`w8JXRzqL3@3QLH6`NSC$Xkrv+2%f0m@vKlqJnL1&}MxDTfi4UOoU9&PXs1%ZsjJKi*Iw@#+0`?-f_# zo)V~)GMv$ou1NTH(7-=H+yZ1!{cF9jOki&<<0Bd{fMa7@MOFyIqh&zRa z{LNXva(o*0-(nbEoIW5jXV|}S1W3(T3mF*KW{L>je$|Q2ebK6#^DYv{j8V*(LW(K% zUyB^(jE#8);Doh4Pu%dQFG$ZZg`;fOiV4z5$%!pKxs8~)%4-dWk-j%mY3%_LqLbIOcGElH zUEDQ=bIgE40X>RAXy-IN!7puKD_x(wfZf`cUiiWMcj#ILCywX5^d+>F`rzt0jUuG(FpO@QjA>UpQ8p&*U~=`7z2#1q|X5A|O`OG(?p?n=)k%Dkafi3c!;S zmDfHCtiVBJ;r{xtm|I0T5NG4`awjdP;^J94X21rIKiKTCNy0`xnQ{o$Y-OIQH;Lvj zWXOY# zkQ+g>GvzN|bkfE6ZI{#qErxWR*Y;}*RminwR@&4wfmW_V&$Sor)SGEOZO)Yqc#bd? z?K3J|Qvry>A{AxlOSX3h0k!7<-ou2QfE{ z(hSg$X&R?_dQZ>C!SElgVe7Ikp#2r@H)5zmU4{0S;Z~sBzqp<_y^LiRU=$Aqzx+$z z?O(s%3!o)tjei}RnpAz6Swr;#wCvfmQ`>lYGOm$>@~KFJO$PNqJJ3$Bv2xYCsHUcO z+v;x{s<3M>@)jr--b$;ATm4sB&=dFD2Iv#)9?>Q_+wq$YZklMPZh?k$* z$DR?7I#IHIV47)_H9yBQNKu?ExAV~we7k1{G9mKjGKpUmk*o6S7(Xv99wmRr%~+j2 znp-O6!m4nc2}B=KMz?hVvpj{b$`@wq3aBc*{#X&yS-1=eI~_y|D&52G+2C*lGvYk$_>Ua|GB*{bA(y&;9)^{+~YKAOUs|j*w2v^P8t+xVnGH%z>%R zx4(SB|MWp|kMC{i{b_dX-wY5Kfy~f*vN#RmWYzBv|F6gX|2A|~z4xyig;2Zu`+xKi z-H&4(d`#E$e>YV3BW+H658iI}wqfP^ZM=RU01t^r!ePbu|6yL?2=`BU7*mS&+ueZ8 z57@urA+6QPe=WYhDO3JG%}@K+VBlu9IybhZzmGmz7;rrpc_8b5{|vxbCGRKY5rTd9 zKimt}y-7Hq#!ig>?^g;YFj`d{|6B-W_xO8GKhyHx&EbFc>BlqReVL*KJB!}_W<)W* z_uudKENtS(|Bvq>20Wz?n-FC2+e3`e2=4bPUV{JGyZ^qN7|8G6cx#isIQO@eE&k{} z_eqP~Lil^o_|IYe?-?1eB>>~~+?Z(Lw_!2vmjouIuH?3E@^^dq@7?uxTNmHDr@zVB zFX{~b6$0K1%Kdx!oMeU$rtpnv>eDW)Ni2=sv1J?Omwn4SV^*z4+UfZjZo#Il zM0wLN@N+dXa7T5bZ~ZM0^M|!v+2jW5EF`aQc6qOwd@Z+SDfG0RZb)i{+oC9>)kl3T zo8DB%5!w_;e~5eMgv_lHdbqV@?q|t&2t0Xga6h7*iHPX-k3G$WZpK_Z-_BvKmCT? zw<$64lSIg_e(WsouRSsqJg4D4o|MfQmHigC4j-2Sm+bkv!rf1ZHlL|+u>7lPZWWTM zqab&lVMy(?GJFix*G�?<;>F*WbV(Pm|znKxR!6Z_(G=R`A@Qt!v4Rro4R&^fac3le>6-)zg7H`tqy zqWyZiPv+h5$~vhfP`uvEQzLFDe!xdS#9R z$TokLKDAA7r&zbZWM-Ye9VMgR3KiWelwCkqsR~PdC}#vvRM&FapjjK^MWM*ufU0Z) zEwW1q5--yZ|6Ukxu1G;b{9l4~1WIK-BHp-Q2OmAt4g%D2?-e<1D6GSPH|f5Wr@Wn${m=)XhsKe@A)%5g zUz6;7_2W$04U$aY)UW4hg`=VxyuaYrPHlbKY5?z;DG*tR{+J+(t5sz}JXfMW&sY(D z`oV`w5U3n%+rsCdGVD1cVl4S`C8rkywWS4exm8w6&qWZE775nfm3mWV@UEW!YKmz@ zb%OB{?-=b{+$su#P1Q73ql&^LoBZv+75p_ zeEN+;s?f}x%8&1MXgw?h*20s)YxfqL)clRgc7j#0)$uNmq7X6b;_IQv4}`P;AZFyu zRqYj;7a(*4f0tW4ez_Nkg+Q8ER=$8zeJ#K@KBh7Z56o7Vli4CI= zoKxAhY+pPHxi*06H7vDmoz6w_sDA=)`EX|fJLV|cR};kq8RpyUKz+{6 z%k`_8>%qRlW}*lN*n*aaA61Ns>TaPcUOgk2hn=6M8WB63`7xrnfFG?Z(G}#yciig* zP|Eh{u=z5wul$GIZaa4lGl$Jme5>_j1#?9extFNx6{`o%P{BWWGAhqryuHRFx#$tj ze*%Pn)-~18aui*_9)t|d8KukO2s#*~Ew96Ec12G20mW;l1PH}sHES&!4R+DitVOsg z$y3;EikQ|Yk5bds8pZCni@2L=5R=>>?R;z&6W7>nLqU&W9%tvbF9PSYXDM6Dxci>o z*ASONwTQkX{KL5yCjh%Hvm{;>H;YM*I$}RCe`KxECQan($$6*_R@oEQ=Qs@EN9mvV zs&mg|t(IapYelx<2~X5%y-b1FVID!VR-+4}W)Zev&Mm^r85!DFNk{mR*X^v_{HzV5Cr_P4npQghm^9#^@&APliFS)c5 zj}?gW-ml~44diC2tJG4zZhiMSNg7{A1;a31`piBW6~Gz8f0x;xpY3TwrS8RgbV7OsD>11r&l;2Aef zH_tf{`=Yqa!ZOoCx0zbHGco!Zw16_TvVE7iEc9?d1@2@ao4{eRPxRT$0~yE7)U~F8 z_gAY~Pe3z$0~1f8l9duKskEGw6PtQe;{IeN&cDxZ$l~S3%Tu_-t7qr|@j?c^PSQJ0 z2CtCNH)`L77njZS0lwxv+6?Ho!^1yLx)MO>g#75YoXq`Pi2}i495#c;@#ZD5QMTvV z_9$)|b{>F>9ad_)j-~(|qvX2+0=5vcThULq7f$pLRz5>1^0|728p~lY_ShQZSlXg; zyXWrr@9KawYZuU9wsGkea43s51-k1WPDlK()GH;cP|VTDy-%jl%WgHD=M8HMB`?=t z`0q+{?-Y!r;+f*)cfgi_ls3p7hGW&9_OcA(=FLZ3Zs)8zG*^&Kb0@cvG$YksrP5z$EhwgkLztR_KL^K_KHUU*!9SULW&hkk;lvN=AzfU#x-q6 zK!LOTr<@Kdhp3OlY^d928&p(x?bS|;yaGTMpU8Rct{DwhTNiwjQXh5of%c|bduU12 zS{X3(g%(co{#5}34|#2qEj0?Hwmb0G4HiQsR#Np$TV0KBY|Z26XE+c0ca`zP`Tcrb z*5&j@w1gGgHaKeDp3%*gnVSp7$ERQSVT7xsZLe<~E?DQbQ|)q-1ygq z0ul7iFpHs91qFaiVJhldoKxH@8Yx5_H$kzyyF5R*+}RWpKKQ;fI}h|A)G$1$voj2< zvwHO2CJ*g{ya!M5*KFxCCV=3`_Dw;UkD?aDD(DcBe2P01PkC39I>nw<2AcZjU01q4 z$D{qSZN>BM*A5FjPo!K==y-lA33|<~aW+Ul&`C*GmPI)|w-}_epIifIt=GupZ!i-Z@ zW~H4kJjTQ0)hpW52_-wgzEu=Y;5Kiv)3bwqZrRxKl)qU=H@Q~w(Uo3Q(p4aA^vGny zHgB?`5X;~MJ)T%`!FM^rFzM58c#kxO+@JPd9AkHyF3IOR#)8kV!5sbtvf)g;x2h* zbl7yfx;8OhV<(k|Uwr5eTTC!5-J5koA2e!8PaZFP{S00g`#xelTmZr52H6mr|Kb2? zQD?4)aH~MXnrmTG4CYTq-s!w)KlZfTW2eyd>_d&nSPkl-+e{WUJ6rmT@U58jVqO88 zr~xDZ!jGQjs%)TJ?{7Ix9I>Uu=Bo#(Hgq>b3gOGC{;}?k-I2ohQ}mHM2z*JIU`UuA z-=nZEbFAMI097lwqTgL5kqm2Md!63UXNV8W+?hP7xfD4-xb3|9xpl&)jFBf3wR$-w zUmpaZ1?~!VC`aYcUV>zk`L9-8-gNpsyz}TIN$x!P_%e_yjX%t)3-S^wD3y~<-VudV zSe}R4IokOp#NfXQ=Z;bDBlL{cisZk?^Tudv)c0w#khCv(auS5OaLk4Ti^1eo_bktq zW$NA1{M6Xhsl}VuL#>B$s{11LtsiG=tK{v(6y z%wpu6XD^iP71JuF$hqS~9nzPBGIYeX;_Jp&o!c#NgeELfC}F)*w=I&JfL zgd5nv$BjEK%xOG(-u8)0 z45?dPhlV`K{U;cs(rs>#cK}mzRAus5G#Urq=^Bp)^yxW8hhHEhQT|!miTw`a)l?p-qID__0sFsw zbc?nmZj&&(N(SckCK@tmu1Lg2-}mh^!88;Pc~d{)hb4o`?+-u5o|h^dM9P;haYPfs z?R@NrFxbkE=v;iRMxM(6L`He^I1DfR!Vn+s6i;$Ek>te!pi(^gSfK6cM8jR@01of*hO3@ zqctx+o1h>=1{^IHyXnEJH}fYWT&{25>BV3eDf}@YDVUcrDUJ2 zl$HnCx9vZv+-m~OP_0{gYQxOyYP7WgUaP}_KyWp9sUS@U#{Cc zqk1(btrgHm7r!3kO&%so{;k^ ziB(`34At;Uj=Ox9)j!0R(}lI}ZHQ#Kfwp~M;{+r~Dn{U31<5V8zE(7kO#$iwct{PJ zi4}WJShbj8m)!nR?8V3uK*@{{WtnCNI73oTQ4825)jg*bZ14U{_xJeT{nb-c!vEp^ z2=CpW*23pXPQd+@f!DEg8@&j5>>+XR1)tbhC=hpV^~}LX;b)`;j;kNJ?bC}1oY!-q zBWyx@F_rUQd^{G^xU7UhF9O5|pAcz&qzU1(^haZ`3Q~z1l{8X@_)kFhW&dn zu%pR~J2dw)DztP9wgKvmh&dIT=sP7qqQVTDDn9MXNN3s^NxlaF*=k70a|X$LWnuH8W+>X)f@l?-k|1u1W*`6$fi2MQ2 zS`evRMW?rvVl0sW*&48ZL-t40W`~)V`)0kEh!6f?FP81V3i1a7QVikAXazeXpz)lC zCCG>Ppw`QD5m9PmQO}#Yx-ibP5RO|b35A;4lV3cRTz8)?9i=2a@`RWR^!vk^1l>`1 zvdg46KYoVwmI$Z(LLzC|TWAj-Myq%7bvZjFPM6PE_Q8a%(|1L?T~hoa;rv8*%ojO) zUzyI)NwcU-kbKIg>RTQ|M_^S!8*>Gpo6Z$AeDTtN5B<68^cn?Tc+Xfq|8 ztb!9nQXxMX9{Ww8y%9-21rmp*G>m=rQWD0qjqA*cF1JIiq=3DuhdM}SvRlqfK1*sV z;&U*2Jl$AlYQC(7&7dwh4nhhGwJ#9>#lnyI&YAQw%@Crilz)vP^!bM@2y#^NiIcgn zG4@8R5h$XzEXQvHm~iK+F0dwnPY_|{skZE^K8IC}?H_pz!6hqS87Cf9iG6KoBmqhE zMRFZ{YgHD%^@D-XY~1y|Z;ORsz`6?OqI{JWsLc0Gb_M95L5aNEtjjUT7j;Q~$Pzgm z21du;8|i7Bx-(vTAdFPfmbhvJ=HZEo%80mvP3z71@u3rLR~`h_tGTNI-Mc{AYxa-j zjKBomkJ5iUy2%@!r`?C(?ac&w;7262Au7G+G!h;7C{okQdZX|yRV;}((7PpQK_Gf2 zf|notyZZ)gaQ$ry>iS?NR`&wEiWPh5n@>ClCp0bAG=98-ribcAv((#7OJ@3qNkBvBm%cP&pGp28e7wK$wS<8%j^b;L>+iXz-%DdZ?#pjXcknK2v)}TLF*5fh zGYS{!uSLijaZRU16fA2MVS+C2mH3Q&gJXA`*uiqEZMXj06U8r~&SLQtUZ+o?9PwM<+ruvOa1iS#BCZKxE*1CXc!8ASQ=fFu28-?i+S!GN zyCjWD+y~AjU&D2BG;E*tNl-JnHUj~J<9ZS6g{Zp~22lkA!FC*f1JJohTg>+;kM~h` zWf%*;=l;zY83OS*p3L1F+_RG#wI^bY??EQt6PwVY6Nl;}QTRz3mpzHo!2KdA; zHrcZCBH|QF<^YVQnuA88niGZ?;Jl^5I4T5O3Pd@Ir6ET~CO|+a3}DdS0#B_jZ5|E^ zfvWjZ2KUV|dMSjn^}qZy{S#xy2VM4uRsvNSy?HZ#J4h<>ht>WEPWGl+_a+75A7BcYaFrqF~9p`9QW8c|n zp@{61csV9y=6vDFk80(8)~Uh76yPQB;tj3Kn-<{)72P6T;jB-dRHHH-3pv#e0xCbB z;1s~zu#Q*6$}c8I5UIT9i@278&t&I3DET*M&Wz?75L32}$mqvRRGO_#%OZr!lka(*&IXL$~5v@Qzd{6Y(SZA=$;5pcl6vluo9&el~yS)yjkgopfYMZ;da%59J>@Z-`Q6t}#8!$MXtzj*!)o%ukqXU|0iwVVsz_DTJbONM^N1kAM7 z-WL#dygGHASF@eSzt&flN~KgR3sUv;NQapTADUyT4n1#0&;7`!AQK<{{By{6Q?WMY zJFj6ntg9+{3#P=0nZm=ljPkKKjXEe7SHAKye50?IBt=ZfZP(xMDCDQN`WTSS`O+df z45~(icpW(ol-#BZ`%F({wswc=gbnM`z2rp2jrhj(0xtPSR^{d{5nq+XjC~hl;6qD0+FEk7l4;%zVx0q>NZ3jU%PS zD46z-4qmnRId>Isdxboum$G2mec-v>7;`1+dm3_(u`Mv10;4ILo4d+jA2NirQXUX1 z9_$6B5ai2A;f)>UZLJI6jVvCy&nyltrHf)ak;aHWK>kmE5EFo=;Qk=HVNRR}2><@^ ze?ax%oBD%PmOt-*7R~oVK^;n%w5D%n`P6|Fd2wzI9U=L`#Q04ILuBY~!@h{Sq&a9NK=SMXNF>y5Xid0^ zL_}Nd5ygzR`RSMbQQOZ4pHt~r+Fr@~G%`XHLXYSpOs+@L1gCW@T&D)AO4xm!QHvc* zN7a%LqhJ+hG_OA#21I-y7O~EEbs#^Q0>7rzn}3wsfXk(6l-HPI|Kps2I$0mb_taCr zk-wiJRSUAopeK5X@I`7yd#5nVB?}?Qr$B^X5n5sZGy< zQ@W#&eZpD0GLdL0{;!Ci+)yb~u+S~;ueV0isojy@s_`1Dj8pZdl0iSTKdoC#T@B7j zKA;L15_mdwRiy;DmKOqBtC-x5m*!g}TWn~$#Bxu`N3|& zlOPDM(p5TJT=KLgG87}^g-GF1+6iy;FLPr5ZS&bT0t%?XV=Nh{z5-5LXT=^j(=A}) z%i|Jgsd6|uCUFGN)a2A=a`W&i_G0s;!-Wi`FribS_rT>*@`droK^@03oEK2s71?6h zFhO+>W=Tf2oqkadQ1L;pz9y~FRr-Xtr_g?(>-Jayft&~2FehEs3s)`|JNzXxs^GcX z&E@n&GmjX)CQ+Fxiy^0z?$}efVEJP2)wul2S~)2+pM{B=qw%rP_>}%c17c6#(F}faw=H|j6p%=!X*0Jn23$cq9IPaVbao40wt8xR!x7_vxM>qhVju7vZa?@X}w+p6n}oVDR1L>P)}Jl>wM?XJ`q| zRaYc)lh@e~_w+qQx`9})duZ1_Qv3G`{*R^n=O6y<&w&MPoK9SW{x>&3C5|S}KuAU3 zA<0#Omb(uD%VfA7u-hvE| z58+|i71@rcnCrfh6z^iJD(~7sz)eTb1-99M4v|dnv$*4(L5}!b$>$+usM5C#`d^%7 zqF(dH7uS=am#ko9Q<4v@mVQ+aT9E`EO1=&(Vl!O!Z1aHQY%F(Rxmp za5A78s{WL_?p;PnuJO|Bt;X0L92h;Ca|(n}C>TLb2>Gm%YxqW?-lsXfBBD(gx#}+x z>1mK}`e)1ZTbvQP>Kq&Qeu;T1S5f%oUF=_9@w(iM65)up4|c@#Y^A*lDQZB!Emomn zJR?3!p*4adOhFYE$L1V));RV=yi(ZoeV+$xP>g0fyq~*mVI?;k2uM5II};y%M;*69 zC2l_y1~`ytr8eTqVoHpj)0|esAp|QS`q>}_xM#YU=dI|C_jZ}_Rm_F3m8)<2L#oiR7k0EL1SjPsaI# z4s6x43_Z7u;hx2p=5{b_wG=^{?Ws3rl8i?1^p$edkDA2aC%)7Cm^FZBJehk}S{ueN z&B4Ur_$Gh#8hYvtKiCz~8C+?dVX#T&mmIi`lXKxZ=(qL_2dglw?}=)Fg}tc|H+Q0i zt%w^7UGR0^Wsa89szzX`(Z;<^V9j17Cstn}_I~eJ(B6sSD-tV={Vi_hA{VxP3GU)j z_k_M4*nySyyME}T4EJ;q6CiA@!pba0kfC7eJdPuNc` z>v_wTE#&qzZPFdw_63fH5q&k+?UZ*^kE(U0f{3lvn;D_%oUeZ7GpfN)&5IKinmP_r zK|MkU$XlkQQZ@}Jb{tQ*-T!IN1wWGM7C4%!PPi2=i~4>sUwV92SPw%nV90GQ*m7x3 zhOL1^cMi{Qq0#Tyl@_B(qXlGT*cyBuPWg@Ret{!p};w9UMDv@&vm8^~Zc&yR(rc4jk` zDM#3syt{Y#Zp$B`|LR?SWZ~Q9_i+?B%Z18J+`AkTPd@1T&>AgDMAbMJPU^|7C&Zd| z*Ps;b>`b)a$$60S>-w%e67;zA$E-6HqsdoWYM8}VidXL2B*^I_6ZS}$eIv?BY~XF8 z6JJZCxm?ABGw|d`@wOu&THfS|I<|B1h38w65f*Uktcf9cHgmeCxVbZtoQUUj9Aw)L zBdd%JnS-UF^2@ELceJjeFm1$namGhc_p!@(p(~#Q0daQ-NC{3)kGZn^zyvE_mdb zsD{&&`x1PXY!_GSs$$IJ99;C;rq0(LtIwSb&7z()Jy_TIKb(DaSd?$KuA&HtbSdG0 zN|$tZDBa!NF?2W5piU zz3#Q1^`Pea@KDd6J`%J?=;xtSAMnbvP3mSlaOx!=NiJ=M+F)4)ICh<@gy+pt&MAqZ z<>PjFOYnZ#cPqwdthYJ>)fK{TnLY|ai^)~ouSO?n!B8jI$fN{kS^BpGZW)I%uuf-N zwWCc6xvq}tVK=Zb4#jCYSGAfu)FTXbWF&RIbkcG#J>eY^n-zA{y0B>YIp)SX=qGiv)M1#dA)7s302V0hwiEbsphN`)OtsgkI9?X9Q+4N*|tGa$y;%RLn zt~xoKAEFm=*vhN08vMAERH6i9gHF4N``~{)_#Iyp_wh3vU=|4XuTinR;0AQu$n7Y&_;tf!qkQpgHBxqOR- zME>Ak{b74NCQo?U>>NZG_y9HFpMQ_BAEM?XZJ3IW;+KdPDR>`4nvBk9?3;lyob!*g zY#^9+g@!lHPB-7PZilP240@?_@ITWg`8RI($%{sX5! zJ*D}V=^UkR`y5SAv~WuH1zRv4uh^-jR$Q_U&e@_v_1?@9R-)A*Yiy2bS!_;xm>b)@DbehiE%qt*!%}Qw04rLMv`6b9YNF^Fx}-G zl!V>7g0URY&yRnH(6s*KQnE&Sl%B;7Z*7U8P%k}fVrE%Hr4b+5@x z#sXpZcKRctF_g~i#79*P;)kW7%B|^m4p$FBrFCboFw0B0+F1+_1~u>$t-%ILqH}2U zwv)$U`Ja{HUYthxDT>x}WfbgxW}+}n3NLwEnG%!%KyZDj!sOenNP!U zZK1MSku&f1km!OiDh_yaX39}O%_LllceQcTBB#L`GIDj1kcS{M7Qwxe7%HcBu!vmtyIidBt+ zhu0^vIYAq_(lo2q*Ox2jdd4g?kcXnYZn6hEsjJh#hvlPi;|n^edl~X(|o-y{cLTkxO`#pzLiH&bt{<_-8BCiun$@XWZ}m4uJu8hW>jdG5fRojsy0c4ogBp#?fA9lWhYD%6qsyx$t4?LmU0Qc2>8WcV>(K<3#%2@<+PgNe{VCDO!UYsi zg7nk#QI+saML*N%;cl()tK6tCzX4KPciG3Di0a!U)B?=xw065lJU(taV#?V}@$Ypk zwArer?GIMmnoHB^xUsEQHgxgd$<9$yCX5EpogAvSnZstp0hk<0t#%1YM5;D!zy)qL zNjawC8{B><*+jJpcKXfhxXe_OVT>;BN?=%^L0l+GYoGfx7JO1;hu|K+j*!7B{85PK zRAKXNeEe%EW4P?Mby}1e|JHdoMCprkv2aG)K~^t@MdSXVp1sUM9a7*|u~&@U z{d*HFjkXX^W#=4oO`yP)x0WmA!?(bX-D$cG;2)uQ%?LStnH8iesjm3k-^|xn% z>SFrkz$Q)p@+yfH^#R>Z@^TAO@?_TBO;!y~7zb`y)(iZr z$r_nGH%qqMbW7VF8j5Im?V34!X2V&wpu$f;yKz!Pep!es-GIw%OY~}aPgHims?jlq zVXvLzu*hSwkV>3~VJl)szbH1o&a=;76?a)JUIVx0*S0X&8`-!k*+Rep5Ufm_2?U)6 zFV4UMU&*{|m?vTO&iNT9PXeqAnhl>GP0{v#3A zaj9O73IGtE@EadbYhCXZ^nG!rN^)XFj-LNyuuWKKr%3|Z5l{>x93aeU-ASMF1Bkt1 zlM~hPyr8O*Y(5kYZwceI(U^Db1Xo`MjDI*~U4BBp&fJe9(`LQ06VqYr%wj-tUCo0e zPprRg$}0b!=8v+J*9I$Dr}szAZ{CDw-f-rE*kInS^5+SbD}4Ux!q*O|*C#t1qCW4< zYtBDLXA0uB(Yg?}G1kk>&l9<1kHbuydKZ4p?sfyD6I8_?dMr+}vkTyThLz1L@QFzG zrc;ktUl#4l!OhH0w1}Iu*uTrQn>8uxza;&Dhe(kCtbgdrJN*2fAaE1U?Lhb$J@IQVrw^r<+l~DH><%ZZDn%o$T>X zRcA#9=kdD8HASgG+8PzUQd;Kq=eZ%u&C?8DHQoKbFb5$yH=sSO22efXoH)JJ@W?f_ z@$#vLL}SyZfwvcI4r8vVtzGIjHsHOZ2*Ixq2_1*wyhKh50)cXA=3JOJT|y)A;g@-k zi>kU&zMro_=1|E4SIo*Q{l{jFPa{XSTh12I5(z}E(sj|SF8!S|IdANpxko$Ay~yL3 zMSAbZftB*nv+Fsm@z&g*s=h6A8K^&#?*bQ_-a1$xl*@QHjXEqcSVy+oDI^u*UJB*2 z3SfGxOP1WX7HQj=euKp(6LUxi z$^_*zCd|tPR?g%iY|Y&T0~I~XgO&aC1cpz!c=1eMZIo!k1wDn@e{@fJf8u@OMoG*K z&(URX>MnG;RCVBr3Q7>Rl#M!8g?{wTj?q^jA9BZrCS8RzQJw|MiI$VikpEU+=Cp5m zndfK5~ytN0+RL$z?LM1=@F&3b!&Wd^yW!fO_UE@oIK*2J*Pb z+=QpQSOXi@Xe26Y?}Bz>mSugrH5HbO!X+)k*49Q948GuJMjx^#QY<;-R}0*Qk}9hi zEXAQXeyu9&HGsX5rZL0Uko+a6aHYQv8yela-RrUDitdq;!sCrKxODD!q7WRU5if*} zZs~DZbr`3~gzz1tPU{P0rLS+179GydM~$EIp8)Ju#QV>b;Zj1e14SC!>}7n?6Zpn) ztxqaY(E0j8qtyBd-iw=Pn>iN$WDs)~-d5t{m;fqj|3tcUvnb=wn|>YJU)nP}IqpW< z0o5(loE+W^W>4x&&<&%$I$lF))~IB8jh;16K>!mzFUbLqQta#d8@w0XSd`{2>QVuM zVr*dQZ8J^N{<9~9I+YuvddhI0Yn;TBpXuF{0GA2riyP2VBBM^8%z8L}3N3K$tZ~uS zRe^TVwSa~$+3^m1KgSD&D(hyoX7zMwJNTt7`96=QUJt2LGk&2KS90d94+q?Hgh2WN zt^J&@@$Hp*av3v@Xc2=SzH||}E2izGl_}FjzPWd< z-oS=vk(;RQ@GLpgx#h7GuiHWEd7sdTInF*QbLs@z^2#NNtG0HkDzu_RogX%Fl?5noJndSCv^RQX4~dJu;vg!JQu zIaivq%u%GsI8vxi4XeJpVV0Ks zJ`3N0rzm$NJsh#-GeAO8Bjb;Gnw`*g6*zw*oaOMEbPV(9I(uuZyKK*jvhcc{(BjHj z0elyq2}4h)uEFdJGTHX4uAUI(*(a-{s_Yg>sZ%2@W2DJ(n7uQL&3EivjvXh2kWC57 zq<)aBXr1Et;RQ9RG51Mn{D-Xs`88aYI5C#FspR_EQE=!`w=6lA*sSyVhXv;tCW+SB zsAKgF7`i&LpqFq?aa8w0A?}wVU%{x&97a)}>kK&`(php0HRcy1SS}*3;W}8Of4t2) z_pkW)T_^oD^E#18<=G_4T}Ji|t!lotK{B6w*xK{$QyNf9i|#H$w^4rz zR`vobs7n!@lK?KsPW<*{7Q8rWeaY}Zw{$_pIgbF!-%`mP@mkGLXwnFOCT!MIHZhcs zhMH|DS$Am(1kiCbXNk=(Rz|o4DmCpxA-)PiS$`)5++Bi{Fj@IIZ3#syHCgmDgTe`& z43uEq(AnAJM=$7RV}E^UA_M#V!(sNq3G?(VJ?T&qE9*#JH$ot+uFhwfx}VHvF(nwg z3861R{n|z{8VB}BywXlQ5!c#e+fDx9*xFlX9YKFBPcD@l%$fyIXaEyX4+HY6Ays=0 z!UgX(zdFu?dsZnRcn?fx*=$Ofc39@K2Kt;csaFD-O@=4itq$0h>U(-=4(DD!PRs}? z{~XT;kc-~4Wnu7P2%FtZZE;AKDc6;e_j7%(f4R|7Ur*)6F=4q+ zU=u!n&G{PZ`+j%4b9-7bmT94NaWeMV?gy(0Ab37$!AxlJ3t&ucEW8^KOd^k2@1)O zKL7SviH1Y-F=!sgVAM`iu)*cMkp{@VyUk)xiSHm$!>!hdT2oPkLp|hSF~%iP$9Cl6 z@yl25=cIus<7`1M;Nfz0Ocpe>Wcm7sw&K{!rV}U0N$OcQl&N6l`KpW3BIt+kX2O~A z*WD4J#)$J7IXHEr7aBQz&m)c2D@&`eh6xAopM?{@inawhf+8Ttje!bts?x=NI#~^O$ zApAY_uzyz4&0}%<>k-4qg2ckAsY}cuCAp&d^BZb901wgopFV0q1PTi2aO3+l+@#vB zrIk_P$2!x}#-K*7;U~PIqHT!;0t?#*1@O)Fj5cf`{bRmWZ3s#}7@^)JHdhhX{o{#Q za9%x|`2b|pD0Dl*UTS?&%CT|Jw%s^d-*rr}IIrSQJ(txgrFN}AM{K(Rz`?kgT2ugZ zRG;@OghN>g!P>jKjJEfBTf*=LY02e8>B4lpJUy4@`#d?8Ik(}708U>2nMZ3_4ABsI zemfjJ&96-Y(c#e@A@QID#ljsM?NWv&KT67T_F7Zd0FE!u04;fLoF2ENTwNMeHnHU8 z1XrJVynNh>y{wKK_~p9vXOTaG{kYcAj>jK?bApHIG%KV!8qCh4;2&!U*!{G`BPyZf zYu7C%#_J2-T^@X@&Y9wuzFr4iuP+V586DG*`G2P7$LF}XzryqNR#Rk8B_*Q^8&O$a%4XdKM?PtUv1#B-a5m-oZkvN z&N)Le8-LxiNkITLwR%ry2*`*MobwO zoQI1jp@Ue4w>i_!CfOqP>W3lv)?N@rTEnz)EVSF|%njKU|LC^KzVN`AmZJ4G<v{%)TuQ*cI#0~e~WxiJ>JWh zNKOQFBJ*}iH#qnKG`Sn{&sqTgjl?uxJ;+X=l!bj$+&-OS-1R;kK*L=9t!BNP0F%I8 z!wd7NQOtW!VLe$KfvQ=BtWbQtE-<%0!$|xVMo1%2>B(EH`1^z?HpWg(WSlPR(N~`O zcs!KC;sDBAMQOB*aW{tQM)3M4fq=xLrg7O-*7v%04ieA5o@S%)L&-kJD^tasBURCv za886OTMK~J5n&2DyJi+)W{4`KkKC97F`6Eq)Zr1RwUd)C)=$dZ=%E88B+z4u-K?F3ujh5Y4D;D$^$K_1&ki1V5MlY z6GP*bMi`%2LHpDSNmYyfjnysV(@dM~qhefhUykEX+HY1STeWNzfv`?f6(Q016*qHL z@MN>T7Fm$8FB<9}UbWFjLApFfMV0&~weX8ktYgHxI?tDN#b^)7P@a&)->A7cTm=@7 zrL{s(zB&o;&(C@rlK69nrv$S#%-!j`sdqII5M9_%1Sds#CWvgU1!KsYJc!5!gJ;ii z&iOkjN{!QK_hWw`VM1RJMH+jcKi#9v=mDBU>{IdIt_MK2&^U^CE!&JH1IQ@`3KFQ~C)FqCi{InfU)1yUF(Si^lHB{^L#2i)Y50LqFh;FekZ>B)8&%Lux_#z2%#EYmZJW)j_XBVjX_I4?AhEMD$CS}@vC~IKQ}Ay2veW_)ZMUX2hb_`~ROJ(B zP|Pvr)SqwsI6jT_eYYD7x%7U-{4O*XAKpE)ZnPJ*vsPjN(gi`TQzK*qBPt z^uu~V%IAu%-mepu?OxA7nBm>Uc9-J-ab}h^AM4$<{-|ZJ>w-=KZ)})qGiK_~-{LAA z@CG#$Qs>6WBCOI5Tmz0p&y}l#f5cMdcR6Eg>=xuizRkPBsq*QTuxzSRe6*QT8|I3f z=AY*OwhBYmy>#u|ZK2P+ov2DnK_A`dB(hwE>R6v0W{8vpm$9@e_Zi4<9 zNcoFX_5k(qtGl?$LFrey`_Y((Vh==%0(b}tFaR7EdC^CId~pnxp8*N@1)2LCV1z!U zrcJE3-Bj)@0{?7$Nd0-4lkO8bN~pgOdsuLf7+G)$EedD3<>q2UIB;{;cLD#->sZvw zn@h-c!Nc?Z+J!ua4qn3J=XfZaHQqv$6TW;6;6&es^&RKx-LjbkS~FX1!t}XJxb@of zw|B?$!fDn8g~A`#y3KBh^39?>7-n zX@@5i*Z)(SI_(Tg$*wN4S#g z_G`7k!4!bP=_eD9C%6vXWRYPS(I2@z`P%L&P^r8Z4c>jXBqvnvz+le5>&>uC{JM*5 zMYCN*FINWwao}3mx8%Qmhw9oDbr~Be>`q|&A|q7!l4y7ksX$1_yK}W=qxv}t_wW!m zW+Av`FC0(#TIZeag1~lOB18%K5$i1|H0lB5TC?kZxm>p+t&H*EI)7ZoRpN%V@!JxWf#3~sq%IT0g56CGN!eFf~W4q8pOs}7cHN@ z?jDm6vb+phbWDSxwV5Y0a%o!EWXzahXmwMo4CRQFW~278{kHI+W+Qo?nB(P=`LV z1~2L$I38E};JNeB5Z?5;cgNMEG7F2Ziyy2qG8sKGxbB#!X87xNv=m>MWtLmH# zw-9S~0=2IKU;-2-@%Fq|l4qdk^H|h^Wh$Y+d(X3xX0(sm#zIxL_(!<788?ZR_KX<5 zHRxRl&-dl{vZ_X@TOd{Job#ap36g_uhqhET?<#UI()gM&QJ0VOJ)DA zCZVSwt%W1ktodl~?aro2!vlU}%0%Wb;gWR&hT7v>fUoepD`0C1^?!`s;8tyE+8e&Qw3 zmy~Zt8wzixlyn#5=*0ItwxDfShcnR#VL9{CbiDb}OY)g8GgShP4mK5ka9Iq4=iyGx zio;R>l^^v=$573RIEX{0G3|)qfPVlQ(e-E0H_+dKggR4+8^|XC^8m3r4?Y99f2N@K zfuOIikonm}@6sy{Xq?J?^g*F{QC_>mY*$Ryii*wRm!SPj4T*j+HbyVXKoA}gT;1g3r_MR3 zCY+w=c9fiomDTiaGsswl3CGZIII_MpC=|`2OdyYbdHZH3@-u}5Y2TY5ib13pU3Ful z&OeFCmP_h7uM!G3EmG?zW<(E-oKUV9A+egZq8nG9L-XUaLzRSYUp^4yBq@t1)9<9lC>Pcyjx`$-c}DHhock#i?-edTTt4dq_c06}D2hJ^bL zrn5@DQ^DHJ>&)C&O<@vT=A`64n_aKLWv$dmZCQ~PgWL(gp#+@9bY!D9X1eg64Vl&jPS>I%3Fz^$D1S4{|LTB1!Y`cvhFngcY_Y~Y+Hvt&JM^Wn4OiIg1SYxzHtrvZ)w1tE#M4>zk*iH$ z2Ve9HbV>9%p`jZ6CPTtWwVwIv#f!kQnD}L~063afN0aq5`I!l5;pdR9C;5U#Jeml< z5Zu$-F14mtIM(*+4AkT zy^(;Un0H6>-;3zncUA&X0WwFT{5$DA5qN;gjT4~Qhla+QiT9A;ZC@bUHRT^7rJWyv zJ_J`UKN+aSGQ2(I$#~MXqfW%;T9gt=oD_8}15L z+*s*!-d4EOE;4JDP2q{YCNlUu(X7FbhGCsBWSvMugn{-IB>pr016A;MbnyVVnTCU4 zL1)|S$1{*%^%mu4Wyhb@%)?5!jMHAhH{67hHKFl}OFL*ZwT}GDi%V7|OkGv~2Lg@-mj~}H6x%S;7rV8c=gw`!*ursb4FG?!fy@n5|JBV(+I1?7i>Q=_U}FIScKP%?~-y&jvi#mU2-x*L3D9e8i#| z{2{j=wnyt@X&0V9!vTLHj7REPs+t|oGGV{;v&;M&?S<4=1nPv7&e@x6%&)u#G^xL) zYmF01*4h;peEp@STs#4D1^9jMYZ#DC;#bCeGcAlS-EuyjwTmMO0mG+03-8|V-M?t> z@1P0Oob0?X!Mgw9$d|jpo&U++B;OCh<@P_Jz|iqf@);C}ekl~e-SZf1knMzwrxZ7B zmDjJ@K|IhB{&*NmzoRPB1SxPsOc*U+GhK&Y18GZ%8Y5r&awkI~Cv`a1-+nB5LL;_%e}laEE{#Vmui}38PoSj~&|^!_>0!(N?OX&Vf4n2ZmZZ8V zg73M!ns;;Ts!sFoHNpKA`Il{2{GC#W7d*wdHxt<^?mi6w z_A5Rw(!G^?fLc!h^p}KN6Zzh6{Z(`Q%Rb(J_)r{}%LlwKET7(cv+&}b80-(BX+Qqo zzu6oZGiz?pUu``qFvIwbe;vht9{ati5XZWM2ysUt>i4ITpBi{7K)wH)@!lxzee3_{ zCp_Li5T(Tm%;vuLitqgr!1N?<`>a_0uQzcHGav>z@CLE|`>)8ki%=;&c+vFV4DG*> z+JFBKLf8N>!opTUzTe|0nLBYln_cet&*Js}DxbeS(Py|5XU%oIcaQEZ|Dpx(^x_3g z@uUB7NB=3NffjeW&EhfP#QnuE|E6#MdQgjjtx%V^WwZ6Yg+5Xuh|Iww~Zdw{o*5YINsSQ7F%Qg_& zb)_@c+%vR%TP@ZRn~THC3UeJbST6kAOnHQ;n8R3!=aJ%%>l-1eHZjpqW`RJq^`Wt$ zId9qz=Vj_K#rS+~JC^-_W<$}deb>t>5ga9I`Ssux@fj15pZC}QUytLz2E8Z&Rr+3KZCh3RkR8E^PFU zK@yl;t*xg@jq+v$%8=9+%H*jPGINp`Q@+lUy_LoOkfD+CQvIg6q0p!WixC8GUBDW4xAFV9Dul>G z&c~Ttta>kSg_TRfWv^Ue9J&#oB=>qQ6oU8;@?{>CeDTz0I#*~8qv@TQO!8MM&;(m) zr?KhdB&enw%FSi+a(CQ2gtuF4sYkUYXu||MDsOE-?~Rr`F% zY8%;UqW;q(TF=Tn^Uja=&C`EWl>4NB@i`&=YLa|^2}Z&I!433<%BS4>qX=)Tos!S!)H&eSPktvqvg-ng7LAy64RbjY zDemz%SWZN)CSe^+*Vw2w%6e+~{ff3yh3YQya9)U&=AaxZkg|RG{6^k3MQKtnnD$tm zCXr5w;VZIx5>}7#kCrxsNR~f`LefpW8Xlc~b$KUagcmXv9+MRpEe*|!7_yrksDMJK za^~qH;OKGxa67sv%_?n0_#p$$)|5lx0FB-jhV^{CUq_!MKhcKCM?XCg8iE);;#ACg zGXJ0E=%M_b^$-%Ve0g659Ra&^{Ad+!4|i_U6P>ArvoJ2p)3{z75)iW+^9%T#_gLwm z=BI!B%|@P3nd``dxA6)&8RHBrvG5~zqi?k&fm_Nrl_tIUcTB%D@BX?FyWKC1O3S<0 zylU^VI_FXGz$Hf7X4eTx0-J`LS8kyVPJXd2k&7nIfCx=~k3`&Ndndi+*e|lmirlah zNJ2X$T;hG8;FHZs0>I{r#{FdZ|1k&u_#a`zI|k-03|;%ak;NvzGw2`jj7Fnb@E~KK za;a(7=*x!rR%KKgtBb7mg3K!O__Fbv6l0#VNhI}uq#&DsYQ+#T>CuM1#FkD8z=Q=U z_HmHZPz_5wEzY<}zwzEcmHTp-Qd$ZAIjy?4tDIB<;x)6vO7~3C!fww8%s%~R;(J~wAQm1aP=b? zrP>dHb=piouQ5{QeWt7!`DeVu+ud6bmM0beK{|=S2zGS@Yuf;R%<1CSHyw7TV@;lg zjoc@yi#3xws~!cmQ6bnSGG8+MuIxa1ZQ%H)gmmxo zWfF53v{+~QuS~i&ww!@PkjtwqF_%BT(WEozLM@L?e9h_HZ zppQo!#Vs;ucpY`g{;P@4Ra*yiPFLwQprb$9ZmXD?bs8jj6Y~|alKPVT#xh}S4bG;y zvMH>;kA@8D-`~FErJGHD|LbB>=yt@w|5nCgNH4E6g-!WJ<4Ezw$8%?O&>X7UB0uB4 zmmSqk$@u2Yo!XM6?ckx=*B0zcqo(czI6hmp6!$;B!4+lbrzfEuoU|P$f2@dE?4M?E zgHxBgqQq3XFd@KgdT^DptBnIZcCTxf6ty;a48W9W`PTT}Ed30t_J%ujtRI@cr&%3J z$%w@Fi&^fBd3THUi4cO@8O6JMwm0WpeH#~i>NPIg=we0V{fohR#Apl16_W<)UI5l4 zhpg$W&#gI$xonK5o`6`o4RkI8Wxn*HkWP3Tn7%HQb-JP8{&O#)RXxbM`Ic$-3PFx| z{iXR8ui9SvRH5dl_5|G~s$G&xE2`sDGG6koa5jsfkty_K8v7mY1pR&ADsiR0kAfMS z`kAQU(RqY_kD&9-s2Q05fL3R1K(*e`PnxppsmU~A+ri}>YTCKS>Fog#84iLfjN6|e zggZ&^A(z8NsC*>ih(iU@(wAe8*zJTBLmm2|}XiQS>_t&{bspGUo$7v3Vu{&B5R8C%- zhh~>6i$X4Y*T{`jU9N(+DQsr=D_&3NbjjO_s?wEcT={k|5f_tf$1wkQw7?<61exgg z)&5pWqq(Gd&DR_?; za2nm%LhKh>89OIZSk1pqc4*rCIvYRri8*}k@@GQ|oX#a~FhDW1V&in3jyOTmvYqV- zESOA%AcS|1<`GxbCdqcXi#D$T2*m4o`3wPGbnq0C$>gUcz&~Si>#o{bglu);{||7P z{*ycK_{`!(H|jk@vd95|;ATmpRT6Ci=@brzMzs7?4m`sMdTW($JwAcz%8l2414_gL@JLw#f>7*9Q>LMtPFX@u#);SXJ? z$EgyWG>t0Xdaph9hQ2&@fVr?7-83Z@s>itR+Hl#=OGYJq``k~^*CLTtyHj4K!Ujhj z5&d}yN`2B^cDrez-FoP)Hg47D%2|4rte!k!}i?Re!A> z>x^vP_vH+VTqkGDg$YQ2mdPkh@2Z=Hj zi$P91gX_l!g*M%_@}fN?qR%I)vUZ8b&T?T~4$2DQKVFHqTkG|FQzYh^@_9z5R{4B) zwnc%EgCw|l6UDlXC^Y=7N6+(y-%vX?aID)f1w|=G!4p@(uaWR`_)F_jQ2%`#_#fKX zdH~l(pa&zJ{tFR63E+lE+yTF--u^cU%HR9Yh#>9l6q{cheYLC1zYPr0&8q)`YDtu# zE9Paz~|1SHUKHY zEh&`#O;H-|xPsc`ALq7vBWE^NNa0IHD*mf$#-=!Cl!8d?Q#cuiABgGE?C7}9GJrko>= zw!0m}mR4&E36-(UAt(G)A(kRECa`oq-m+lW=noGp<$QO_aBQ8)FZ$a#tQJ0H1A~vZ z*0?-YzVbWXU_r(rV~h(f+x!%K6#?#88}80Y5&79ACFw)GZpZWaeEY`KPDEe^54?9 z@;*D@#~*FMzT&LxwhL|^Q+G`t5iyUz{jZB6dy2r%4r@V$(yrO~eE69}Ls~wzDV-dd z{GLt~J|a)wZ%;tw6X_FF(u|(ovo8P1O(Dky{1jRj}Z*qUmg zud$tK_aiKTeUV@@pH8<(V$u>XTZLH|Kw!}?**>(UGzB=V_ezwgGfZkUWVE^M5+^d5 zI2WXHm=Gt@Yy7hF)T2?W>>y#cS#yabf>HhA_BluORqUr)E>b&sM9gFSWu~zdI^Ib) zRcTf*ug8a~@mbT!xy4_#L|dzVPL<`GI@X^oZu9c^*O8l&ODTF+&ys?Ilqn;X`pq%s z2q=Nc+nruI)?=c%wY?W_p=^g%dPq)d^U4gZ(oKDtrLV-!)(hDwP^Sf z4SUlo@z-^smJ;ruN(WNGl@i6R!&UlmhkZ*e9m*w3WdrUsO*yvE9^*D_Eyu;%jtVgv zqoBv~X_dc&=JL2(^y?qgIE)Q2)f;l{x3Ct~405 zFze$V zCm8Q|yJU)gHA}cWd?WU*7g6=yvKjk$W0J!_<jE8$F-9%x6k@{Hwg2_l)MBS{s-kSfSSjgykNw{W{w*ijI7Cz=Z{`MKk@Qfe-&ipu#!%^`&LBikXnhs176c(E;lML+vzphsx>Puh= z)|P`PWaVe)QunUrNIX?)`h^1p8yUG<{VcFMtpEIU&ujT+Hun~rc33)|iXi6scz3I| zyR&z$hmrM6PK58=sh;dMy=OlmnXu0^hhsS za23R#0*4-r8%D>!>NB>`jdr4w30jGi5%~^`CJrD0WFTz%4r`ZrXON{} zFO_NbA~UtE=YoYdOloD?Qft1m6%hQf=39G-Vzr-~YrfODZH{)0@4lsSeg!st@@lld z8+v!n)U6Ip6`#*+x<0e&cg+sSs`nXOr5lPcCEM|vDDtI;T)F^;-8E;Yy;DKK`*at= zyw31Z3$9iq>>jcRE^#bg`6K_B3O!sX6DWm2;#+zU7O(@j+2gHIa`ozfe{?KTgQ`?4 z*!l~4A*lB4I@bC_x_rh30onC7Xj3)+Y%19Q_MM7_4{g8C_@da-m$RgwZq47@r3p1F z9~@O#nPfcHl_v?YV9r^;5%?@P4Y$_o#$7bL&}3FZZbM)g&( zy7U+p>#U27u^sb=DEr@DptQUSYB{L}ub5v}I}@eU>|ISHFzTrD!23Z!!5Nh165cDU z7mQ?5hiiip0=&LCm1G58qESmCs~O5@5-w7P{En+IXa_aKvZ^!XQree#2rM&Y#Lr`2zmlm0%Lv7;aOp|aEpJxQA zdg`I(HemyY&mS#msw;6PgOzg%v20r3!PTJTN8>ekP=hvFX{CjoA|N|A4Bx%E+6^2UXAO}Y-sz4Z5BiBjVjF< zAxEi&U_l8hzZxlj0~1iA0P<=DxBz}ELIo0Pc|a99-zu>bCe^l69?=y2O3~M6ccG~& zemet~v4VT!Xme#JtoAf4rrN1yP$G^j_w!Swub*~%<1IGw9FvyWZ55fNlc{2Sy)%s- zPG;R+QzItTyB80`eV7~oZxOljzYQklu&1k98@LIY_ZdrNi_s4Q>!aoC z@@zaK5%?sZA^47gX=@Cjva;r+h4m+nE_-~#2tjKWNgj5}?|B$9{dApF9-cre0pcjM zGl0E|w*LeP< zuvV>Rh3K)A%*+MXjAn$<;h$b90lqqCx*YUDmrp?|KlY=6v+bVKXV(-7POJLW*1J!% zt4`lZ8D1BPbQp6R#N(DzDr83UgBMho6d9bXCZ{i&vv7jaC|=txE9eRTy%Y@Yqs`R> zRB0lG&F*6Kic}7$j6?FFZ}Z6Zx~K2XTPT|jRJnp-0#W{9toU8rOC_Ntr=A=pd8@I4 z0*W63jA#@DB)}2*;7fOi?2A#pn%*zq)h*YhA&x8nUYhBymfMmxxk#E%?nbI@Th7I$ znUXaw?)Tp^eidljb1$5k6Rnd6skAmX>`G|rUe2YSh$`CRy_?c_Zf-71ECF8W`qW_k ztS^HeJJ-RZ4$r?xl++P7#)2?-?P+7m&hWi)V3#5Iq>C1~yZE%u+ zux6UmWUDGupr(+AP;}uJiWW%$E`ssWM`PJb&} zyvVoR3fOVewU)BThizNsF+zlQCGqsUFu!(-_NXn^_>4FB()Uj=-R^m7w=G#{`TZ7N zI{*A@N6ExzkOmD=pGaz-~|FEEY4H_flD-W%)=gO6$6NRKw!A^HwJ3HAGVKb2hpKu#qVq z%R{fYG!j+nd(8SjPZj+Fc8bdW@W|*zxCZOCh;NgY6I|K0nLt|Zg-|pu)%(S9DuLiR z^r;L5A}4~LWDRatN&=)o&C0??&i+Ktcw(+XlLcKO9J#*-flLj$*?RFJ zN#6{*wRG1MBQ$B$^WIV_B&nL;Hrt^vYF3L5RjH_bdWB5;%`r_nVc0Dl<@V!+0Xm^ zqw6Uw|?LVE4QLi%N4= zn%nP));S)gV;0VI-M%>8l#iTDhH0Kw+TI?`xiHs86OecV>lq zjZZ6+O9xaHJ}K-?m8{pt@CJ#sw7oKNva$rOPFvB>-^%Hq)l<4Y&SxMsjSWGcs{^j+ z0uU3?{6=m1({qNde+5GJm(}bITDjjw?<1f7kAV7rQIw?wz!?=ye?k1i89jkR0cct| zZ?>Ad&DB`4(b~1D<}Mq3M22H}Rs2XleSFLM_1pS>M(vkzX#S`eF~w24oJrr0lC!3o zQ%chbnsSf*x4i*Ych~kVyZbh!w`1MWx$m!u`X7zz6>70p_E03$q?D=j+YxZD z@zR$>KE2ZpLQ2hAJEUFbbm{r)Gbr!F%;{{oh1h(pLkXa7fd81OaHV%=c8n}=nfvZF z^1i*AP^FiO=pl5)&o47N69#0R7)tWNGZJoS^Gqx;TXFAP07-ZtiCaYQlQhUV6n7%q z#xl<4^P!1OTQ#NKN}F+wOWV={h|Jz(G^6SMq7J1Vuc~m zb>riYv-WEM=dP#Sq^GZVYq04)@<1$AhHz7so_?;;UF7@ZOO7>>4J}5`;g2(==5cc! z@QzCCoG8GP^YiwO8lGAn(A}fwbR1WRDOAH^29r~epChLxP6Y8oFsvY5W<0kKjD7yxNVOMKx1z&*@(01lAby6EJD8h9Dd@CNV6Cu{i*xgvTgvu1{Ma zufGLjXVyw3m;cU}k1>5HDce0yva z=bQcL)Y+LR{s^8&I_4a1Ti@O5CJNPLp%$9^y&X7~?Nlz@5BNL+tK^6$iW?IWYdx<^1m_X3iqoXj9lPqEHq^<+(!}FU)Tw>0 zJaN5V&PiS0`L85zLx7UID;1&2eEi1(g$ZX<(ayvR4DMPt0cYDy8>KPghY~B=udI2G zL+Jh;&J8v^SOG=)5nFHulMO>l8@w#ZmwNF1|2Ixuh#2} z1@aI85bj&LuD<3WJb4~Ri(F6im}a%RAKWB;jCh=jPXcREF(!WY-oQ9}eQMe2MAooi z@^cRp7@kE2!~;#0A`S2ov96xBX#B_T{>XcRTn868Oi==se&WMRUCnoML9E<3KES-S zT{SHDyz`Fj4DjM`N2+J!w82+N&9*K2bdU!D0)|T{&t0`&iRy)Z@_uUoa-QZUkBw&Y z(4WFuCP5Ztz-+UEqf7@V&6YGYbTvCbWM8vItOO+ zqyHzs$1hhP&vYI{QvV;``1xa%e=quWaXItsa#(k|tF;B+a32|W-EAFSpYkE!dCje0 zih)w5CEf_!^)Dwyg8Zq%ovD-qNZW)H03xTbIczRH_vv` z;zTxq)l|GttuhUr(mP3mzb{N_Keb(IG+Q#=t#||Mbi5M(2>Hbj={!qdypf<%{Q1Nh z;l7NDq?9Z{#w2SWE3Z!3PmK;6 zdG#Ek!*^ummeFEF@#<7Z{P^k{4+6#mZ_Gom8AP!%^#CCIxs9W*b9 zB9&T%rKW#0M`n9%n~}+3<@NSu8%fAucBs9x7B_~t^_}MRYtW=v0#lw*8&TpVUq$!4 zq9ej}xuz+V;u0`Rew_o8&29&kCftwlF){E%sQ;wmS2kTh-;};PX*shCkRC21TSIby zDRehp^a6QH>AHKXa)U+tYfkOYa2%H8z0rB_fO$3!F*}oG(RjsE!iFQ|gWFTb3Q+=a zVdh&EZUfRn3vl4-mgP+Hz(xn<3C&8oR}huD?0N7bBZ|p!`6Ik|>=?gn76LCCxtQc# ztz#CDA@v-8={_01Sfx>t|6ZK%>3gvoF!%jV1M5q7hx+Aeplne7D1Q3oj5+ljVoW}+ zwqU0p>5{+H>{jXn(s21P=^m}qy93>3L6On%{7LXPldf+m4w0+zj4hfLpnlD;&MY8e z9j1O683XA8z(e!ZJXvu^b=_h4)oVg|8|%iJK{`wlNdK$@1kw#mE+{(s3#F>9No6Wf z4_@i4n)h_7XQRN|n0TlTM0+nTsmW1ZS;b#tX8mb0%zS&izUCdTLt_HlDFSW)t-SDPN6$3nUoiz-d|Kf}wvcmV+j zCQ&-v9|6d(-bZHPFCJ*&tjK|6tH@*RGyvwxqCfoX?ivxv*#Lo1F<*RqG2f_ir)G^Z zUA`ByKi3d>Q`eK(lYhEsX~|Qy|AKoPwh<2OaIhd@yFn*k7c;;QpjT17G(%rliKP#GZ*kOy*$ zQJj9f2a@Bx14f%$#ch=Ao%K7OYUr;8$Nr?a4fi_VN{huzmdF-G2OI@TV6-Wd8J{J* zc#5QcByf~=GX~FhcqJ#!FVx@Xs%E-exD*aJ0o}qbhW#lLP5ZoG7KMR$mO|_H^5V3M zn$2bEO-1tv#D}F0Hz{6{dWo7=5sM77mm*u z3ouHZfZIi=0XD?SYG&kyQd^$S8(2!Nc_EK zZntw^O6T-VMw5*=-8F8wOcF4ag1-23McBlv;~4RdU5RzvPND&5Hev2H5lH^s|IaLg zKjMf#;*i6~Faw{}>`$WBQ4~M{rTQsTj**TYEX$M1GqB93TsY5XI&r@hVOS|!_DKnH zVX*`UhA0(NYJ3e-*j1*@b*(CrmQohT^J>1!(`mhy-c{XuSX}NEms&%Zh+M z_aU@r-F;mwbcvv%n29UJP=h44yebmt2gMcr3-IJ6HQrON*qB)p27}IST20(41X|u9eQ-q<}H8MxGgZc*~iFJJ$T3dWN3Gn{*l(hWjb;{ zQz61t-!y+|)#UKraVB9C09|IY?AKNWi zCEuQ{;q9FcOD)em> zV$EAu}>i9>EIy@pP98@v#zj(^6$bS}rbcmkV4-ZSQ1Ohu6;S zmK-{*T4fXx!L-pKvlVBVvVugoo#6RJQxWyIzO-X1_LzW zAE4;8y*fcC>oaN6dvD@aX>?E5Fu!GAV|HAsRIWS8!_IAy#p@i{<9xxQ6ewwKXu7oX zYgxDLuo&?u*WF6Ea)voyKALqg;<33?im`n&6<5XRT8-_q$(!tfY+E4UQ-YlQ0DDv} z_>5J$@v4@Zg(i)|`Y96L?IzbbZr@x5Z|^XV{xZM6Z5&gdt9pq@NPTFCsFg2FSY@Fz zr=cEdGUJPyajoF6%8_zuKcAZMaSc;dN#dcQf0zCB^=vI-x$~}c8AIEy1UkKVX#08b z=X;D4=E~5-s^EQ=1hZDJJmduC^u875lGDXfb8=cNrNQz-qh)VE%ca!H^}rf^Iuebn zJBi$@e0BTtbx`2`1#26*NqWv7$>Ejs*~Cwt!tp(MhVl)V;P$sKKVb36_RKw(`17~E zpaW~7Ua#P$-ltKCTg#dpx1>_i@uUnSbJ9Z~Tv;Pv33a#Jxgu84_2dc+@P;HIIV7nX zMP=s|`U3`CL~4CQLljqQfR2vc;exvh3YtJsk;U5nIzP|C!biP`wGIcnOXwI*0?RIJ zjqpu4Z+c1-qrRy`En!C~c{Bt)oh-k81?%H#8E{1{GpBTYGL z<(HE`Y1#Ef;Z;+HjpUTA8QvZc{=w__B#FJ;)6guNDytE6zp9<9aSVTEIZs?gQ`APB zmi0^w?{yt@Wesu13)oC}j6NBQqF4N^0Z2ksa@DzNlXafoUUyh66_{=g+Ht20(rAw+ zR#^m#yDWn`M+MzJHFRlWA-V8~oR}@FFEkKY57ts2@X{$N`f?<@;}_yHho|2ldkbSU zah_d@aXFV1RvVIC+}~a;1*vNo7yQJwb#`|v*k3SXhb_0zXrerQizhKD8|OAflO?tF(t)#p2VS|HPYZUQRpYnJ1N6K?W~vQ zCJNmv9w1dk@fFAi>dAu4WY5F6fZ;qrb%yzs2&{C!;b;DMt-<)=-Arxk<&9EcMzga5b6!wmiWi5_X&>~K;n zEGz-po-f%Uw36S**ZA8=1xEyyB4fauo4Itzkn2BG>Ttj>$8n%)!)}_rG?$tlgf{ zez_9BNw4SrtgjwVOPB#@R1&*&N*Rq*Zk z$i3>=cJPl))4#sUYmDk!nxaXA4V%Sp3B5K+zNL-X>U1RjBOCXZ(0@HnFTF?M%>?%T zT6k1d+Q+rnV!DCF{2t2hNA%Yl{(&NJy#bzP z$}4II_}{-Aq5P;y)GBD3Pk%4|VG92KC7;@(Mi*-9BoY3h=B1AxiAl)f|G&A4|Mz43 zmjr`SPx$jTKZUY-|9-n@C~%s<9r^yc|2LmW#0?MZ1VJ&!0M75{_R9cx7-rc5B7Z#I zfBy1F;L&WR-#Xd;evcQ34%jmBCIL2{-;4jjL;ZrnCks#d^tXslo&`QyJ2)g|{{LW? z|675*&>vL@z80Ge{kJ?`AaY=_r;dz`+`J318q{l$O$ z=zo3bHAW1`vr{e2CjMIc`4;p7aSOv4wT*_67x;rxj68mxS!2F zIE#|gi^w%{(T$xjBfGLN@4BvP(U|jG%hFjYEc%qz8ZhlfV`AcQU=jCo%`wWKs&z9; zAP@|21yo;>mv1&%)w+@U3$(7T7d6-#W`}D-CC|)9dZ#4N`O8jjLyZ6`CQLtCt6uWQ z>Fv-U_WszMi5Kc^cLs;saAmEV$MPR}8&mtf{QPD7Y|#=fL%q3#?53A!<7N2?m?qx2 z$Ho;kwg3Grp{%SCo)TSr=uUl9>xmb572Fw?nq`Yc7nvhsVFF|-9I?Ohc3<2BJs$!B ztw2g6a3&r^*P@$XSXTP_;ZXIfn>c zzj(umk=DX9xUpO&Ct*2y2rEG8SSeXwB9Ev+s5L)=tJ6iLBI55cqu1Uct4#fKPC8f_R<6kyt6IW z^nGxpS+>j;`&5<3B~xJLm2Z6Vm$%whGQKVux+xK^l_4N%*KBm>N+}q+;d$B7G<3o( z%;V$;1vl2urt~Xi(@sM#-~&T)93dN7>E@@ZoaI7s0iW6wKCkK8^hbHs+fLul6JVTv z_AOtb%)FfGJ&gHCXz_5LBW!mx#0#QUkxt-ZA(DeNrM^Oj^Y!P+dX4ijB2I#$;Gkw0D56YTA77e_(N z`M6im{m7p!-3sn^BJx&Na7uY6RPNKq(eI!Zhiq0IQxYd>cKD44t1{svnYmiYJ7f2& zv9%RHq{Nq;LmRTfT2*%yTDiXq%AaD)ThRF*~YUOP&j|T z;d%Qv8plTj277dvkE*!0I zVz8HmV|P5F#B6$8VV!w>CU=^RISC%~n+@a6;zxMhoWhVA1iL#b3&tp{fxkE^J#UqW zP$}{as2-J<-D%K?+j0;sP!P~oCMe_LaYKI9g<+ z(F5qXl_ICEPAUfn4^7-1HZ;PW)80se8KPWjTvtA{JrwC_bPyiAmgjcdbfH*b^CD4N z!c;j&nfbavlRB}|HY>V;pVGC{i?Ji6Vbd=S)bRDyUFznPgwv!UCzv}2ENAe{;Z0O+^S$p#5>s)wH5jw=w)@p7(&ou*>RauDJny}8be;?-t^-^pu!Fe>SFge7E5B=@>27u8 z{aQ@tXly^;PMP(o1H1={m_qqI#q()3?M?o}!)aOiUNZh-+AF zX*{=2bQdI0dkA;i@|hu^JpK^vF3C8#0B-lhX_26Y#mH>oQI)*d?n?u8hozoM)*Y*C z)x()s$K-@&1_9q%$!&GRh4h9$FDpinbY)xW0ijg)5ZNtv-PHK>^hbGH+4eD43-;hf zm-!;Kw|u2K_$&oZoE-{JOc7I^!`MP~@ltOL73?43&7o}A2n;<-A(MHKJ}A{p#8hmu zpiOij#JT$3xXr5Q3n5f!hhNVH5Dc@K184hEM?5y&y=hkqIPqy}gRRt&&J060K z4`?*k5^w^q9kkecOJ~M3Fe6xW4S>1U@uvPXCj&J|RoRE-~We<8r7X+ufZuo1X}Gdel|vccMV@D;ERh z#LsS_As&*5uqkm$E|3OLvVn)b7FYxqq{d#2ZGN*!jL(Hj2_b4%xXQ3Nb8&X0QXLv! zxp!WU434yuNJ?#NP<mJuJwa4vrUjc{o>IdgTfxhjgoLoq}paKU$=dqSN!GbSC!l)n8xGDMh@3n=zp*}bpDxr9c66!utQ;-n6rYnTy}swL zE*(-Gi)ttRSsQDzi(19NCGQ(`19eAP1r!0*3%vjJby6*+R24y+JaKWKV(m{{X2_4; zuYNzbrpVjI6|2WFR%e(cUe1tqYHw-9n?fArJXCPj5AJ@cr1yFQQ-xKNNZ8qyhA zlH;(l#U?wOQ>o(WBM+R`hJXeD{w1xG_8;Efl|!eX5^Qe|zR|d+QG|TM9d~{YyN;e( zk)@`FDE@x4&S*4J7CzKT5)5A}hjYxnK;9^Fc@K0g)5M@OO66~2?iMUAw$nyd;vs@{ zu}Fl^+u597)wer-Hn!H>33k=~y2$S~=PDf#>?($STtP|KSk(ip3R(|NZhp3$3kEl2 z6q}Sy%x7*@ldo+i`Xx;E6k_&H!h1}kaT`Pru=D}y56%vfhf%wlVL~d0Iv8j)aQYU_ zg%YbAH!Kv!R|T4R$Q(D&^5j7BtUFIT+n_fWtNy05SA|?{i-{2Hw5wK*>`CZ1>{ONZ zo6GNq`3d?qLle3yj#0M2vd{mabaoIx#?yJM5V=Fme?f*WJm9 z9P=hdo?3;%vQnbxH5w`^Wn3!8EFQOuhBT`yl(r{@cG)#$@brAr1h;DQPP++}VzafC zbPI}@xNa9iRV**Y%d5+qlB;&5)U%y(9)8-tTK0+UpCK-=b6zRdcNgt%! z9Y@FX`T5?rUPjp!uTqUg)QbK>RRii&i?b!MTDeAh8wzi=~AWOId(J3W4+&07=%+<-P_kyxn zwTh`Vih6c>n!oOyUb<&`;0GV(X&dcMjanKG-bRzlY^6p&bf4DS-1%Q7b0A<)tBKBe z_Q+`w5|(M2FII@$&%QStz*AJxV3jamyca~r!H}Y(i{Gl4Vb+gZoAViSovXDCkWOZC zi`DWh#BA^E9No6X+P*w~)AlepTF;}Um}9_9b)ceXPNID{*C6qb#3sxNV>x`cgyMXM zfJG%L9E6?*zuu!(K$C+k?kDnTsM}L!G>;LhYn&M&hDIwyt5`+-o)|ltb+qru@Mh{b z+U|VEzLvmAT`}V{EkhVVA(SeHB5}i`aN5_Y;NEuSohIt+b8fiE&CjAC!!0jP6`UC^ zN)3Cn_boo+V_}gW@^8{>pl;&O@Tj-9+q(%b@NV9f3tc3$8RWJPCN!+_HCQX!G3`$~ ziVmC@wkjw-YqHtrxU;?Gy>j1Jj+9Wo-`H`gR6bI;>7f1_$C5?=n1W57cTE40f=y8Z zDOmCs`k!@W-tv#b6}ED*Z3efc6*>`Jy$tnc)lYq9@r7iAX5*ylb6XRg`&cwQKY%{n z+Nje#C@4~#bm_d+i@&?!R{Oc{rQ!J{;Yu(sxUE(-BfR1w(uu8lFu<9%=-{qZYKFFj z?25Z==t#zG`_$MUC8Cm%x1iosp7Jp{@>4$?t@3={y?*M7kzTsO@9eI?Pf3qD)=e)U zPA;@1y7;p4Bpqt1=7Cb#Td-><#3fX0X$lh*T^x`Xzr*OHljd~E6LQJDm>;j;`~7v8 z33pRvL1Lmy)z7`m{mxPC{(8DrWwTOe$yYYLQF4oZ1i|F@0l~Xho67+PyE7BewD+8^ zm8$$b_DmTrmGdW6ao9#J`o{BF0-tE7~ml`=9s>n?pIFZgeBh)tbUx~ASL<0Gc>RKkk3A|Q|I9h6$ ziF4^ch>bL`r_^J)7;`yKQ1+4UzM_+l@qMe}fYsc|77T@!q^@VaB#Fqk)OdARCZlzAgLdwBwr}kUB*cRBX&fxPIp=I#w-V!l_mKaF^HV$g1~akhLoKb+B!k z=HjQyK|=4?pZ&I2bK2 z^j5!~dvv@AlswxVSMqmLi~}wz9xhcfUEjk!Hyy3ScO9czHHAS{GaB822}W+QoIAlP zygx_Q)xLXvSSHnrd=?yfy4bHR8h}j5I^cM(&AAeAuy2f~h|TViFt>JRU^!Z6$y{!1 zi^OYx$oTnQ$3Kc>XtYU%HF%c$HA3bq1;f$(S;zi~NzX#E{izBWx+-+}rK5qP5{16X zRg~E{d|n3i0R%jJ*!se^2sJkA?_4`lc-U4BQcx7v?uu0FyjaX;I-;}iUAcG^Wn+Js53IfS%%;eBW*rIzKYQy3gZM0Hzb2^jx1+tyO zn;`}KYJroSeYFJN<^x{1?0X-PJUX&syq8?>Tl_R4C7IPb(h7BD_D3TpVM&;nTn}VO zrfgbQHd4#)hOq7A@jbNaHGByW(|;CeJABB$tp~dFbl121N#0FdXF}JrP3H3M-d#YN zVDlP*c^lc;urtkE~*O5=p3_5uS+HDIZ_T z?zGK0+;SPk4?XWrnAAICK}z7W&K!5E*@ZL^%lG!>}@y;poq=NbqmQzp>!KGyD4(IAVAXYbZ94RAJP4ud^SYP-nnT#N+zI z(wV;!&vQR^ouH0OZfUPE=K`VVsqG(%ylBu=WDPEk6VlIRS*7o_59H?KwcXcLyNx4$ z-JYJxDs1&Fbb=5)CPQc)ZM@(GxTCmQDw4t9BS5&~9TM_8OwbG!m2BP<)gN$3)*nLD zYe-b{^t@SiW<*9rt90#W?RKp=xOhvL+O3Ubbes zj&>_84W7XdAXz~C@I1u{M}?a@cG7PEwxcUx!WqX?izSSPzedQ9e4K_`QoBh53H8J_ z5@YC+yM`I1^FyWJ(~5Yvit#S>knph3{??GNkcPD|nCLg6!i<$K!N5@pY5@XT0Wz*e zlg1>Upznij(hMDRb3rT~7ETXMh{u_3&IOCez-A&RZ^2OOBh(D%d#G;LS6kYw3s|@j z&Qd_Af-+di_JeUI*BmkEX(*{K6)93;>V;!ccQdX>FW%^UM5)nr6YHV!j+nWK&P{

PO2(K}_Df!XL`9bdLVO_lXg<<}*noiLiUR*E`!0%kpe?Bi zRBwklCbqd#LjAdH!a8T7Sp1F(tER{JzF7y!{Bmn4QrGUFe))#wWSEr0X^U{Ws#+VB z>K#>#(X4vt+wVQ(DILNyY3TH8vJ$j%ft|vN>(h>x-x;dUn{T9YyAv(3t5#o`^95E> zlTz(n1_s=cl9KA5rvoPJO+W#U#e9n%=Y9yIw>;37OrHOxr1z{lQ?j7^5S+SDsmls* z(cug}BUm1>vrCBdXnl#Qg1FL=9{(-!4H(rE>DkMRjF1)|b(;fz8`zZq7T`Xh&E3xBbU?PipcXni zRmG3+Z@ki<$XyiHAXiHYv$q7QJ2O#)`p6am%^kbD{*A7f#rS|cMPGk8k86jiS@61QxW(>C{4 z`3+k9pXU5qJb>mIy!-sygcCjSdJaUTXf*QIAV9qqJs!%An^TLl-9BMx)5c}U-1s|n z)wOR)oM6w6^-6Iksu!t&lG4z5-*Gvvx?d5*QbFfx2_=w)jC$n#$QMLkfY*@PklmdE z2*2E~@=;W{(JnXZjjJ@=SJMCJpkw7)~xNF7QZJ->>qk~Kbk?S;&L%pqidU&U1I(5sB z+50?t93RvUe6D2CqDWfZ!Rb{#Pb5m`g^3UTnXBIh=Z8x=X8}+d%`0KfU_6lR%kZ)B zc~cwY==g?|BF0H}C;BD9ScPhfq2mfbrO<^Up zR%rGGbj@+mLqW05*zB0L0{KJ+hqteQB&a>Ktz{`fc8Al*SED~h-bArbuP|Su2vOYn z^OtH*!jof$+Z1#H0%F1;WJGt-f3tT$lnO1I;V|5$4^tG-(Ynq!lJ9aOYYhj1KZUE| z3g0PSqI7DJ>J~h2ZXo?Jn9pf^>5op7lj;o4f?rOHKel&(xK<;pu)Q{y0pJqg)1#Z2!`g zo>r??Ek%CR@$IPVVSOGqcR&sycTk2F%rm_Yft?0T1e$$R$4BSOCLBP!5o=@kMZ$Q2 zTs%jmrEYx+-YqVpwf|+Kv*ZjzI-+=mEBkH+&Wp$qIu|m+Z`c2T& z?H+Q>@S1=j37y)2bQCKU2z)8OydiKEbFZZqcrQ zsGv5}j7HqA=T91)U2c~aH^MJYpJExUjIrBjRzNUvZ38Yk+z%EUzj|ozXA%+?>}dq zQSQT}r8=64j01!Cb|Wf~f5yD=#HkmFdzh}c3*N%}d5L-FJSAH}Q4-)#fef8dmlMZj znfj2yw%#3&qFAz?1c>DRp*?oL27bIwkiKUJ`S#~Eq}fB66|ib#c6~DEJ>*=H?YPF& zZ3To{-t{)`x$N>6^JsN277UJB_;kTPgh_rVLB0sAztDPcc{nXa!&FJdUKqF*Jem2@ z2He6W_YfOPZD&mQbu!@FVWMR4Vc3v;k}Yo8GUX;nHiQ;>iGXr}#$h!88!Jt_lHJH(2w5?=g2tVHexEOa!rqS756c==IxG4DYp=j}S z7SKD=dv|t|jR5qOCLPQpWOzN*T`v{Wg`ud*PKeOBJPQDFPJn7WDmOelwx7sv^QNeD ze7>a#T;bt!ZBO2mk?mP6fcxBZ>5BVKmdE#;XBUZxk{2wx7vXj*iwdRkF%Er?+_?VB3)pEW zz$pG6(GaOFdD!fRchJJSbOPzkAdy7}X2b=fn^1L+U|b_v{XQ_Llx zl9O;2iu3kat+5hof*Qhe-9yjivwmdO^}Urw;XJ*kDiu9z=g!Z;N!&=1(!Y8nbMhuB zNH`XH)K1&=g`M=QaJ6UL)`;)Q*BhM!gkoP?#x2s=<<`!kfy4*cmY~*pdwHv?lW9Xv z9Cma`7Tn)6UDHd@p`cJ|EI*+R1v%7ORk{||iJ3Aapfr)d#({LH5q+weRrJae`EIC{ z34~56AYp%l4f$kmxglka>y$7A8PXuLuG^!dl4CCT!8n{oV|m*iVtO_R7O~+aoah!N zDKj;b5sIO-@J#+50mVbw-%C*J`6&w>ok&Ah`M!6Pxx7#~>o=}i)HYQgYpfr-7S=F^ z2C~t^qv*my1Gx`uV_7Ua#t(Ym2g1`tg|TDLl!v%-|01L-<@yY zgB@ak8+3`9xStZa3c}9>$=|s;0^Ag-!@!xl3bSF*f?H-U#D6h^em$(xl;#9>9c(kd^T4tF3~A7`>rC;+P$fnbi3IH zG zzXef!rr<@yol;MR94q1G-Yn0GDHn#~PJLi-yeHwXyCh_v*yzKizc19ecCD_^%#{sn zl#kdcMr_HY5}HXHJ0f<^J2~+gFjh5?U6cQm1UC!)xt2JDbcX^gTst9hlK~AKZP~rTj zdS&;?b?_$`k~|Z@0x~`%RIefSHr0p$#x(o3v=EMrj%GDMv0F5+QF0yX^N$iakUTVN z1r*n2N<_knYs$#psuVuhQs32DWCA@Kay3Fj(w464J@&OER1(&>I7A)%TBFt|Bmz=~ z11XuVP#;QQm~3?hSrxzm=;2E*E=CT0JEKjvO4!;VosByQQ@G$0TX-2$c`!wLV|vIx zfDJp725MAGI)1|Nw_MJj`~ZayH)80;ANB3K$NIKQQ9~v_rAMyGR_?1aijU$vAf{YP zM=JR?m$dib??$Zz`FD4@X;s1Ff}=bkrk28aAC6@`%a-Hxl-}**qC@S$8x^+Q0p*N_ zF4yQum93-37@!r{F{m^HGMegCo1PrC-21hlT+Ap=b+Q?@THk4(j+p@}3UD!c)^$?# z$#a%7%U~#BP{}rBDWvYC}>fL^QL!wXwH5%8X3fSD;9!`_2I^6W78CE zI)ETIJe;26;#DYy(mV^nM_+<*hF$GwNEoyLK)P$lfDnO2bigqZUYxmTPnvWlagj71 z%0(Sd>xW01Oz)Ya=uC-gZ?Y%c?$)0_p7$;KenvI13VmR`(S%p*MS@JAt4cr)l+N@p zU3;Oj5!T*wQH0K6VoHLRpHig5>%?_|pRx$O9V7E>f2am-hh_Fi_NMJ&J!i5fUb%S; zTBWb`0z7A18(sM}aK>Q`=-EG?ky^1D{Uq+m`_{_s_L9N= zU_ymJ9cYdj?krTt4v!X;c83j*7%vGyXczK&b_MXA-mWwyfGDHSA?Y*bc<~c2p93I| z=)!^)%l;=^|EuzO5k&w;&!2S%dU@_=*T8Y&vWo=T@j%AfUrzj@EKCNAs~DivJcb2$ zh~ldGTICLR2G{O)WRww~!S78CVq^24;KI(61&M1{R?@e94#Uz-5Tw(#ziu;QGzU{) z5RJS9cZ6aKX=yo&;=Ytmr?}^`1Ca6q1;3m#%P)gK5zchC{z7f0RBF}#d@B<KYT zI5-oSyx7A>T)g&V@Tl26`n7Z-E}_{1FZw(?hX~hz5@^Q9xv#ZO-A9&4cP`Au zO^(K)H0~CY1@nBkEauUPrZ54#xAneAcw7*N#5L3618v#(5&tKSW@a?B8nsau`@B(G z4%f2TwXoLXAB(iIEfcHu(=hT8y_EKAj0WYJLcz`wkG_5cR-bksruT2iDNmkF0NmDvI^L7==~( zD_vKjJ12$2KSpj~2v>G+0+}%gN-)(%c)q8M-R}XXwZ<^}oD?5H+`aCAj&{x>7p+3_ zxnmGF;=yN*$>PnqG#^9KF0~iY`6P-H^1i**s_-{kP6PZ)l}08;j$26K5tuS6cpw%15obRPkHH84(8+8#xHN{Zi|xc(hYS~i9-#``ST+8 zPlpp@q^hLE_$dv;Vf3S?c5kKl&nT{qDU;hccr;bII1!lN@K;2u5DwGH;S8iqWQ0A^ zI4Y|HY5DWkLSRHX2?)&=omq`1o0d2pWR8b?0gv;qV!=<95@o9NFw~pK0-0 znl=7uaKO(u0!>sX8_~UfJRb90R(bPc{;zRzV=g_==neV-C+|5(H+%fc?4}KK41Mt- zMjV@p6hLdiv1uRx*x*ielnv?}I&AItd{_?B*Qg1ZjhJxWo_`w{&N$5wNJvEyBcUh) zFdlyuIEo{ROC^8so)h*%ci$armwlgW5(9rG+<8HkFG%P=9K^i4wN>&49T(TIl+Kie zb!RReKZg3IwA;q8kIMyJnEM7PGCj_w7?e(+d2e`X5yMNkSpJIi4WftJcxDqfRdlpJ^5S7^8f*@LjXzur_>SYSr^v}qff|g#f zN?cTNYy^Vu?Q#iq+ti`n{Mdz(p%20CMwtF0sQ zlP;SGot)Oto=UYI7VVU&K2#)(6`co^5tl5XDM^WGQl^*~chVyGi>a>#3H-a@b~X0GG7u;>{YI|8k<6X2mS|3w2+rDbV6 zm@7-gq*bNdtbh_3hQOH1?gUHk=Rw#Anf?}MBn1vSdxpRH9~NRH^fAicl%K@~1qFp= zN3Ld%By;-re7v|_-^MCDSxx0XyGw(J~dfW$GFxVH56t6A_~gO{7;V)I#pK7{AM zog2;iDpeVuAIR2h{bJfxN+k#m8EgM#G-Go{TsV&%W_QG~*) zC`*<1<=alN`##fo!fRD43ZwkojtJp;vI#}m9<_@1t`JthmZQVUR41Ur4naVDpB5cx zK~t0?PI83?{aURRGpy(D%iz`VUT$&CA1&DfE}Ov{o{Gtd88&o3!Ult^4(BuEe01 zP3^zQw`9BHeQ%a+R}F~pIc$;~4(3Hrb^He_^3^dD*p2O?mffV?fp;1O)stv0h7S;4i4tSLYSG2$^9{BHU-JNJHU(Jg#^ zAMuDY`@Ky^Hh{ELgkG4y?(F9^g^TqQ%L@)#Q8NR$C;PvJhldY%Xr;bDN6+uLBVV_; zOFO^2MmZrga)$>;rAYeuHH!7JOlHNmm|@W?$2^DK_wCbrL#h&H?y8j-X2Ahz%;A2EF?mivM3&f0Y;zJ8Q>}{n>iG^pDt?p&go?Hp<(OyQ@m0#qa9vYrdG5?6@(H z_Pfw(4`Ul>%hEo`j8*Fi6_V?Iyk$~V@Wqfb1yU**8`3DE;;?i7F>Al^i|^&$Qbz#> za>*{+A}N5C-bM1rH%s5^vZ`!pa2LJBhB7ZP!d%#>obJTg>Qf!XIySa!wzB0T+t(e9 zfJGU0qv0Ov)>~naLf4Cf$@On70Gcpc?Zrx}gkLuN2GBNtWtj%SEZjX*Lw+kKa<`Z_ z-gNU&UsJgC_I+Fv)THO_Sc3^~fcb7^u!H9t;bI{Fb*EhR_mt6wR`PRr6sATu$@*a! zg_b(OnR8roFhWYbBzpM@+~FfbzfGqdzWLx!B38Nu-TMp8yZ=&Lk;Lbi3`SFS7klu_ z8%cP25M29j{s5st`Q!`RmgP{_55W6!*nZkS;Cm$w-^4$x{U?AROZ&P92;v{m_=HDs z$Wa0>w_pui3f8*dlN<&7jc=31a0&a4@HBw9jYrgqC8a;o#DFtROx~ud**8C z`=v@q6Lm?4{I{X`dUN56y%`ow8!OZEo*l=l8d4?!cMYGxy*18(dCU;JHjK7 zkoTE|Qy;pK|2@1j!K2Av6{rUcfNU@hc0~&*hcBZ>(zqF)peLKKQ_Ex%9-!C3&E$zh zl37!uGR+|)xjy&2^qk|tSammJ@}*-UKu6#N&=WFVQNv{ga2Fi+39q&ghOfN+_@2x7 zHB;Ai6n-om8N8VBSDq=+kjsRVr{BWj%NE&PW<+bxP%)yjGT^}}bP6Tz!+?a^=l%*W zdwlK1>5=9qukP*86-F1?45wb!&s|;vTQ}7ioJey+D%I6#RDX<`3sWCxtV)|o0nC5l zuP@lGr)pa2OD7Ahv`aMOW_ewXV(+U?>@~w-knBdt7lRsbBJQ1z_x+mD?|Qu@Jgk%L zSA+tdIwLMe*W3w=+Vi{C`|XIugjSH|RBz3@(zN<_RQYJj@Z$rD+U0RT+{JwT(%jx> z1X&truM6?kY!E+WHHbm_{6_ZDODsJkf7ZBPVWQE&C$SVzw3B`pev|AN9vxlC!2kSb ziP?3QigI@6(Iz9C9{zz@7w#Ic^y7vC|2JN%WH8{32Rge8{s_;_^?}CzDThcaH0j=O z5P^WnrVfwC%u8t1c~aKLQ_ad--tX_rIS>q+u3xG%NR`-q|0*K4_(TAj|9;vYm(z`^ z3ArN5_ZCeAznmevE2^*j+Bto{Q+ei#=sRX&mO)~=#_z_3-Z;!yDb21Otxph+H0cmrch>A`)i`x3i- zUcQO!hP`ZCK37oA^*Do8p^m_+$o~i7mlrIN59K|TbJ6)5Kd{^M z;mee!-KfR9qY$^9Zl30gRQ#9<@bIo!CEHye*MthTo7550aH7 zYXJ2L_qHuLV$h-agsv+6+EK~DnvC+MY=NpkQ!R34FYF=@<&`@3Bid-iZncf&&<$)L03KpYicOHO!>4;FQynal z3s$c#Gjp0LfR4ol;=Zkn;o|9>vJs<%J08DJIqB>Ip4W@ zm%yMJi$nR_dZ|O?p_&B;zfQ;#z{1Xtwb05Ooo~$J0mx09Zg1a+w_jthj#!Ey_XV}Q zz;>->2i*CCb^IR z0IAd#Q+TWcH9Hk9wtS+^CRC4FTo^;2V-VY*9z@{LJmDkz$x32?8pMwpN7@ZEGH9aN z%zI!c_vgTM0H9Z^;goa!dKO~uAY&+N6*{g7SSKgXi)r=kJw~3QZcdNs`3x+>!5uP{ zfdk`{5JSA}J0{dj#7qoMFGXn&#_5oVqU*{p5cmaB+zM zcEpUzi1<$%p$6E9crWwOG|b1DO|EV+TRr+g`uAIE+<;5GDB2%l9N750dL3Izn-c`OE!;l|3bp-csN`mlUnFE8gMtai4--P?tQa}V zEfQVd``8Q<8-3R0+w(i6yF)j3RKy|`61!^K(qDIro2R^!Ml0Xq_lF=kjy}_3IVh;4J%1bbc4_46cZ5A|CZjn}~v3^9D zbxuR&JY3a8$ame^6?XD&^041}axVR??7i^2wZ{|=WGaAc*<+lIw`;m+pOcfor9gkM zvm068@>kmE&)fdm=T5AQ78=xrCrjQ-__AwiZ+`^pIE6+try?nzY-RNmFoS#NHVQCA z9vb*PZ$}6a+dxwjn^wp<#2nzH5bd*#{qw5@y?D-~@HE)3gV<=Xa&CBQO;b|vWpv!15tV}p zX$UV8P`I%M6MC&fdIlB2v%_u0)0RYigUWp397OJtGIz*t9)Dk)D(Emr64xcLde-`w z-m9pqlzW&4rMM%igU4atyxZ6l@@e!r_G0fAO+eGAnIv1?7r>5kz3p%q!*r23I-oJ++oHIXEXg@rD=bFgi+QP~E)AKwG{ zcul~eMD(2q6&aoKbG#Op9Z_$waE)@kT(l=A)&kdA)F($rw;qig5z2YCkMUTtk_LF; zO~*K4o>+2`V)p=DcEBnlR3EiMr&S)az>}6%zz%7xH5qfACoVE`WDb~sa2NrTN`^kT z4a~-4;G4gsrBmR=i0xlJr&Pn^>l@I+m!T3lZ(=3wIngaZ3YEt^pd-yj5;au6fC2 z?pqv2-3>qwt#NcGfn_rxl;DN*$(?v)szsT!PNlq)#qU+GH7<;ndtYFaK(UNwhh&7X zx-JY@+96w!3+E{mN`{_LrqMp~wyjuIB(PF|Ik2*B)?gpW=M;KMVKtri?J=AR7|!{& zQhb?XUq6RqC#Oo2t#C&CFp%9gY^ZzqD=+G6zX}sGF7Lob7ispC%NP>(@bs^ayImiZBHxwhFZ)YAp_`04?`Kai z0Go;bGv2sC!G-%dHwF8_!jYI*$55Zu1&lfetXg;g=Fq~2T$IbQ-StuW!u85KfG#zGv1PAsr5Lu=eBvhFFTg>>s9vD8p&!B5 z>-Kr!vy!HHdQsUY8T>^aM)vY=+RmyK`!(KqZqq?f%*wIm)Nz##4YliR3B$ER?(NAh zbMIR4@n}V`EkmwwXWy~gU09dG`*rqOuaNCv3W919-%8Mb?C!*6Xw zZmIN%L>`H?T0bOzZGA6ACm6PNnwL;MJ_7}UvMTj$j&2ikG0cxYbbnW0sBu!%NK45d z&s=AHA#LZBq-M-8>GUc_QI5CI4OjhGYpSwnAmD@%z<#&qWJb6Y8H!I5j^6;R?ebJbGf-0 zLHuyU*o&k4^_5an+wP9!6m}PTxK*#xxS@$(e1uqMmRPyfGFFy%2$ZA-v}!ME&8i>u#dHdXUkYYqf;L)31x4dEksYb0#0c;d1j{G^8inl>EF$;qhSmgv)M-Gmz;g`}tefgaaX zUP;y`EY9Y+xxgb=uXmVW?_jPD%#^jY7}=&G^t_T(+e`o?%oRu5J<01WK#Q>r$)kBe z$mJRD{T(EFxc30{S>z+XftBiCl5hoHUq#1<@6A-mXEsYaK9tN>@>9MgRb;tDI^JB>}chLm$SKf+4E4t!8m6Y04|GVg-!|217`MM^Ov@yL==&FYfP6^W0|IjDpmbh?>pAv zWC6Y5u+9->Kun=3ZA;N$NsB#V&(-LxfsjlQ+Nv{Cm+bXU{zv`9yK%^$$ z)pVI%RM4HBkC{Y_n{@INrXxlAf$tnX`Q{Dda-=Wd!;?_5b2 z15UH%`LFtd%Fm%7H8dzILnDXXy&(ZRu0_$@SW^ve>MwYi;6{W(d8WW8htl%FFw=z zH_?|__;EbauI%fZR+1ka<3_R$wT_vIw8{e*5}@#@_ELZ*%+?fH`TKWCnDTATR&2*Y9=d&Fwj0HSfan!OF;U!&0eI2YvTE5eFM* zMaJ^rLGHwD!RuOE<@V$3w;LW-AhE6PR~_@7_k>W`YH6y7q$~})D~62@%#6N=IpuK! z6Xd8DcB6mSq5wL;U~dkgz>YWCdZv9CKI-~*$>XghO94QNXnOKlMk(^kgvF#UGorJZ zBlYJui!(%)laEBQwCFA0_kln`CzUg!+Esy2VlHVSvd_$R=MTV33C&78ol?X#Y}Zt$ zB6Im~3XxYr?Yp@cr9~UJgV@GULM~Z+4x7T8oSG=)$m5oO3PqBYA{N`C!nQ)LJiOq~ zuxVqDTQM28ptBYHt1IvWEV>Vp*L@4@HrE{NMd}iu!B2X`BT_WCYzEt(aTS+e_T({( zl&h7nxM4Eay8fn_ueQw8;d>BI+^E*^k_V-yfLwQw;Wb8ROHFn%dWaRsA6reGl)C!T zADHy5!znDmi;J%hg$X8>eZ%P^+v^g%>K5i$(E)43^@zn`B-(Lh6?<*r3^pRyRb@emw-~5HIemeol&OlWm8kOHps6lxSSqgx zIc?I&1E5@9K**g^ctj;wKyRvI15rf{K@FJb+(YUW>{qgO#ORJr@SLX`jtZCU{9GC; zPfv+SU6{S63^dJU@cr@m%Rpr_SlHNdOYUf~@F&kNa)4dwf9|YD@q@pXh`x8B@}9$k z^4ea0C%Bn^6g`iq{;5>AmWJK!46S+la$s`xdNgiQCws?o#h_U3J5TCJI_E8Z^%19j zvPy&+`~Zl}tS-H4KK;hyw%h~3A1IxWQGHqU$|aT7Zg@%91G2*7IyRT*aUm#5c(@{k<9~{uEFI9 z{7^$55Yq+i8`ulXdw>5UD*LbQQ6!3Y0Oeo#gT^1UsF;ceqdc1=D3;DAn78C`h~_4H zT@}cWh(}ejpVo>icEzV#^hua=!bGg>FpSu{!%i^qFsNs+gzDMpr|b`_EBan@J-g^% zbeIojdFvbtSGr~s0>YFGGwC?2&A|uNMRezo3o2x^@In@`#n)bMe|zqkwdMikp_)@n zL@JIb8hKo+_Lpza!vhexd%c^?yZK~9d-GmjVG^a)SDBY+VG-U0Ht`IV-!QKy)3Rcp z+2t>Cqi$Iph#HACt;f!72G12#HPT{Z8aDq*mVK%`(;#OxG&#g&yWT?KLhxXgj@@>g z+4x^)ht-JHC=;Qq$V!Zm(_5`=yL`)ie(Pm>C68*i(G)6ivQFOd6ve|t>Xax498DXq z+m0!tekl4?+Ax`_GGvJL3SH%BYn!$&BiDWeBCi_1nl&Z;if@y-M|JQA87e5w)~V)v zvF=~;nIr!ppSkY#_Wct}*x_VA%u9PGCvqD`jMEcnVs;7zXG3l`sBmPgPE+tF{k+dD z>2&GG$%683awe!>wlK6PbBvYvya{cjOX!h6oZc{AloPhEg3E#cRA80tgZ{5r%p z7P;H*`&Px=ruecIZUl`AlVEc*4-DpH7S*1I`dWKbwbdHjB4@a=!m_<_upxHXVykGh zP(Gzb*i7s_DP)U>FIe4>{kysQ^eVCRqJ@G z2cVA+zAG{KJVFfu zL150|i7G70ic6!B8R4hdI*cssu&sY)t>f1<$#12mjf5VZTk2xG;g(3fFAnU{uD-He zRus$b-Ai~~Wo=!e4hGnO1Cl!Q8?dL@Y^GO)&5-x14$49cp7wZx69td@gy-8TxM%_0 zL@Ok|(X6TOUnh3})tBNY*5LhtQ!e^v`8-ZL=`(!i7SiRE9LIKsFE?Hdj(kCZ)I7z=AS}?XLyx+iRKRms0A(%-1*Ip&I}xWcZK4OoRY8-C5&Vqa%$;L3k2iWWGMJ07Nnry@K*w zr(Y^?i@mQ1rV7RMG)-s>(F!JVfXA-YR>DOI+_D_${BGs_!5w4Tt!M4n#N*zK9rND4 zzQL?vnO8ZL4@LR@`clpCKIRl2gClFjM_yV!x*(QA)pl$t!2I!o6 z!qg{;?|(?q{kJptM`kaC_Xg-)M$^|x`vW!=I(_g79i&izRK>^ZSM}5NLHL8{F!^Pa zyYnw<@6or2)OW8}9{o$r5BKPXZZAgYsdI&q5Q|djz0jGbSK?+{rK0v`%M9&a4Htf%3Sh)o&2xz z@_$_DFGATw>%)unLRh~2L%#4eAR;flE@8KYRb*zu^CHU7@gtxgn-cL_qMD1Vm>zaB^2n z|8bk(Unl>^xAE6Q{P!=Y0u)B67t+R1{#sik=?_JGc}6~YV4KSS9}n@L-}O^@5Y;Sc z{lsPP*UR3`wt*2@9BfoMVD9~^&cny8y)-iIR9fD zG4CH{I%d)<%0JqJkINqBs#?t)>SOdnTMz-o zO9FpH7v8k2$NmUXNL_^JS0pcW{1KjXk#=2hg@!b>;zMp=`>9@&3#^m6{H?sTHx@pi zgN#Fl8+E7Yy~1E$QUj#7PyXkRc#J2eR6rq8yZrCpdFJnn3pB9_X|kd{7U_J+gn;`0 z{*g67ZV;@T;EnwD(cf>9{E|teSaw?_!lQJyjPyFZNUb~tJ69y335OV77w?_kmZHU3 zF7E~XOn~xr=WJ&DMv(4ekgGz%xJ0*By6p6n{j70K&GkEbg?N-BLXaFwjsdTTvB&o-P8+T& z?j6h>*pkaq$B@LpL+OO8T&MQiUBE@^$D<`Yr9!o!Rvc^?O{1Ajgr;ce?Ca?wamOCP zaPy>b5Mv>*Kag9(0Z8O2z-qU!rd7)?*S^Q$AF^E0ZySX!HV8Obn6Kq6fAzZV?nz*? zp2th6{oWtPEY5S&X^OjJhaZO^e@a}pe84#R>T;c>-byWfWU*57)FnF;i^oc9`%--q zy0BRKE7!)0kpdyuyD4UqPI@c^=iFMp-U8ta8u6U5T?u}kWO6>)qdJvi$Iep`a-t`V zVNTo3_KM1iTiC5kcC8rWV~cR+ZA2ow_d!&q1PR_+EvS^c~-Tb z2))Cx!9o^nM7f{rqi|ZJMX@QUIcIfpJV+vq(MI^CNYm)5W&ZcMf{pr^HJ&EoaXPc* zn|AmiRSJkCyA!e%3Z5#LDeUZ1xO1TKg7&JP_g{Xkv%UK;UuBmI)4AB4E|S|zbQC=b z_*_cQiSa9S6ek%Zy(P#L`G!GTf!9hcgh8;xM1XU)*{T+&qA;8|d2>V13-rDYf{qqv za^YeB-0OD8-eS6l2Mtv1aN067K*bT;0!`M`n_1y>%vPftV;n-rt(1_UJ1!xxH`TV# zmlo%38D%k{0g~?b(1n)*1$ly*8Q>VSuK?2q*L~AcBN7F)EiZx z9PMJR!24&~73}Sd-_wfJZT9`^HSKX>UK|i5z{PnC!hwf_QFP^PIQm+kO={Q5;B;fp z(H0k)VY4re7CM!9?JhDnUIQbwBvlZ0%+Du-XpUp@XxrDAT4rwZ*7#3;zY1-^F=4E> z&HUc#%1>x-Yy|JP%A2WU#J|;$7Bhg!^s~S-nxGf_y=*tlFa>g?@89kmq&((Zj1PK5 z@{fH^!qDm<^_Xq3H^d9E=)dcgX0$e|{dQ97eU#ulszs!R`}8zj_IJ{LW={DYH2$aVEi;+*H&G7y+&%_z)0ks zpS}QHoDgra9TZ#-mSS^}@(;N<9M&z8f*XMKDn8rJHa7;E51Ev)Y1j6>pK3F-#b3)2 z*o0v_j_$n>Tg8|EBB$L@iQV{9@#;p9HBR8t5SQI}AF{%1E2Su!0s0hBsJIv$*^x^z z%luV2WUqFE`F*~o8P9J~^O(l0f{a7k0`byJ z3+nM<~Y^dn@knEj<-A&!|mUk?{O)=IT zNv7$fDXI+7SpE?-I5+yuzW9neKaY~eE=0^cT);$Q`BR@0d`oKeo}ks+yz*f{FaK6g z=KZKoNkoD^YfC^=)V77Hn3cD=r9^>tzQMs=fCBpM29l;wyYjM#D!aY#a3L`2PQ4nF zsyT_Ve*%1enxAR16vcI`ki<;Acb8z|l@P@FQS(|e(See9mj*>ZNZ_1Re;%;HS1hk* z(En!8{n|u3QS(pjHDV+E$0tj>CB}$-A{6wx2Q&*aC0qBKbO+h5PEIXtCVnbq4vu!N z5iPl7$3-7ux-#lBdS9+7w`}eN8jf~;)WQ*U#KEBxOU849BgbJ6DDIBoQ@Y8XO{ke- z@pa=#3;*;uMtkw>Fjd|n!Ki4HCF3f-A2q)w`Z+?4@tsDZr|nN#2Y!=_4@~hb$16MM z!*ker3)*aD^1Rk+3*qAaXROdlu5Z|}0UyIC0gn~7tTvI8dnksDTX3ovL&L_0GOcKF ze}`#4etO|Zf}fo7LG!rG<8Cd;arD>?Jj_Ed>@-WiTjc_#v9*`SA)JY|rPpNRC=sKH~zxR4V7aw!ERxoAD1 zgHcoEa>F_wbdOEoT;}iKHy!4Bf(+bs+=R2NF5k{4hG7$Bton#~FF&a?&)uIs^I?sb zj_8?I7|x3J)+6GggJ0KWb-Q42xVm4Ri8=O{zA$$3+04+M&0UbU*#o6w z?L=0O1$F!KB3ji8rd8j7WleU>lI1h_ipuCNAGwVUx0(CnN^cA+YKZ(j3OIcQ*yO=} zC8@fdWJ5-~Peqka@xKoPDhiB!W6le==^T6W$tfJxABGYc3y*xR6TGiB8b1XV*N$?1 za|V;n&Cyg%*mt;MN&HCed-d@p;j<5Gay0T;nG`jO3_&rk%t3^^lX8^xQs8bW2DJS5 zBU&PrUd=yg@7n^j1ED3~CrJ8<^>1Ccb)lE2j4lkhkDsw`#bqXqH?FBGeuU~dUwi+K z6hjt9!k!vsngx6cs&w(_{ASq!&;0J-iLM++f`^WQlvlXSW1!g6%XSCd9f zWK^nBra7_|LcYA&@#r&=J;e>zX_UiqmtJ@wQn5f--}WI#E*CnSF=~VG<#1#6axNoP ze@!W)oEeyd=SN<~3}|aqNRo`YlV5q_mCxOEErA(CTat}sOB2mZUCzTg@LoE1{_E^&e+SyVr4ZM41mln@O^#82<*L&z}CJ-*#E` z#_#HM#mM}~=W49~^s@-TUE$F8cnRX6(viU*sl{|~x=J?{wTqN&6;0CgB4FtczOgAC z{##_31q7NR#kYi2T`v4WYm~ADC8{N3di<>q`*m$6Z|R-jr41-rEVVePw7Xw)M#-h~ zq+7QfSRa8Qjo0|P_IJsCqZ;A6p*{Nxm zXgn?6qSIkQg8$SHlf50^z1QeP(BB;bOZN69Pib%>YBvUga$LnY;vZ&3Bv0+XW`+M3 zhuXR$2~7<9`sZRhai!TBs)gqyxt zYTkirSzJ#(c&Ba=`SZ+w1Hbn&vzfT)pD*|q9Ce`b^-vf_jD9--RxD-fE_rNXB{eGrGCB40B58-J{Wk3CvOzIeED_fq0>c^w#%7p=B_X+8)%-V+B}%aJ-Es7#-FT{xEFTnk z+@M+(^I0WZN|q<5Xl&O$Lpx`X#L3fXAEm|__PNj5FX}CnMi}{Mt^7RfOUj=T>sl8a8-wkWm_%$4Agu7eoX116dU*5xk7(V=FOX22lE~Z+l zS2p9y?@ZR<7LS$a<_pC4_+}5HKE@cQ{&B%{y7(s1skooxV&kN&Ka;XDr}ij)m)~W; zaCoPQNN!?Nc)hstPD2>etURB-Nx#)fHL2l9IYitW6n{5=iF44T*)F8hHKkvhxFk@b zk{MY8{p9H%d;SW>hvg^b-pl9m9PI;v#f;~B@fhXC-YR#SE6OiH~la8%s3Ql zgQtr`&UUAa_n!Mbd11h1K4W2_)2KxP&GpfjcX(5t%OPhm#z3Ge84FP?(LKXE-(ehb zl~q^T>zg)D9jX5e@@@8YWBs-zgaf7Ebm-E;yO^~1?w}$3-TuPqvu1$auZN&n*+-U`P6B8ziPpwq+1326L2X zCqS1-=N2;ezV~Rtm<04s3E;$)J3y$v<)#|Yun8-}sbt(Ch*@n96e9ROvd*{1c|1Q~ z@O(5xYbw3Hm17yktNnxTE&)=1!a6(lL2qrLCidy(Z3iz~Udr^xSwSnR>7(yzY{9dR zzq+U$W|Y(`U@^LQ&pJ|n&&Nc=s$p$#rw>=g(LPf!DF5;;ZY7$=Rw`%7;GSoD=gHL` zW6tP(Ys4L|2|HXMoqu3vF|gr_e)w#~^*dCrfNKGh5iM}H+8CFfe!&Siu-?YOyAfm4 zYY{ZkQsJRoITZwvaIz8I<9O-Szc;?cS9zD#nlXRk?flak&5cn`_w1yN*6XDcBi;xb z5?P<)X1%jS0k8k-2(oHKGD9zn(-=K9%a}Tw-rdN+hUhgH6pd-qqilLN@RuP=Fp*EC zKABPen=FW@vL$$EFQ#$iOrXYdTHI!sFC~&6?jW>FJ4!x7_Uo#e4?@X&5Ypjnp_z_8 zSDf7_u~}kehnDZlyd8G*zAR_Ajn6-7DcroC+tX{}*YuPx>{;8<9l}vHNM36Pl$Hoi z5wIcj+l)EyOg;NN?4HqGZZWpf${9NZwp$o+k~Udd>^xtlY}1QeW5yk_9B}t;6&+UY@r{{aO*DsRt4z6NorLnhs?3bMh=v{M zb0|jjYcgOO`yTb5QyJ=?!iaS<(^|(0VmULT6=k(+jL0i7>-n_EkPomE8UmZK}e+1nq&F$8X``Ai+s)l294M=7@fyf~NvTM4k5q zrx~rbgjVxqcvf9?vjPk3bk&d?&-Ukrx*1(JY_8al<3{NNyi_K7oe&^^CwCr+lSV9(I72&UnLoR-8TyW{A?4)C z08g(Z9AM*U`*{;bYqdLw@d2*1+vBpiFSgfC7452Udn+0bBRpeyUtLQ~;i{kqpE z<%b4o$>y{6)FG22B)!)idrs~*rM-3b2l{o`qx8!+D+&bk86X^ss`4Msx9Q)$a=Hvv z{XTy)s+re_=$O5k24aiTSK_0Cs6Ykk4kR|!J$P_&Pp_$9cH>q|e|fHyD4&n8Au_??KZT|dIvj$8(2Y|50v@$!bt7_)Wy z9Sh(!k3AXoP{}D{I9d+Hlr|9THuVSu9<39Ho=ye9rbse|Q`=^~82tE>D({q>sr)Mr ziZAU4eMHC>bv3cfNWMC7NC$f%-O?Vzej!+)I%b^jJRAgj86+PegAss!=(R?X()x*T zgDl79`&GssPcACa4F8x1h!qWmTh%i03?6Bo15bWGw#Po026E(|+#70J4YEF+&3CJg zsf5RwBT=mIE)zI(iivOhxJPJUfRt3+5MNVLBlFX_<#^vq2@uS(j3O&Vd1ut6m{#9@ z@c_d}7ScM~YW6}#phR*KyrRj7E%Hgs<5%?oZKiZdo?Dv?*Ulp&!XElSpj_F8bM2HP z|7!syjw$n%5_0j;3jMamT>(R>3et^!MV0Mu91fOI$EoO{ZO%xhZI`s%jH^G4ic;ncX;fG>NA1;3NP_J#&{tV9p=spIo>beZ@Ba>1tljcaR9%d?je_$tQxZP=sCrRwAkmDSs1C6w*1#x0N8%(gm_ESAW( z(tf1i;2@`{ge7 zS7Xqs>pFj861e70%k;QsQ`mkg3nrdpSENRcXnkBZcMNuUVPdP>_ktg9(jk)~$NCEG z(>iy#0p0A&b_#=r47zNMK9u1dE>cbI^C9=vW9no4HIBT z|8H@@vL~n*2A-#x;R}@E!OJ_x+V9;!r0ZU#Ke@=R9%TwOUv-^v$|)&=pgYM#BD#Vh z4iTh*{!Hhw2)$HDal11#YB5jFY{tTC&3hW(^|mJ(C0uxPUPeTFAr*aTDKhIt!Q@Y{k5GL)^xyqHdHJgwE{ zVW?-9OWG|3zciWS=WRl+%K}>jY|9B9EiC856G%|_g{vz^I|uYQpJ&bl>s@==Eenjp znBF*^UelVR(_-pf^`KrGUt2WC+yAP=Sy}S_(&KpzQhEo&q+a(sn|BM2Kh@s!m~v7$ z1>IhewX97ygGX0pU9#6~GWV&!4tQ~+x0V=(u^1CvlkDnISh^IzcilK-+qWVS+E1}2 zk(p%UQ~)l_E<8B0m(lH0i`hE;!B{_(&e7{?e&kv?tZ*C!-CSP5t17EU!+_E4x&}QP z^-vj^v6omQ{5c=z0xJS?$>zCN${06dK8@FNew@ve4BixLY&>L>+pSorpR|!Jo$aC5 zN0~ZIkFL!xj_A3gSzeD#&`q??{m8ViK<_RD8{QGtSXSE?=*-@f>&Uq9MV?%_L>+(Q zSw>osh8poZ7)nt-a(({ep3;qIXT9Zqhn)(UpWs<80PC3qj2k_snkRIU&8xKA^ov*e z+)oAIA!kBNcF7(gHDoLMy?c3~ulvcJZMMB^Oj$L)en?F$qKIZ}n$Tl#Bts=bCyWW@ zSPenwTQSziHV!y#e$^8YDuQ^mhkA1(!SSj6d>j`+jLqlCfh`QQkU#d!RlHJ|GkuDt z__pxO#h5&*(6#)WH+Ot=neW$PfQUb)P3Eczf4w>;4{UkaONmU@7<6x|sP6wti1dBA zZ}|d<7QeQ;$o&<*^oqx7L7cRB-Pdc~ynf|bY)3sj4~BSRW>}S47m2K^dW-k@jb27b ztd994*DXmBPd&`@825O%Vv<5FRKVJH)uULZH069hfiTtqYE%#)tSu}>FcMdFe~n4- zG|xe2)n{g-5r{jm9x^y{@_;Y!CydKGLm$l>_!)?|#%q~85sxqjHF^9nj_1S)m6nNf_k8^+jZGvvGO+hmI31P6URmN z*25({czUP&aWN#hNTRUGX}9Lb&wsl!p>#WFPLaJomxDs(3#A#7b0u&>fU||rY~~nV zlJW#YQpAOKrh^bGj~B?O`CH#}%dVkcap#3kn}6-iA<=}s8NWRyaf<_$>A;;=_(k(E z;Yab-9vbd%pOgzrV?1>hI)! zbxOV6LbPKWy((JmJ1ZmtT^MB`2Zp1MlHUBJvo4q>hC1@?YYH~$eMed6zhj1_xv+X2lwg8O z{aNnk^&|qm9X4CzuK+}2`@80UV6M`E59PHHgMI+Psd2<@zuQE*Hah3jD}+G`DG8k} zyWB83-$lPimFjH6XD`yu3eg%4M0u@a!V*96L!Rmv`G;;bLW^-_sQuG-h)j?DXGgS;?|Oqo&dG^SwxiP?@a9fSz*RWP|&b(K^sL`b{>!`r-J%k2#aWE@>< zFkz<5N8J0WsK@sUF0H=jvB(c&j6j{Gx$Zpq!gGtn@!BYFj)i8m&JZEy`Bk;8f`{_F62ad=hQ zn_3CQytUtt){HoM#&*E7TN70Zm}ieJ_glTZeq3%hM{U8+NJiYV_WIqhO!v9VzDHd= zc@d`UVJFBPO6YYfLuUV|VoOGcGKWHAlqf}KC(6;(Mi|L`0M4!-z=_u54c`Ep@7o8OO}_Q^Tf>&MA^K4H;TwOdV+%u&b%)+ZCLH%;p1|t<6wcuG z36bLSnK*3^r80`ly%x%=dPTgww}F9kjrV=V;Z)B-@aFka*l${1`?{LI8eN?SUzmrD zKyPX+`%R?6;ZW#hD3z5e$M3+calGpY{%=2MuV~@C zGH30^F;#i}g1c^`WO$UDk6%Lzg$wI1Muu*P`BWf-dodKE_)BR!0q@#E;9P$E$PPXZ zJ0?8o+3|1i5Qks6{sx2Y11b2Co>7i5Ln5^kYHSKw<9}R3IVQCvOOr=#)@K9NZycY< zirTq*IT~e(XRHupKkIu|GqakKmqDQU$t;R2xioKPqmEwngxU#vVBX;&U9lntJ+U{Hmt#3an z*eK<&4UE8+G6uo9{kwfJWoUeMt7GilYbOINhvsVAttBc(b^jfRC(VimZ; zLheR->POwsdf4yN!hn?1U`w5&OCT^bc#V6NA%ZPtEXl>@FlE(F%?D=%cT9I%+bZe_ z&+3ASRLVh$u~v<}u~dM@`Zp}+Lqe}Til~Wph65NUpB(YPC<$S!1o`d|Ix8}Bm&Wps zkOB`mL})0LZ&5(5w@!u;7VamijZ~!2!$Zq*`nd zv=q$~1$LLh6;8WUOxO6&!7>$?`MuS> zCM5WI*34zGu}R`xG_@if&^p5JiD^Xl&I`JS9L*U5+fa<6D8wY&C4yHP6%sU21}mQu zba1;YMt}y~mI7*NC?Wg|E9CkLn$!=i=P>L z{#HFKn{fNFs6`wCxy3=-kdv%p#^#Xp)f~?JsmlTms9cg}a&-Y+`jV#^NUm`0`b3YG zs2D>>lHAXk<8%WSuos+p6rSdt@a1$8uu>UPq4ZnfTwTt%zMmZkf-Qfb+;u_T`=sLj79NaA z@v~M4KVX!jOkaYM>&FO8X9l++8)S>ca)9-U%gOM1OQ9gBr;GU51w=0atTcx*zk#aesTI({UezA>~@ag$VIbPwZXc z#{uJ1kYAzY>ud?r{`Gc_=@{wh^OFm4SfUBJ-)Vk_{2Krpx6HTlk1l} z0~xflKL&eq#*l<>M~m2A%;UHaNfM(RP+2;2bQq+$J(AVg1b=bnaod)A(J#7C!&h&a ziZ>N&(wX)2)oDtzWm;GZ98&8%%JK za~Kqi-8aaizQ52qjO$wZuKrUrQkvmLUq`^V1T0VZkP#w|(S`p#Bi#6BMp%}MCr#=; zg*l3vi$<56$!9_{oT*RHoGo(G;3x(;D^?qh&EPzH@v2&cH*&_I%7S` z*rjGD>|QhDsCkMxPJAE}sJMebZitp|Z734x)u15S^LP?#?P)F%1!3`r2s{Io<8iB{ zwwu*==pQx}eowyK>y4x8H8&=Sq(m)YGE+jIFf#p>>3EZwG4YCNEX>QDh(rs2({DYs zGhnWww#uo8{gdgWo^v5v*g5GACfq(K^_yIW!Clm$*3i^s(Rm5uE{y38V``=LZkF?>t!c}9PiKjZMS*%VKCOe+dAQbm^lDT9xeeCGm5 zxjb|+U{$-&nAm5kYm{Jf;{n0Y27EjE7~a}@Vbd-(CdmN9Ewt){`~cH)2Cj>+#=Eaq zh2HfuK`gLZX8^r1sW&@U*vovn{RQca)8-FROPL2Fn zCCFt}++QgncOev|T@nt5rpHWue5@oZ8P)dwWpiq58eCU#FgIO*d2jsM=!@1bWh>Ce zC@kPUa;X`odf^Y|O95>BQ1Pm@Z@2GEesRd(cqP)HpX!G(IiE`#1Z$a4dy5h%n)WiK zpFFtbN@0Ri_!m42xLzAVOWEK@B_R6~asOP(v^`DaskzU(f9E^}hfqHhN#^%>JxAjH zOxvZX&2US;l6l+SnRb&RJUGqQDj-pjj7ZkyWCH zndHe&iuNM=M_37Xmg3Jmh|X&Og-eyv>u15$bA?ke63!c?g}qZjQl{Fa4=EaEmbF*ppa^ zC2i}G$C1c{^sF+3)gJVCW0m##qcS_0O$qKr&x7h}!KZU^j-Z54|85o|lYfwl$L(0A z;PJI(9;xrQyp;R-t;ccv0VpbLT3Zw2lempA}t;$3ds_Tl#H0o=sz_`Z< zgN;F(VplV7?@yX8jFhiUvhJD>aK(@39xjj5!o~aLdPD{?Q8(ND+lRNu;1evjBuTQF z)xN(iuGv2;yhz1GjPS<34*@sz5&-m&-G#S3yL~4)oa^pY&QZQZp80Cd?O1@i`Mf`+ z^ISlO;_{>BWS28uQGV+er@Ojd(ls#X9n;y~D;k}5fp?u&dmtRJ9gV3F*)<(|)BvWk zP1bEh46rLIgbnfkFUP6N0)TSmPKxv0T00ddJUrtamG|w;*&gR?x|MHg$)=tX0>@%1 zXm;x9NScaP#p3}8eo+Ij%`bH$RbIzBnLK=qn<$mp<5j5)d`uJR@skQ1v_}E!kB!2F zQ<`s2g&Ppl*rwmXj2K)OG$ifD&@MjCdI&B!tS&Jc^gzWRpEr3Jty%BzoI_06Kr~8U zH0wI`rxQMrxG_~5dI>Y0MQe0>0Dt?XHOL)HsX4R7!pK^IX!~=F3bh;?IbEB-A3L#O zd?Kx8roOmrDpr&`x?pCGc=b5}ISI|Kk+9m2qo!Ho8E)x1g^B1ID5gy3i^`f5;I!Tg zKniSKrqH(GUF9QM<@zr&!Xq3!Hq4igo z8I7oSg2tQxA6S7=TlEhUhPMnx{?pXQYy3g6j~Wtni;orP(V3edAE-+oC(PA=tYZgs z9i)~fZv*LF(Ulm+#=qUzhaK@u`!UH2{75FnxZVDbx=TFXbpvr6X$zpd<#jGfXO0@r z1r{OuN!e}s=egjL__~1Q&$xb1boIWsVGSHd)L>Y?Coi^@BL~e&UpYL|hnf8?_H;EJ z?)MDAySeUfP$oHcfnx+lyJK=>h2HEl=4S-eCBrG>Q0|h7ku%?CkMrxwC2S=bUf@NM zfAbWhlB(?>(;o8+FfxsGacdWqbu!ZU?%GyzzR*m&_|d!H9aI`4dYw}t0}>&Sq{b|$ zF;5!wdJxA0kHq z%hRoDk6ztlofrWrtz6S?~I?o2J?9X9fnBU9(y}%>N>14=odP z2-YXVv^}zi!*r>Kby&T6o$wplL<7c0Q+|8>tEObHWK^kCFJmL}5~BOQzjmCcjQ)96 z;Zsug5MCaWSmF{Gz`Z5kW^A<@fcZ@Xm^0+qz3JY$(cQU$oEIJdWbrH)1*AbmT zx$+UZNU~U)R}tgY#?c5eGxKxrjJZgaxvo|S?PaBwGE&XHc%P351fyXJ^B8@xwRjv} zIH+e|_>4MRD0xq2K-q({;g-ipy3^2Ok<9HhBe5|p<;;4nhv#duq)3$^Cr<2gF9V)> z(yrICqny+|!Qhuj0YXEBvG78}T3$7B7QD7*dYxID0B$MUHr1CaD@7KkVJkR0q(|Qp z9x831t{@A909|5`PgBiJCRe7EL#aw7;xZMZ5N&;Rn%Us6 z=v_@@KmWa$gIg?q=XWJ>HKZtm3ffIzhQ?6%fe!g5m-@GssuF+T^FFU|!GcPOX@W|v zYEn8ymlP%%W@bEa&P{7fl{b5ye6srZYdnb7K}L)4%cFxJyE(ZrMlyTVd3O!zcGHYl zkd8oeFu6X1?~8Rpc?0Ei2OsBWZBp;cJg$eSnwcaGu5)1VZJkoROQwlIvVL}i#r|j% z0R|UbQ3tarlg!m$pQs)yiABkRkAEgFj4yuCVtvMI@sghf=nfilU-^v2TK>%cm}2XB zJNNYmx$1tif>ARHj?L4@PxkGJF8SWv+q!31aVvfFuY?mnKULZvR_35vl=kJ8ymr3w z%FT4U^Lk3 z#9dEV-p3h8^JmTxqh2gRefmY~a(<`29dp}7RTB9^nF^D|_N6D){mce7p&{9s12h?n zKSHC{OdS$ln2*<6`6mMpqP?7bT1E)GS}XvDy4d}UwAaQn{>DI(?{H?dk)WpHqZt0v z@4D(+@*10YQ%WclkR>)rU1y27v>DiOzC-X=4QC{`qo1knT6?=lZ;_`wB*TUk|du*@t7RW6-w@ga#eFe|H#SpZ-CL?&)XjF0NLjGQ6Je!s`P` z90>{jSPRjS3DBV6P8N~4Vx8y{*;e0+4gxfV;$+-%0*urCB;ST)EQHkSX`JD@c)J(0ted+y(s-I6nQP>$?UgsK6B zYSFxoMU2NPXf)!EqyOw$^LTAboFzB!=USp}wjXR8^v6YVYwHc`z(m6fKufHW(8vag zGw@R^>WR&?Myw1qI(A3x%_a?{^8Wet-x>FDW@y}-xc27^SM3Ur6is*ucDAe5$xrT= zzi`M~+DCB^Pcy-XE97KGo`h9yX4Kw);0iS3udVIw1-AWxa(#E>ymOkKdVsp;gPpB- zJROO8cLOi<|NZo4^j_U)O@Um2g69|&B+r@5@BT}*Da`!IZ7FgMj8koba9;^vfDQf~ zZPRhRMF&uHS01a&Y=TgH-wBG>yGS;BP-?W-vN(&ihXHfqZl07VftTx0QX!fS>?FvD zn_^c87e37UnUa@NMk2U(zFFm4p^iK(@%J`$;J!%QuV*t`4P)mpQS+@jo?I!uo0m#T zcC$N+d+3fHq(b`VhWbk}ry@A9^j_=aVJ!V;;DlxY8l=BHfu8{P$^+g>R~JX(xXWv5 zC1hp?MqSf-K-dOuPbOxStL7C!!Dh9;iHtx7O3ZFWJc#zegkwO}xdpLkcvd&kiDZU3 zbiV{lw?Q7RxhhV$`~))DRRy4t$l^oQ!+l?(-B95E5J)B?LHr&-m$lk`8%rjw#_gRp zRCf8xFd^4DIcEi>e#+3>#U z-20jvcm5`f>B@^wRus(nHqCUh>jd|9Q%Kyvx)Y^W{6HW07-P ze;izE53A6VTqg6RxY`s4wfFpT=YOy43dJWOo>xYkH5tv~3XyLz^2Kd~c^exbQ}n`N zKEbgo4M!aoFYU1`UmL z8nddmg)IbrxX}&WR84$4$9aH7EM}cfwSOoO>UgvihrxISZ1rVaeRLl zrq1x>m_u0Ap9l+s!mCdJ5_e1xGR)|@W%F!KT9>JT@C(Xc&kbPfpiDNvbJv9S8oj6w z42_TCuWoUCe|j8+(8g}(cRf|id}DK(VDtwR9oyE)GUKi}FeAHNqZ;MBjaEk0TjN8N z0`fK#ixaLnXnh9K%&bp~qY#|sI8N&NJn1YM)vE%#Zys?&@CQsR#=97`B6h&M+FMQP zM&j-{?*n_*-q5d`z#t=pALN}5(6B>$=!+bT^fO#5r)W35zy*j(a$H-QsBK5UjYYQ`eOkM*-2$?4d!FI5H`FASPegw8b=^xo zzkc!OUcs z8>KJ3dtv?6?IzCAYhOvwq42|Q3yh}eE}MfAyti=DXyU27eewDZu|adJRGf^c6Prmy z`b(^K_dfnUndO22nY5qjluThQBEkgcHAtcdZgz@f-tFY1aU@+Y)G3 z=Kn_KZHCXd0h%W!ZgkF4SE!9C)VRDw-q&~=nehFzT)Ugm?%VxNAEzSXwXv={|IX#T zPrEFhz7=kE-=*Vk2qfSeUSd6EIs<(lQ~WsQU0vX#j;XsBibw(Lq;SHjBG(U$ql1u1 z;aShyBupi;yOQ0%MVHks+`*&RRh_WK_df5#W!kt4hf&ELbONs4l!3`@ocB1s2NRV!`Y9Ln6B(rXWw#b@(|YpwjVti+mlVFHQD6 zM_lvIiKy?>8LtEmIkt2Ufe1$oM5lyEV$ClFcsX%{mkm!TdrI#1>CO!$RgDLa1n<9r9Z-P1-PwD<# zL0!&siqlq5Y@yD?)KK<^jZjnJeEiNo8oj@;3dpaPzqoE^#y=eolt1Du4fyeKrE7Bg zODzY;NSC5=imve8K95L+OY^!bsjSLfsZz{r8@0l2iON@13x$xZmQe=;8*=+gm3cmX z(S#|hnp7+d)L;`ERn~bd{naL@I@{8eRYu<&SvM~6w!O8xip9H9(o;<)hf(p zTAQ~l#r4lPKg>4?pZyX4^HY7me({Gm;LB^j8S3X7Xmtl_=@?K;c}4jMva!d3tr0{WSAJ zx^>V*#4QaakzFHxS%Z{EmxZMA`gVuhU9R_CHrT2|wrZa=3ELY7d)Fs%rf5m8m4-kj z5~-W2NPRp~3TLJQOX_51zh~kPG%Q~nE>dcRr)9R~I}Rt}H=3cHE<(1V-6m&f}a4PXuKG)4_Z7;xihTcYp zI;Jy5r3w+)ATc@a&Lhk8NfcQOQwL5Dt3zlrO_=60YnYlE?7cFP)`RMHE=(ai)J1V% zWO-K+BU8;b?ma%`E7tC;c*%SeVMM6jl6ZDtz+#vCE zRDAJJPidJotEe~C1*r?3W*p-0dc*FDyYI4b#Ua%*mX<&Q0mOho>V=mBQ-H+o`tfba z$yzIC4-=WYVygOj{s;)*etZ&Pl@~O68o4Km2862>GF5J?Kr^T}J)y4dpF5O6$(&mb z!>3Ujh77)PQpe}&MoHk8O*zp8xX6xFjTkvX%rkdhi67qfr%#c3eoh^WCh$=^2K9tg5QUVKb8iMk>3Y*?@WjQS*hEKxlLc#+F?hAW)WMh+43CsKCmwbDfuhw-c05Zw2bG~4zI(e|qPU!vm@WShuy&3hb$=E`=iKSOkRQ1Y! zD88y3EQT<;Q8;fU9YcKG)8x;3lCpbP6xZDSJ2wqhUzXVfuzqNn;u=MLsIZDjX(%Y)+Wr|#z zQ8O2Kz*NvTzCHVXLa_`nEBF^JK^3dLf@+@VjA@xepz6cc8(XTB; zu`UoZKa&2xdPZ3%6`!`MRqIC*rP$9svZ4VgJobie9nXii3)`X@lbCE7pN=|+i^qfZ zTgX+7jPkg+)M5Z{5`_4_#GTv%v$WtXdcirZCJVVBMvY|$U4J$EACGY?r&Zsje z?m5lxw&zC9x)&fZ<6`QYXnsdP$`cXTX=Zh;lk}nWy_`GriF;3Ne*5FTRsCfJcK+d> zN$|H7YjMr~X3ZuFnWK5ZUGMD`LhmdI0|Y!GJ}??Cgk|4Bd&2DV#((3aThQ|0 zjSa1&ho#caT=Ab=4cLsODJbqYklGWW6aoFI#QOk2Z9$1@FjS?7Qzh@ z-$CUT@UFLGwLWPvW)Ccbl;G-znz=R8`H;?1IgX`k{L1mPi6~HL-8TqDzUU3uhI#Q= zR@RiI%7uD>XS6W&hE6;{XD$$?bEDGk49EplaD}_$wLR9my3U(ta%BJ{s;qz3f73=4 zF6_R6*V(pAgby&9d)1bEveN5YTtSTE~pa(-p_$F{$Z_AN+HKIe|t!OGeZ* zz_BePK!ATBj>-oqKV;-fnScxlee|e#v@}=|5&rk9)lO+Xbsakp%SiQHnd%r77#rPo zT~~qNn(=V^FBSSrk-6m?R|j9rHA~%lrSOfbj&7fm7?kj$a@xuZV9o2u8hAiIo&qbk z`6>_Nj$^ub!2rvRy-S%u)L+qe-I2pTJe7E>^`GVo(vkGOMJY9og%OS#FC{jA8G8enkj<7U;G=R^q)2|wsp(s!4853r(x_azD{2eAm`<47asx!*NpZ^* zVDP?TCX9nz<~X;%GD1nuS`BDx&}&eCpKEv~>1BY4pXpHrD`_ybiSPqV1#heQdCu%3LJjR(1G@W4ho_ zEiXT=f0}RZ?kMV{#<%0&*Vq3; z4&(e0m42te7Y$UaR5`b`xkUiCEvePKuO_6iZrpP8xM^_-15Bf1?BZ+}&bp!p2C6G+<|KyXDa{iTv zS7O|%@QRG7q=FPA;mVhoWyG7PugF+XXJhdp`winAQUo8cA10JPd4Jn3g7A|5=3&=_lqR@!dMD5x@sE%nZ;t(oNr|h;o?>ut@=o*7n&*DV&Q0Vr1R5G@_6C<;Xs;rw{ItD|qgR zvos4pW+c66UC)W*5|Y)$UhG@}Ng-|<$=C&1O5AfP-`BR}#Vg^5Wc$wDqN6oq)u)B2 z-NBnJ+`R!%j^cXQg6|CtV8&;s`bjP6R5z1~OW>f$dZrrtETuBO%GeVaO>H!*LMKeO z{%QgsA0juKbM80MJ}x26lcEfh{T?jnqjzT-J0PTx^4xQ!{gvBe#fN(}@zC>$H6j4_Dj=s!Cvc%4^ju#$83@q{a~tvsI$;5pP#$6++!Yd=-p&%)(L^FV#!6mmBwboKyRHkCt#u+IBV%6 z4BcZbYD1{D$=ZeSrO48HN$sGV|0ALH0VBuD5!V+Tpat5xZIFpq6x2>dq=Lv` zFhBi|w#2!urFIpK+(yf&3e;&Ep{7{zN|Ux!y_#2t4nFH%1UNcX)}FlZBI&z=u>z+< z)WI+_O8&ML3Nzho8TmR|Fh$|6CP>HEH>T0ZxY)F$=>X0nl{`Iv1UR#@5rRjlciG7! z_(8`l1Ep>)^=x?sOQL^mRyiD!dsnyN{_l0J(wjo8gpitR+v0?76sA4s_MpK+M%A{F zlREW!UGG;txb~gb_?Yp)MML8n`G=kYQf`uy=5}!q8KFJ#anZ4#tS7N}UDIP~SLyj) z!1fij97M$u0|99L>MgJSH zlNCr{2l{6F+qT@OMekkjbcMmoTL*IG!8GoZ1GwL{x-QIk@u$_`ZvNece?r?ykk@Cf zTVZ$i37_Biy^Z;39u65pbt#WuU!!soV#bI8G$Bv~F)%oM zu$Tg=DyzwyDpac3gE(R(S0pZRFnb*JthOHW*8#tX%58ruo4hA~&-i(b{qTixhzA_; z85@05Q&_XmlUButYWJIwU-Mate8}^KV68er0k3Xc6Y)@IFoz1dwrMI~gGj%Qk5HZ~ zp$7SZvX`Yh$@5nU6XC)P#~fN)%1poq_pavn`Do$}d(^DRn%-C_A#>|&POYQP-+U!w z4+X74#-CWOU$4`Kz!DN2;FkSvgVU>wB3g4AypK$kS!7Hrtzc7`qaOZ7#)jnMo#M~3&yFl zi;)G`gbV)Wxa;F#$ujmtmN?)q+&fs)tgTkwA6%KA$hKWV^C?wLWA|P)RGGtr9u^ua zxdoiP9&qQfz}D97aP4bME7<$J>72OUe4yg5{?>xHQyHq(%L@#ybnZM)vue(%8+5uX zn36RAIO(l1xTl~fnf4P3jJs|z+?`1Z3vcScwrlfhokrwhGggj~8<`DtRVGa`sDcK( zqL7cX`9@scSH}cJ7~MBTyOO~!x?}mKvR=Y{gRlU#Rbc(!GR2ptR(M$X=l6dU*$y8Y z?*Dx8(qb+Tm8MiYWkyI6MMuC+**?8BfqV@6rf_PZ2>c#WNo=0A>!U{YEuz8?VcMoP zlfHC30oR7sJiz!AXJ!pB32=T*s$pgh$ymkg2 z-$?2H4~S3#%W3~%Uf%#u2nr=vU1@|9r3nARb9q8<*+&FV@BFe*VXsce(6@ zTMai|vs1|xfv=*#Mgu@$hUL(j`OguknhF`Uxk&y46@~`mXA@&LA=@Bp88EnPGGJ;S z6>xu<(~T@O&qq0|j|#JR_T=PGm0e<*S`yz~9ZINXcUs|cutwCAJ^m(Ck@W%*%Z~yy z0<+~Or!+o15;{~p8ov|xDOBwN2~5%(;Wi zOiShFdegonz%+?+lxz|@n`;kjPjRV&%@!Qe0{q^sMr1I*ea6Y^@9a|ZLCPf&oUHQa zl3fr_VT3h!J4u*C+7ij)w)A}4k{tpViFRVdS;XJW0H$*#xe-o109U|7HXFZmHi7Ac z*HqL=4{{1+PZp6X?YHoYnlb;a*{4^eVfPe>x4+%*U5mnvVE3O&isf!Jg!TB;VIJcs z66kLLY(qR8h!$t*(@9gIioGJ;7$@*Op_Y1=WCeJ`O4{_1P>Ptx>gU0mih!Ili$OLUwQL({+<|aA z?^|krIWpVdqC)IMP~dp>A~Q6p^Y}fWVx>1VhHZE`*+!Byq**kY z6L13C0E{DiarnN|j)#uUIV(`cN-$sn?tip^)&O11$Vbs-iXWNcuO1=r0?PK z6~B)mesV4rcDH@u`?Ig(#MGfULjoWYC9$@+V`FpN-?FhM-AR{mAs^TNgls9bcuW?V>G<;Wz6%vgUQbxg+qtPlzuH7l)LdzzrexW ztAL`4h>Fhtfz^wGSG z-Fpf^(7U1Mn{ghM29!8VkY-7*(n!gDYs=4;kgHMgT#9cN%oK$z_~k65M(!4zSYZxk zD=6-_Bvg5i;QFxF^40!L8Bnth7jZfE>UZG*5WJ$*m`Fjcoo37x?wGXqJjMu$Y@S&xY@BL(nFciz}qP1hd2!wJyPMyTvXWvRz-?NU9Z0(dsf8!Huv* z4W_(uldQ%iJ9kLSf3XU*n?-Kf?H{LsQ?ce#ZCe3qjaeUOa+)JvJWK{_(b_#ZztD#n zuC~=nHrl)pio$cOT{4Bgn8IQT5bPJl(Vl?UNkLtDbNcV7N1!u2QWcg-+z-wh+eN>; z5iVD}jkkL>t^Tsik#n-@sUM!tcxB1Aqv;_!FRs?GUUXxV#2}KPryf!F;X(s!BM|Dq zr!!Y)HZnd(t%pi_W23!d0U6s%_f)C5kKbiHc0@6VhIjN6BZsZLZ65=uVu^*2GjKvb zb*g*l%b@y!JL=It!1+a^Bz&|sAj^GJw>p#eSyMz?r{kspgG-}L zs3`dhe-KUZd3fUa-%KBu;_dDWI#fM{YG{!+1wx?zJ4XA3@31*NuE*S$)7j= zylWW_dC(5w3qM?RJ?a`lf?VjGPgE*x;51RBPC{+KZRT%B>-O5d#zbq;4t5ddO#A*z zOiIA^p^}ISxz%0p!YPmFHUdwcMx9)+0R2Gh^rj@bjbC%= zD_=IeCDg;%_})bU@b0uVim_h|^Dt$Vd?FA}my8i{e9aY#Sy)Td{njuTGDPe*?~^FP z%R+W2(p~|)>JwZ^BP;r6V!Z7})vDPzD?L0fW$8;UIN*e>Qc}v4e61Pu_l`5BFDgyx z=1qlsOVIkzH#jJ*$M;;~3;>wrBxTruF6gV37M(|O99l+@I;cA4J7B5o^jnt_;2f>@(z2+XtqbPOC^sA?0)CU5zox2A&E-PJ(*@PW68jsUJX*{v-OvJkpz3*Wzwa zwRZzRbkEVce}em&pM|%!$HRYL`9A9adm$6XfOS3jlFt=SS2SFG;Uu8Gx&4a(qMDIB zn`RJFM>tyv(I#m3yu|R{YxAmTOjr&^SGb62xzD2mgq}Y0ei~pmap@g#q5PjPDcwhZ zvqE8*mqRDqxW9k)(lt-|XaJdlmgUB6l#4EM(zbJj&fk>=Pw%cP8n_=-1(tj-ZfoSy zJMN!;VmdVFszn|2fChUY-wX^8^G`zq`Z49cDsW7*rm8qgCeN7n(*ga<@+$t#*4@ab zs;LUoA6V#fTB-x`(lZtF86S9J9vU}&Cl*@q-*VeMsZSiiK$F&vW=UbtK~fhxiq9~K+G=6sf+yUjYRire+9-kmD{ z9K=!O*R#}l+b@0ZMCUW)`|QojW1l&9BUB;y*8e@Qv(RQDjwNvX5~AC-=#3FVEFN`` z4<6@cU%0oWRx>AzgKy4Ec?^0V>9-lnI$t>JNsY0^1*LG8aE`9X5cd3PcHBa;8%JC4 zNLY}aKQSB$)Bd@BFtf?@c*#E|cj!UzZ+f}1T+~F4iT`zK`$jWdpwO`Q2(Vl@6%L%z z7M<0bj^q`7WT&cmo>b1Xcc$f??2MzGdx`@791Ly>x_}7JqC`X<4lovtuLiKhuX=yE z)1s#uDbL8OjKJRq$fHYtHwETX_e`HT+>!=-WZoFVvBRqI_cl=+dv8O*3C9e?m`SG>g6aiHGPYl!r(!d??_P|X0^X?N@~BAwfy5VM7HRXrY0Ipze&K4Kp8wQZi)+ftxQui2~2f8LZN zm!DQdl$0|ge>3HVFqp+J;XE-)8AxfPFsXQc7H^~?#cYrKOu=Z^RM|{OX&tsvdRBOu z&S=wrZk%7UUR;Hmr!3unNsqP%@U4vF<4&8AUL7U>F>JZXpr4ix?TE~m33pg zlnji^wDQ(dI_mDf$0`m@4Q&6ikGfGFg$~2z%+w_ zAgKE+nbwFs&q)4EFNu=!*`)V!o|)f-27SV4sz`9FI5r(U-u+}M3xHj_y~W)J<2d%X z(>@?hZz{<~H-HZ2HR=QH38Ybf$0I~f)If(ynN_cJ(J+Q%CN{*T0 z6_x9i$;M0Z{@U#ei>2C@zDIH1p2KPzY%QHhm=^~ZJ!TW*z0G+)2t`Pp&lft&uCcOz-brg3;durB9GnMBI}o<#r&)J?x>{QyeCo#U+xcK5m3wO5yk0jKsqrv9X0{0AEEUmC_V|CVQo!+%r=bvp1m1_W5A0gYM;~N zHb3Z^grV(;t|*VH$DDtUuN_r07E zer4%PVAi?)A(G=9)cBdMr}{KQo*WmXmtE^9AnxE4R3?zY$b1T9I;%oDKLH%kX7ATM z{2w$*+$XG{_4tdAkq9EE=}vbQ7E}-(Z{ROqtd${F!DtIi z)SKk^5u<#1Gb+{#+L;!h*6{AK+59b#QSht;L|zax82o2jbQ%o%-ALO$K0Xhf+T{Mc|xb}TR5qi}ex2ED5Nbihiekd$qbTHk+Jb8m!^5_ntS$_;|%$^Mk^J@U@QpQ-zi#VqU z^JSzqLlA)y!s{ND%&)A%8kh`}CMcvL5a1&Vj!T@QQpqL)u>$=4qxdNyA)TCDYpDo0 z{;hG}$?L99IVMf22XYVVq?LY9^d=tOVDB#dZoRvlu}TMSw>`$v>!~4_C%^$WyEj!T zL5#`LQoMLRn%2t-s?1D39TzN z?1gZ^FYs5L?lFx~L@E#Z9TRqpI7mJWA;$*^#ko=3TjMby2+M|`bZsiAG=h8pgA@uz9)7-WzVOSQII+)(KPB2~bVb$urg*3)g-i%4VM3L&C?ZuLNEK;~EirU9ht-4EK9#^vC4m^Sdy!W1|nbH!GSP;K#DdKh#Ua-IxR3Mg+QxcQK?R8Kf00 zn-EcXDIkTb`J{u!|Ag2{L+9;oIf_^^l9}O&1(Y@kj?1q4??*ZwUHbb1y(c7 zE<+z=i>09Bb~DM9Fkl&MtZcgZWTPWExu zZpJcTBw>6bjq1-&aW@zUEoe>_btyd5LB3mB8ypIjf0VOTdJwxlzE0#4-R1CkeOT%6 zddXDGI-%k}Np^0xda3SAThUo6PctnH`g z#l~0Qk6!iDQqAqIPQ16Z_2gRk@C_lx142go?TAR{45*+XB z#yP6PS!WC$h4`NOaHcO4Yk-ywUg*%mNz~9)K2k_jpKf5nVURjx+ zf%uz2w<*GiYa;N*wOHLe5E%n6nlP@YA5#!VcD!qPeViTDC!(-k3XHWX(wfEwW?U~8 z$6A`QE+hR7DjGNdPl%=b3Ys#lMDuf?Njm&6;Qs%~e1Bixc(o8{{KJ}$>rHFMGaiG1 zyvnay(VOaJ3IeT;D?py`ADI4vhCm-ZiZ26STKLpmU3specoO(XlP2X;KJ%@gK$0fAlQKS?7 znc`&AY%zLnzAm3I|2xOziDeT;UKLyG(y*1|rw`T0eN+2P|DXwof?96dJ3NWdnWq(W zG@c`2v-N{$xUD7;HM#|TPjqAEzcsEMTkG3QiCK+%f9AClvVSL+ss3eV-MI{!dd?Yt zTP=76tsEKCE%?fisCpwv!>yYcaJ*|H z#Dz0H{~9Qn%=~COx@*L$C-t?5*84=rru#P}xH)ulbloms;u|iGJdj+9lr3|gX*X|E z;`tXCUtc(0m9lnh#N>s$P5xinMu7ACoYs!z$6vumAxgKFH;dS`*`BL}Z=sUmboX2`Fz{gP2wIalIv*MqF0R@Q|Yl23GWiT4y(GuWSg>(ln6EV9mHx@ zaC;X3Bp=dv!Z1ymxCYJGagblZuvbP~E28j#Qb$gS%RuT2x({2=EI5-Ks&r4iKKa`j z``H!aFOLzN>VGesskq`K+ZsrGO*M_rO4q0N@%uHY-+l6(T};<;A)ghH__}-agtB&z zf`2xu&HV|@AWP`U1WTL82D82HpMFyBxVv}mzEjZECN*r}n`92JHPiYe65u7g1HomKrdSwO z5o55;K5zYh15ZL@9)vg^f+O7aXk$xxd~h0cFQ3$SXnV2gGT1>+s8`Ouo%X8#zj8@rSYq`ZTw$b^7T_3Z|DJsu0k)2^BN(NGVp3K z*f>kUF-Apx)5qrpneE7kIJ*uGe+oPDF1*&AYB4V$G}JAjH|6J`UDE0E-ABj;^5|8)!K72uQ?n9QRu*u9@m zOK)^)??sHitdp;>Rn8RBCNE6bSi&*6pS4G3eLtkC>^Vo=dY{tAW5+d~;_JXgA#q{G zgq2_;ZkfSv(7CV5rxs!!z**f?U25OyiF|r=3|QU*r*dtEaN7ad@TMz;O}EXvNq)*a zDSf~G-g`mLLJ0vPuk@hmUS@ot)1m)9waDG!5}K7?H~_z)UUpLA1AOCxngsbfHzf%n z9S)${u7u?`V9C{62EL1FY+pFxn!owtg48ChzNil`-yMW4iK8A)jN%qMZBRa? z2M1mVx1A`h9P-ocK*>p697tGnLzn^R4h#qA5yc}PXS|QCnhABC32l3-&(|S_>0^Pyj&nxrW(lcIM#c#=O?cB zqrd!$*r|0o_iYzXIzvje0eiJ|3vP&SX8ZEGbwI6{Ww=uwcVB1=P>7n zs}zA>QXq8MI7h+*D;RsrDtVRYL_&X1k8xt-g}a!VLbD`+6nV&FLxw2MA&$HbE5g&OhPi0Qxd^>ErYuYK*N%|at??^WYF&F|I7ClaVK`A{Bx>c+n^Cx^bLZnGlsw+8={=Vz{ARh|OZe?Tlnt?oe>kJYOIJu^dsImqAa!n01+ zUwZGpu>6seZEn&w0x$=={7|RbFTP+{DtuXkV>(am`N}Wn^!i;c02;}dbSwQhH%d)n zm()yR8(UVMVM{j731d?{T~5V32^%Ya_>Ix^^=b|aE0Io8wf0a>Fht^xU|vR8)-Fbc z%G^Ca7AkYskP-P0&Z4BaUV>(!kj!2gGu}>LVXYzsXia&FQ5eT}Z(2tGtR{gjYm5)( z_V~`=t)GeYD{ue)D>h8pgQT%k=hyCA- z=e5Kz)KAYQ!Qjy3Tp9RR;rsbJe|4|jPqX1;@I~hPT>R`k%&*mSffd51A_{jqe$q!j zb&>uZVu3K*rog4meeM{pKO=pV5eAJdzS{O@@BMGfz<><7fq9Bdqju@9vk_eYG%x(l z^z*vE1M~l(#~&Q&7ZQnyJ&su`@BLbA`GisX&wS0a{@)4TNJuY=K|bDxO8Bo(;1XbL zNG)sm?O(4A`5C9bi$9@R%*9rq`S&`-zk)1+*;aAiudL@-WVf_EYGk$D08KnM25&9it4$j*$ zdssm8KLGxVWMi1v7^6Yh-Ws8P#W#o_!`q9wzUP0JzL?4W1hap3du1CAhP^1p3kS#l zFtLvWPPn~NzgV}XemP%9ZMJ#fO74iB<4$&Gb z5fe*Fk=9GOg;RY7@jZ$+HIze}2Np1Qvj}Kt*~%rc#!C3?GFgB02}S{D5o#3cYrQS; zJ|f)rT6vN9$Q+j=&meMdr9Nrh<~VJZ6;*$4}GT7Q7>YPz8WyYkf!_RVm(CsyvZ3z6yiP2_Q&@jFNNse62tr- zFN$9mzaIAhGSNs+4?U2@|l>kd%& z`_P#x?~c{Aj;=0m*Y%5nsP3B&KX1|0Zmz!9@A5l`b(zxlzxvpjl+hP2%h255%iXOO3h+B^D=jY)X7GynOtl7)`v9v;wcQ+b?6qzFF14zt*8b4T zKx&`G{UbA!9OEayEtVKsk%ThAMe?%2Q!LKQzMo(MW-kHQT z`{{M;T**T%wib|;^Y>?ObaAi7J_e=7fyt9%;<=HYbjw`U(L0L)S+*zz*7}{Bbc(Gn zr703lB8R-h)?g~E%`MG2(>q!QZDoS$vO!T#-8^=_J|_rYb^I1;#xg2h?fmAkTMdwQ z&3qYT8qL~th^QzdN%6%`Tucwu=c(NbnQ2R4uMs`2pFj!*D5ME6P8{L0Mif^*2MvJe zI*!mT<9Uwr-LlZsfDV{;?M|(lh9QHGc9561*jzD>WzOT7&*NK~*)LCRdnJgREI7Q= zxaT96y6N98b%R*-cI*%L7f#4-^=*2N6t5zVW{Xp<`{hMW(o>pT&E{(A-bYF&?4Yld zPu3`KnrqP^SQOhe|2tWMPlX^@7qc^F-d;Z=Z1z>)F5*Q+p{?o@n&s~H#nA6_{!T=J z27?BVmV2dL>2!|~8a43H4%xn<9!l@`A22idVkX(K+q@Ur;e5Cn(?|Rfy%pMV z=(E&CdC0h}LZrC}F5`>YGd1$wRkd=NpN;^{tAN($%a!>mH#NxYQIc}uBX(N0)R^H09;Asud?>YA}Jk$bE}2fhk)Tp&Qjk%cf+xS_lb2iiDUoQ`WSwl z=V8r)J}3T8Yb4j{^VKUdYDa6$!y-aua~n=3DaR_;qN3yyvW4qLd+>#oK-wA96Nc#K zPPR=nY>xD}lvGLA9*t&ym9D0{ZnY9gMNj0xZYv3ZN!#T8FRA;p6-mhfz`urdS6}~! zbOPTy*7r?^AdwGf@8AHRP4?u`&rV2Ged45?ChJ6s;&xRZTJS$yiE&zE6#^R~?h~~k znO3#AA~(M})~B#sC1zno2NNtYR6ERtZI)h{V6E&vPn3P*Fk;{3SeOb4xo|}Ax(K*G zapvij%a=D+uiP+y>~o6ufZqA&9v(o3a@zk_GSF_yj=HYEkLwrxS9mm=$?&p6PhGAi3^E z(~3cE#DNJK`wz*uaz()!MWFOP+|szTih!(v*2n7gLcVML(S&{V4R6(u83np)`N*Z* zu{Po>@lyvEcHvqs6xARfn+AGVKiWc#D!*XdUHD{wKBJ~sW1tQX;vr)kWwB&nz`SM{ zch!*CWIBhb%xziW3Z5L@=aAd+!4{KrunCdtW3DHRM+9VSs?~PiGdfNUqqT~4tCMdr zs(dmD0fXXVW5zct8UtPv{LsQV8C^TV+*_BI-y}47trW9xDqjf@=tUbxojsP zH(ZHgt=?);kvCX$@?6nCeW|{PlMBUd8|%9;$SQhQ-!~+kPvnY?N}O+%z7C;=SV5Oczb4_kny}gXE_z21GVbWfiQf9X9UG1<7zr~zh zT>k!+Mt9LsnN1TdxARh1mP*Ubv)kVBp z^IT~aJzjf-$oJxvdB9~ROIrOyxq<;}@2QHN7P-H4JcPrIOWp9vpymRq%&_i#5$x@x zZftZyVX?>bob_B{=q+Y?0)8bue=J+H>yR6tFhg$8a?MG3C+@D<4a;g-^g)@`+nqtv zEDe8^XsO}9bVs}?g_PO|Y_L>+C_BOp13<~0c-X2;bx3%*Y1Zr6cP(c_zfe$5kIlq> zje4wqv=9U#gt%`_%CR~3z@+fIC)31hI32ECJ-2P@7c&!+{vB^ULyGFNvk zx7lbVw?c1x#BL}}uF3JZk?9c$QGH_hO4x*H;p3ucts0It>yK)K_4~CvO`t-3bC7C| z7peLCRS5q&y%F*jf4Ffb}*9SKwF_;`vo-SfE0@Xj2`9}`(bxwE(M zHoCcwcj}OA!D#w$WzmRtcZG$xJ6L+x10)*?)&B~Jt-3WLah2|FfAl41iqf%ybhWvT z1Zv4WrIBjONB~W>{qIWo?2vVXIobeZ!&Xj^eN!M7VMoei8vatuUImK`=d{PpRd87IhKZ4L?GYyd}&> z(dnEl4k_{;uUaoGJ)DKuE2X@CUppRMIyzl)h@RZ3mS8rdCto^RnAXZ^aCyGvNd+}Z zNh3Z`OSgHt&YXj*?kNyZaFdMNUGDJOJ!?(_TXfm+)|kUc4wn8}`*wI5WVGVmBgE

KIXj-LBZfC`9hW>921!PQeb=tpK=#H;$q3;kxvqBZ$SKZcoZ1b*URHLdYEhl{F^T0j`S#U%URpfbY3~*6AWRhq znA#%K-0mq}t=WlZgcierbF)Br2?Hm!PJ0Y`H7}iPP;f_Roq|$Z&Z~E9KC>hwd!Lm( zd;%;IvXNJ-SFKS*fPeD9it2|u^Pl~BU%|_OLlKNSfuk8u>#F^4-Oo&DvrErH(rtn-nFT$B7gtj zbR~5EnR`lG0`-%LLfrc>Fh{ISxNTrTWqv*K#D4Ui*<@Qe=*5GKZk^0`mK9OKZNX9l zQ&k`_?f?~c-^RzVEd4mbD{^-!1q-=tajHi7Q}y)WDV+k6p6bblMEnu>B5E)-eSbB= zX9r6hbF%a9r|q4o(MS2nOpVU6!(wOJCvUkvGV7Izs)~o+QX_KHWuzW(%Zi8LXCE!e zJT&m)H7HIrNr5*y?5%Kbn;ad1?|iJD0mrFZz(7w1r|<+ckDRdclSj(h-NA;l>2Tc) zVm+@M$@xlk;%(A<-g-Pvt43Hd>S@o!L-W3gfw=7!O00%AB45J4bx46tzKl(UruTQt z#<$LwH*ne*0I#kR`e&pUOmLG(lz`$~bauXmB@kb9ymT40)V5K()A*^AF@94=A? zV|R;XUr)J8IBk7UG%JxH$Q4zt3DNS!O<2Gc)By+5@k2|Sju6S#hXpLF8%k2NyKc&F z3t*QkEhdn(^f|nOhj#LAl|482?7Pw-_?NL^ZFY#03CpVaF9{QBH~Z%MWs}rdKml7W zh@+5gFO7ySs@{#)JZpMWb^M2guB+#uMSI|C5pJRo+$awC;n1cLUB^B_K3tlG@9157 zbx!B83jf&k@9+p~odl3VO#nVMaNUB8LVi@#iQL_y$)TF1SpBJO65~rsg2^cMV)(f8 z!q@BDBa0|x*(Gb3T%ySjbsHesso5hj_K8^To{w>z$o5w! zHW`5nuA&uqEq(DP4ikmq>jlW$IufXo7U15NgN0R1-YB(v`a*ph3-V|B4ae@O4^P}+ zT0?%H;BoCq)5+^B6AO#%^hMo%+zE$>%6V3My+~^%_(T@E%vADIny{dIrmvot#oVbv8v-d=n>*IuNy~Jv{oW9RJ=tFb z`=rc$np@;%ETV3+qI&xQiNf>*#0vC5!fB9HOSuZ`Rt>9 z(YdJj$RL?z_bNCqcZyTCAgyo$vnp~EbEeoL?7*><^8Z0j%pX71eta(gcA*ql| z0R7f@PK$4v5cs8+s>&)PpXaVbl)!L zfA7Q93Y}N?T}$g6Gd@Pd%8waE^3iucwt6~PNWpWiWE(l*;JO$ySR_@vc}-@~LNUdI z;PH84^ep;5x<-;GxQbjBTC5Zo2%P6-`_2d4mX{uQlp%UOMXds)*CQTmb7^l8M8Yoi zay#N4FYSe(u!;boR1w^@Y0MSf{wiHY&yL7G&5vx0E6D^i9Oi@gg#G;H*qlD+8fMbo z$^002RSLP5k_;s--i@A~ENzl0uH8=_ceMpWH+QO?Uv+7{?O230$zM>bfB_DFhdnW~ zF609hl1-RWuV=oqCt%gX8eqo2jMvjaNhf+91e|ZV$VuI8Khp0aL)3a*H|THDZr&1d zP*|&DU^iYJ-KG7uJ^1EBjBucbXLW&wVcP3;MGjH*eC-~!%jNJ}i>Qv5sGKjRyC(zf z8^Zj0aN0#&F?3KE=Ld2xYaU~WxL3R<-?SsqCmo}R#SM2x2G=SS%GpXFpu=}5n4LZw zwb(7`mD&O7FlFL(S>-&HuP>J}EVmq5<984W8Wb{-ch9J56y^ONE(MIuimk8{j#b~o+&bTX#lJ7Qd{z$z1X>9)jkj5*v z9VBNaJEr8EFpS!BTV=T4`Awc_6}Jt?yUqF` zY`qweD`8CpW}4YOs=L{1nH9PcZjynKoOu@VhoL*ulGhx}%Q=(%BF2ayZ7ICDcz3_{ zCHlKL2N7EUo_-=9@{5f@g=vcC&gMg4e174d_))KJQ{Dv;nb-*VE*1#!B;ag9@?aAe*0?wWE^ zcz)A2-}5xeZ#66CS#X;9!m3)m()m4rT0T+7klgD=E;2T9DDKj(1tR%qxOBSL=gXa& zjXguCVJd>wU{$k3fwi|(&s`T*RC0 zkvpUpbU0gOvNZy|4qkE^Rtr|x&0sXESRZdGfHJAd@Kjox52gruYwnbLvZHXy^A#f_ zRI-$W&B}*%uO(G};RQA3K=mg^Ly0aq9^h>eCOTgl9v2LlIwD)BeiERPGFBDeo4wYH zK)Xu=N!kVpGl##*3r0P|zATeHtkWHM=23nU1CqdUkOfBsw#&U5zuk3!hEaR02GKgb| znM@7P=l15*y+`rBePwya7~gzLDNoizmor%yR?p8X{oNm1H|W5#5-wF{Tn(e${*Exx z8!ZS2Cg9gcDZ9Zx7!BT98S1LL)anMll5gBP)mtHY*Q23mnEk~k9aEj)`b54X2<)B- z_+}ZeM5)8-C`S*tvOrv8uoQHxd6eP43b^ITJkHH(9;ECAX|l7>yojUa^=n%vGm3K1 zJQ01z`U8dAC+l2&w!Qn&H}1UYGSM^xGohmVWm`@;ZOp`B!vY@lmu2MDyB&Su%`>_u zK&ZfL+*fYN%b@=5RXMl9a=S)Byg7TrFP8xosOAaFY<&GCK^Ou_u90H%&W=Oi1gDN5&Vj zAzJc6MBl_=j6Zq?8Kdk6{%BY4V|V_Y)@b}6Vn*(X&=3ISKgTp zMVQASW}~RJ2^LXF#YfAd8J|+bt#sOVMKs?4ZrJ3G z+xL;6t&M=|hP=xcYmk-Xyt7}wXXaQ8Ww%4+`x46MO)_!6E@fxBH@Csfx&F3+1l}pLlL1?xfjk?m(?GwMmD*Je;^Yzc)%# zjgwtwDn})FjgfTm(0~jfSxHg*z~pK!G?J$}=w(6I>+6t2uJpVae0d}J54B4y{m3yB zgjVF5cMaK{&q3_;94?>EwItn0H_Qg1Y)02^>+rz)}2r0uL_)N3c)V1m9YTw5;M@IcrX3!@|Q z!!;*%*YA;V+hmM9Kj__dSmy^n*nPci$3q6`SNr76g$Z?;@R zy$B;>w!ieMARSWf9!GS^At=K%kFZ~b$loC-H<=ET#jNcy@*M$JJ8-)dewOVrO)KCP z?b;JhNyDnMhU7Ta`?7) zZtfLz>Qp89A{YD2R=iTxlLxvclh{NDi}cnxG!dtOspjLGYM|<}+#JBL(Re^v>f9>J zQH+dIn?daxoWFK}DFBnZ*)y&AgTMTYSpWxk1UW_R=y%9X-7 z`m{BHrZ7vITseNa(>5ZZGB@8WKD9e%JY8)s>T%iA=e<{Cn8;jmja%RF^pu{(=rAlh~ z-ScD-MG-*GhV(!tu%QbMXYLZmVN88-%nB>UQoT}&N4^`sGU595r9N#K38{75T}ynO zs&$C{%?Bccsi1mI^v<%*xQ-JAu|c3s#*HPV4nW+c>er#;oD4)mZP_L5D(|F(7bglc zN!oB_zy+ELqm?0-xwHl{ll?&X+K{UB{vpQ(C2%-~{X z2-7@Sqi?|lf1o3`+Igqhoo|C-mw+QNTyJ+C=6YcmXQ4g>h+;iRM@^8Q2y$Bwcau5< z$i4HmVYRpxK_TD)oA0OI97OYFW(0A{qKBg`fa?R4+~5Kz(5SWhXB{Pro^Dy=y`gP+ zWFiM#dhPidGDZCA7!fuB1wSZ2EpO{E{O;WT4Vn2_3rou>HTOFI2M{DVA(ajE(qwG zBl2h7XZq$pkD`(>1=L(bYsFcrW?vUb#dJ%PJc_^)JoQpIloPayutB|Waf7*VLZgd< zkdPyUxEkdH`nY%gDh4|$2HM>zU9;Z$_3ok;d&7o%93J%fPA}MUWf7A~T9muRhlWfr z6=@+c)2`9!twgl|m2SapNWb@jH`KD{ZXHfIWV%H`>V$WzEqzjtszQf%hj@?cxLn-( zz`B}@`^I(Je(Qikm$>7DDGH(Tu1o{hRdsMkvxFLhMgN99?C6VV*Bt$jWQ2#L3mP$W zk$|?ITemmMzL0H>??nz=eS2)3v9ligUahsuVmu)DcML>C4vhxrDbcqJ^baLIc9m4oIvzLS%HbVvj8 z{#@{Xso4L)bFI>V$L3IzqrGEUS!TSFh=y-P|31c{R)` zRnj7^dxmX1;(|uj$a{WMgKjg2wmy`Jv-!~7CIP_*+ZrhqsCo5)&vVy6&ukIJpXdsn zg*!`t&6Wl|N=BT#4<;;t+hyQRH)9`leP{|cNryXRKylOpA<)G4>Uh;};c4%d$H!js zg7VrP1BFd4ugoJ>*AtP0lNRwDo!I%bcx{65`OPTw5kXt8p!%pa%^GN~q9%Moy6{PU z!GeW~pI_1WCuRoohbI?sOg3IsbrfCNrHG|l7F@E5cy}^AWiY#J+rV{ngz^-{z3t@1 zazoDJo6l=qNZizRY@)mqmR(7ea}s=Iv1veAy)ehLC%&{LUwKGdcgyF2~u{%$f} z=nI8fz{x>YNf5-y-M>F*IJY}b(>}93f=t}?13h*P4*sf(su!KzS!Ca{?*SKURe5L1 z9op3_QpBE1;Mf2XhtX-!hg?b1EqKvUVI&2#+wNQa=8Ge?Z?u4$Ev}pIi;C>rTPT-S#_+st4FD)?koRVT49jDz=M7LbuMomz)cYSIv04qfcE+PODkzaIht zDpP4rjCW=YyVBL-rbuQcDi8DyBafEyI<#)vH`E1=1bMn^-adBg5@0r)0}}ov%h0V= ziHaL(B_SLU`1BPv$O)^#VpSb%&LwU6o_g>ES|t+p;I_jidwlGTWxz z;@o@RsU)*cqgZnk$-&a_QaC+P-)sshr!7c6sgpzD4p4Q;jpV6)fk}%zK9*DAlVwHqBK~Itp*V;U;anG!5b5f z=|@XOJ24#j_}6wr)#RnB0t5A8zJ1$qA9$!hq1e6s4`)NX)-{tqeKhrdq-sN57xrd=MgWkchjmT3>R=1L=-ix%9tk?S=~ ze8)9A6r9a`Lp9dJjj1tIk}jp7;6(xQvC5b-6PYD&q%4rYDmee_E@U(h6i`4sn6NC+vT_B`eJSJ~&=r2#hC24aV?v4S(- z;m=0W!hjJ_c`w`XQ;SazMg&Ot>0EBwsq?s}%SkM5$}zV_bBDh)XkeIb;0GBnA;4Pn zHR)=@ad%Y^D12~_2=$XkLZkhiMIh3^JP;U0?e!GYO6bHGHN|-?p={%&=gzcr`O#zy z$qd-pikawZkL>fD{%DY1LeSVw6CevzxlOaJ9}pd3p>z0@u`KyqGiK_|D zC<-77>vde{FK^K0qoCz!%nG<7j!P7#JexJksNViA!6306xH_Z^S8E^^Xb7wJWJT6U zm?_DuBM@>7V2&k|-PzN}839+DTI#LqYl|=3T>FrL1k>apbBoGZAQXg#W3O;etXA^{XI)1~$3deM>iLmi>F>i&# zHMaxcx|vXhb@7Xh1^5PeaK%&AfwY}x*r`r;tac5SXEGxND{7?Ggc;DuP8O$d2uN9X zrxXd7c`|sA0T;V9mJUe`yXWBtugrvsU~jcA!$-9$zrG5wgR&Sz9|%5#H1wlMvf&$= zq4+@Ip8soY@s93T-QY+3T2#zHkK(s7r|_-mScxY@Q{vlHKHy!Y$bq1&_lKwI@NqSMm zb%bgO^^rvu+8&ebleZ*c`~D(q$1L~+uSd2WnhNz;hdmo)F{pmPkAqh(7#l>e#Pk+a zGHd?2(=5u_(<(qEw>vp*%`50llXGApesR^i{5WnTP7TJmxIaXDzu|L^*N^kZ-jZIv z-AS#{m29dbxP7`d!WgQSXSY3Q&1MfAvuSWzl1f}xB$e4YbqQG;o`V{mAA@i2!fKeC?E(Q>`QQRwczh(Ethh`lX2f|AX@n< z{Ge{ZE*d{0nTJT(jR~F(;W%hace>%tF;13`TAVlEZ`6-rp#HQpwM7a|hj@1dN2ET5 zgS7${w=d2#;(w#nuXE_~Av`zlV68}Q1#+GTantlr}dgMcstdFo=1?VDuv?jjd5 z_JRg(11s+*X?kACxyud|-Dosy-AsZg8DV$g?`5#V+7+UjKUh#9Y9FUYzp{HOVmLVL zfm-95bLTav=VMEEg-Q{@oOA%2;p9_?(E^dBYoYo>q266Z`t=q?iUzhkyL<~~AD}=h zZ7M3aBdTGrXei&br&~$saL3_9SIk#qz5i&^$HPesXSYx5O7|!%ze=m+3DRgy7$%=o z)%Ma=unhku3I!eb22>oGyZty)y@8I%{*6^0VFsi-8|jZg;7?))>Ik5+@X3C@;4EmB z#^il(R4X$WCmu!U5$RPtKl;GPtf5Qo%9Aj>hwo;(Z_^{3)6Bspi)yacJA5&!S#nq) zr{koum-gP2OQVKebJB91pFW#6RmHHdr#!MPjrXH2mg+yM6BkxXZBT8ETEVlq9P`w< za61kkyJ)FgAGdM+5%+Ip5SAEyffY?k6$Fsm^)B1bB3-T}6LgjLs-Cd84wSsJJyxIC ziRYYps(2$`e4yF51y)+(Cz4yv`KWbRJD8F0zUZM{BNSD^*t=ZP9r@)LTm^bP+ZCfe zcP+$oV$hFK{e(f)%+rvQE{x;>3A<@YV#jh75}gJ5{*d#000EMGxp!{->-kWk{J}3k z4^4C|=+hLQgBTH{*C?6WLjiDG#Ntm%+wj8|(8}1^`fi)tjzyW0dI4T6oaPTXEXVT+ zT(w+QMwx)llu??zXU<>m2LcUd8ej#OEj&l_Xu=@n!@Y^ydi&;s#0Ef90b|Fb%%Huv zm5~QHq2t|etGeOmM2GYPoM0iN9h+xYT)hD&X(KvR73a(7#hcAv>54x@#YD}i^b*f1^a4vDXsDX2Uun29H%4eNZ zaI$R+1G*h+2d8Kpzx0!&wT8s`z(Ulr)jn6f3Q-x{*2)lLT^!ux=(C+T3};(xS~Q(d zjr#mpY`Y0AX6hVj{>|dlFJQ|LzBYG~Iyk-X)My}NHrp`3{RUaG{tkPQ=&Mm>oEJOW zd5rg}S+1&ym)U|%*y{R=w_my6TYB%xI>w5|E5jFXS$V%h^0>{a(qokfrc@a=FtZER zsP3I()F^ab%q%YSTGp7!Jw;=D!5A{`@^&fk`F8D{@ftLWugYdb4z3P=%8@X%%6l5e zYAesSj0S&ZLd^?vEi|SGR1vkYr<@+WMY5wNbZAuGJ53Af7>BsqIAsgXXQbuRE%Vc& zgTq)6{CW*V7O1@=)tL}nf2~mO=CLX_@#Y90F?MyXg-QO1H{cmOrPx7+5JBL( zLeAzu=Mu`aVSMsED&`$&LwCn&%Uf|8HSBStnm<|q{?SVRX{Y4LZl4b)+1AMQqhUI@ zNqjwv*r}P!(K+?Avf-bN1POtt3Wv*&;mn8JFFq&Ca9YR3;AV;wVGzqNS)9ibiX+89 zlyybD_@T6Hqn5J8;_%dsH=J=R0`B0wd({tI=dQ&ZbNA8~UftTXN+HTF0P*mMl~|(g zwI(nn!ALV1cf=|dNTQ8s5?v7-%ECJNEs! zsG&X$2mfBoHZikI>4QKjY;;CaiT9Ve?)JLUE3jrWzY=Vl^~52(v=FCVy*JhCc)@#Hk|PD|oPxG2;MG~81} zs_OMD?7xnUp;4u6&A|#95h*VOW>CT^yMpcE<)PyC`_3#8@TZy)UzJ*YBn2YD)yXYSh9ZF`fFpu?@mNdcv}Rp8 z12twsd~|G@lOB6# zUA&aDJu-|3*}+a0t^slY-%D%2y$v!&=gvo$$U>=~c)2&E1589* z>YmtUIfJI~Ilo(!8=lj2*)D}R$5S#_WDY)CKfQ5g>;Cg7N(6|PvZL;HnmqW?U-wBK z!0Pz;6WsF`Gj0Pbp9oHexZcedzssGEwTL$F%Zaue`dG*uBk)8jh+u^#K}2N5>tePQ zE+CGS*gzAW%z3T^*)uQktdsTE5g(>s&-WwqPmbq909El9i!W1tll2>sRcuc}lCg`P z2o$}~l8tSHHqjlF;sc3oG)^SzP!CLxu^QpjO6?3G#UWNK4?#Qdhg%A{GUbeCn)*>z%usU68)Hq zG7z)>=R3hUXyn<~erqHx0dPS~HCXP}*<`GkOhXXQ{9kkWzoGOQrrx$^N_NJ3QT`Yy zOi${}K9j<~THYUU3&j8e8kc2D5i9wFLQ>BG>|bH(*#El-^R>mGx;a|;4Bi+Rz+%Sk zM%54hMUNkk`8S^Wx0g~bn9y*7Y1aR2GEWAeI1Nn<3;ZpV{{%gCG62QPaueepK=>I0 zeexiZaWUreKM3-Bkg2=`=DKB{R4sKj*?T~?Lh{L4&UQ@w7V^(wRcZL9_pk2GS)BnZ zEeCMQ4TPZ_@!tik00WrEa<$zVU~d2w8S?su*BK6fZ{+uh7+{NilrPEILiS4m*%GYh zmb>;h0VA{npm2Cv5^_d1Uj#6h>-EYrOZzG_uETlepcgs?DB z2yYaGM=gGSAGCscPM+Xf8g-O_ zk0tq@0`qSnJ(@ax3!I-HlnY2;?xp3Qi1i-WL+WtB@UJ^K|4r?0sAAx9H+pmX3@#sx zpBH)5`0($-<+lK!TAWUd_s3j9A-AP)g<(JxD+TPY(FGTC?KE)NWG#3!e?C(JXDaW% zjC|KGVqsxXQ#$)*sh;LY$4aQ=Yxw8^=5IO0Q2z`O-uJ(C3&ZYm*|Y(h7iX3#>>5g* z{iRR+!;${S^P`Tjd$Z-K76Y(oC@JHRdJh2?2SFqEliZ|T1Uh4!XGb}%TtyV*9j~*H zoH(7nIjd~{HwAIFM(UT1@>lTjV;bL?-+Z?5BvM)l)a69>>c@|xZd^QOrFULz16O+< zSjc>vC;gio+!%@N@82tZ*y!ZVE((y@k@3ytAEfaW#NF3XT51D>rI&h%e6pS?oMG{| zMp0PU{s9-zIc2v1Gq``wq4n%nti&6+>HXn%xn@@mu3Buto_U|&Ig{>xW5GXrc|+}6 zjxl~K{rU5+Ptg*9hvc3~LiR)MC}CIkRh5>W4VzTEGx5P<*?j(`P<|Ga|MH%cDb`|d zPBh*J;n&{ai`h5Ow;%nO{M>n;+Z)jI{@rD+neo6!J~_{9&KCDuqbPrDHWR$~e%OT@ zV8D`OY&kld!Qw-VnH*^Gpjp`h`^^7!f6l)%s0e`hUg@V#YsBi@fxw`xyY-ASj`bQ1 z{8i?gm}U1X?EbgOd$Z)J?gMyn-v#gzmlk^_s{07vhKt!SAB_4#9{vprzrM_}^D9VT zr-cLhS!{NG=8Q-u%zzV%Tqpdypq}>uQUkLruKsRk5=;3Fa4~>g_bzAN&`%%!3ec|~ z+ygYDn~)*#Ppdk4003sNc5(a91mMrquMglW&aS(@Iyi$O3ll;x9PT-8o-O>hMub;^ zAC@*|{G86I(!n^%x92^7YlgE`N2y^FJ_EchwP)P;AH4UUr0h&2{?BLwu(S*mlx!D& z!q^oWz=e5hO!9X<^_V1pD`s3ItMC1IjGr-PF0Ba#<}=^w_gMLc5+Ki5Q*4ia0&J6r zlAg#w_>|XsqQ;rcq)pA^+45i^$1)K}?Qt;?qxz#_CLQ2VwG#i|rdqSD*&DX&Q%%pi=PnONPuZw`!4_b0|YqP=U#xm z{qmG;9H1Ea8F^h+qNpSZ+F`L~)Ru!CF^n2E3>uXM7cXIdUVRYTs;t3m5TAROgH1!V z!JFTFtWuAyy6jc^9d^@#j+CRqnt3gg<}i{NCLPW+J%h-56SHYPoU)iU=B*h9fxA$@ zKT`@nq2r%lY2?DfoPnSQQ#mkwL=hpP)E4{(@261Sw-~bq?Z@9+EQgA`{MBL{y2dF! zs^AdZi_t1K{Zi|O#mo z;`G?F0*(=RU-+_6rz%E1i8INuE16q0}>B>~a zG^??}!Y!H5+k0*S0Rgjd8ro9XAw<;Fs-`35W|3Bt$B}YLoSkd29G2<_8%?b-f<6Ks z1~Qt`3+ZY{d*r}gxdy!1b)a7orkYz{Sb*U^jwP<9TOc*5t=abFK1X0p40Hj%y(U5y zze1f#snApJQda;m+17|k;IJ8&gUuES;}8B69mCE`gVE>FIG>jJ>Yuzxhp^lt`6*A} zcku=g3JaB7d9<=xKx-m@>Y@=>gK5+Q&I+lt9y85R%8_GmTQ1xLj-q(QL38ActVj1E zTop&^JR(OlDzrHg9(mVJq~1`E@lXQYU%wAznfgtv{x9#Jsh?j)&*&bxtUNi`8?)^Y z?#oq#PJIqd0vZ{XL~Tool9RL6TS50 z+i-=&i5qKDz}eo1o024KPI<~h;0)956BRb&FAKZO$$V}qq>rCJt@7DwrBf-;Wc3)U z&JfiMCZLF(Zi#FKj)$W0*dvSKvb`rAOB-IE)P6WC3!Q21ybF&}A-#f!_t?6otw)_` z23KXQrdRo?QdB$ZAs^6-5T7X-DYZ84CP$M62zK>ODJNfq&uqvFvcUj2gNSCN%tRGj zwly|i8BwlpbQ97$&ttcdPQs-s^=bx8lvs-q=x{Vhhb$rOdyAPEHlt&&ljvu zL{kOR$i_c(2rg4B1|^pMg*5yurys8@tQ+&I<;Qai1BaNjzF{?5jkQMBPh89{AfG_E zLW8c|)lS;W$5%sdw%{I^^+|AN)i-28S4Kp#pZ8>_6pr%GAv9g)Ulaipp`#2wb5JRx zq5#8oLuJU-NRrqzWHq563gWD*cE12XvveRXQ9;p_!_iDhSFK8`YVMZC^A~1Hd7Lx& zk-Y1*R0f`#fr~@+qB!PTAFndEEHs?+IXNKS8Zk?2TkOl#8+E#qZ$I6<)ho*ad!%Lo z<*ZXixL99NeS2T8&|)K97$E#ptf@UIOp|_bKuCxkH+=?Z*qZrkt z{qFg9hvcLEOvwm{qTrESj$Fc2uObDqK#QGSk}=QN`IWTN(#ODUwu4NDt$}3jLBqXe zQG$UOKsXW~K`XY$tIJh#74$6ork5cM6w)x2Y`xKVmQzuIiVZ4DU<`VnqOXhuZ9PC0>HWM?8xoMoshs5+%7-@zXSXnc1P`nV5h#)nw zP8?!4>ym1TAgA(jZ5>(s|JeJ^uqLzZ-x(DdLBR&7C|Ici0s_)O6a=M8k*-v!p-2so zK|~Y;lqwxWr1uhf1f(~mmq6%93!OkhlDji!#_+$#8g%g#L#6HxoVKZ)c{kRdR<7~JsG(W!J*>eT7UYTBy8xfB&nQ7DJ z#vz5y2v}YPv}}*Ep-gEB=gfSnZea@8?cb@8el9_>CFq(<8^+`Pbnnd zIx9=3ZzFZhRpx8_m_h~(z2=;loYWuY!iWT>%*SrxFAn}RlRNt2&Bu@{ci$@~yjYf) zGwk!gjF`6DV;RmPc4lJU!7eysH|C~4C&{P*RvRTYxEZO6SsGUM!f!G-z9OUym@ac) zN=h13C;F_l@*tuo!9TPBLwfba2n8F-171mB(e;G6?eruzV z8FtI%&eNiWjN2rrhZ#U0MEB3m9e{JP6&p9wxa~|wwx8TEv}U;aBJJ{&g%a-wrdP|b zSr<}eTE5V7FYt^wmqv!VfOn(-jKGd+Pk?B_x*nxQYM8Rm8c{@Ld2TNBhc+^0CKGC7 zm3*Tk-YRHl#%yLZrt77@(<^h(py%F|pXnryS{}#b1>?4{?ki>pyzJWo8l52pU!4k7e#T$a@4k)6N^)^>dSfqPRCs-sXy=P z>InhIsYcM`9@JTSWM8v5+ceVw2U@Dzx6dXRC!Z1GwgJhfWrh{FO}Ci-B`SqE#H;zW zM)!2pNr_DcRQ~9PlgW4@0SDp?``XY81{Cr_)~IUzXd#F#X2*42on5uyebv%oJ^}0P zq9@Cv?O-5ixaJVHc2Bh@WyP6Z z0F81oEN=tBBfUYq<+?R!-VV|Gg2QeuZ~_Dhk8aOgub#iw_3n0n(XQlKkRst>4;cY* z+N?@p3gH}XaTT`P3i#*Im2PXL8;|xN9Wp=rertktWwr6XX-inT*=x7;8Tp{I!tb>@ znF=;ArfQjeb$r$XDyMHeetlIFCEnEvkXk!WYu^F`cWk5_Gf0eA2xZ}ULFl-DlLrVp zowuuGCGO}8KPd-V1&cEnsn^hlO(zWXf>y5wS#H4g-1;k-XN|ar77|ZTnTQ$bpWVn$p!~)+gt{O5)P!$AU;<5 z!|ecig@#uboMQ|I3)~8etb1D)IrU0kdtm3x%|>>NYo0EfZm%3O(@K!2($Gjf=v?Ty z>%ISAld;qMfK3P8tl?U!asoGi#zO5NRBiMta&Y$`-H1IXZ9_t@v#&uPl85GdvM=q& z(%Q|P&{whS{t(zz?!@SLugcv?LnQ;zy4v@gUuCPF!BD|_pUVS9)oOQYg`J5}kZzZ) zQyfOmrQ;skYN8=;g)N%=R%D#wQ%JaX}=ay*C9oOfTYAx|#Rh3jv0H9umO-_Xw(|MbDJs#~#w`>~iv5vPS{ z76{2$pWBMZpdeV2_Gk_Is}CL?p_?nV=w>N^g7B#rpw&nXn+n|Cndf4gj@8mz&IWk}hZD+_AFt2A-fLuP zwWGBR+-@6q3K)%4G5>-I07U1Udmer5(Np2=OPc|K(2cL`fS4kxJ#lyN*x(qvxciex zP)H#lby=o8PE@Pm)yr+u!M?+26+~N^Ww_-U(y`NR_e!@WqO|gOn18^T#yzOjF6Mxm zf{KD;Ips4k-xQ;+S;3Jrk==l)8}w!CE);Z@jeBwRIk^rmiwTx*Qqgg=yq8~=?dKX@ zg5ftKpL{*4x>f4F2{lEQ3QY>kICe_1;xJ1>#=Ah>2obc=37#A&wU@E%NqaPeBp$b+ zh{{mU=710}4ED3<0Vt#95zC!HeHpawi_?e%dThpXEw?XCkV6n9{i}iZPxz)>y zKW(i)PeFYH_NIvD#@w0e)UCP=adi)%LO-2|FPI?9W$FAGYouryMk{KSs4q}4pZlLhDr z1+-S4m2yInKq>Nl20W6igib7}m zaOFs@uwehVP20e$W1-cNJo+Izw35~?__wLvMdS)M9y0UW%oKjJXP;{VgO(Z+T55N>dg$pyavXiwB0tq(sUt$t7`(WOey$yrE~*5!kU5RwFRYZjX6a2MDf_+DAbHp z6gcJR&X()fBn}{L^v`7gwSiaUlP5t{-|F7MC(Jx7jh~8@mkp|>;3;?8$33wdnh1)i zK0~jnz_Q1y7o!9mBP_gTq8jFT+LYgJDnTkQz;X<1`T#+)1JSv8sXk^~g2)Fp`Y~2_xb^z@A*V+TGk5 z{Pn3?f(PGUYMx~^T7}yqizmoq&=-Bops<$gAkBaY) z_h)*(|5MhfYwz9}?W0ruEF*lvRrVx>^k-rml~K@(^A|acE~c5+UY*H$OR-<}=+SXh z?*ij!k27smRbd5xA!5hFHnSYtn~8?>2f4%s$CIE&>r@%x`NDLcuI-WF<4 z8NHm2S_t49#-JSK6Zl*Ys&$oRXd|4&rlGKej_U%DIX06tbng zQ-9WLu z4sSv8KdK!m*c!6(#ShC;-WKbRpj(~dusbbk8>=VaU~;3es(9|%38g=9XgMc(g*c}^ z?Ji(tY#SP~&kdBAT`ZhxPK~GWlcmHIk3VO8jxW8*?$SjsJdLgU-5h?+HB%k$zG)IT zW~01i>?pRoQdcmaF$mdqUn=gO`gB71aOxD+=NTYm{pti!p;uPrEgbm}NP$&_mtK;( zM0_?^mAT6W+p zBPR89S~Q(R>;lg?0^``q1%K8ESXQVEjqqhF+oAnbDP1kRw0l^V<iu-Nlw_Ey=&5lFkbBhzOkcUNTQ zj#c^(dG=|$oI4Gh-OkIE?2uS!yXW0^CGv1+&^edB`;Qhz^`;4bj>kZ%4+SerOrM-{ z{&Y;67ae8b(5&LNQsY;<8o*Qiz%L_@ z=e|a%L*fn#xvld!UrDZlmU2#q7E9=B($&7xz;o;=xPoOT)#qsK*;Xji)4D z+VVUt>a4e|U~=RY2cP?8HjXpbOS4>C@U+w7B2FP&()sGZMK+xhX%GPn(k2eEZZlC1 zW8RdO<$2j?a5@5Ui-nLp2I2*hO!#i*5jFI0(57(z-9%W@|9I9DGYj|_wyJn zO_5YHKG@TRsCy8%8(a@-V$4NwBtw{!(`ymlrpumN&1Wk*(p0sY&$+xh(x{goy*}tV z*(Y72T`>kR&}KhN1S-MQTjQWZIJ6(-T6zZyR$1n}K_kKIK2kn1P z@bCF8E$mO+-6*+NXW3mSU}Z)4(up)_pM%ZVgg6wxUM=O>U5MxeFfS7&G z1ua$xUC`&-I6nB$=g7s*U?xxKiqK2F6K+bB(5*#w)>ylzOlCv#9AWc$n|EOg`*_|(0v3299jCYXO0#-=r zxVuHSSz+7fyFmD+9)N_lY})&3f|UNM-OXI9S*li%$V`CrTwepCKr9{g{Doz__&dfq9oRgY0eiTdtqZxznO(HzpyhB|)6i=t95~jh zcCia$7Jb%7w5_ruUe0g}l`%*qg;CaC*kj-Ky)2caenC9Yq<@0S%&f!Xind)#8aM#H zkP`~R;ym5eL<01*Bthvk|4dMmN6v~FX6%SO>*n00z%xQYufEh!GQGVD4?cY<0k%~} z}gewZDT=x8T+1 zom8NkUC*31N(Z@bCl9}ZT6JY9n>8;EI+_q5>n82=T=i87WgwS{IP65Q?-szF(H=^>r^GPQC;IA4|`OzY_q z@Y|EYU1_Su+K%n?T|ia+wP0teP| z)^Rq6e2V>V{dqI}qMOR7Daq}K-Qpr{UF6BC^+PMFRY5T%KY*qy_X!T}qZ(`)t5;Km z>*$;TTre2d%|ph-eNc?iLc^g?u)lpLm0`xbI4*aV8$nphEJ9^e9WH;Y z(r$3iJBu*x$E*;{Bo2-uOP5(;`&6lMa5#{m1yVUdt>4tSe+EvyB>ewutBer8xwiPyn<9PP6$ z`ea9T*bA`ZSJuv#9r?vcra6%0XzW{xJ)gcqIuGumW_u|1KT+BqLjO~0={VqZ&f*X> z{kJtpY&rnOT0q$_{i~e?9sA3L^p$A4yIf1XHG@%P;Gq}1;u zQv68kclQS|eayEQSO2L6@NGhS_`Ua2(tHz{C>eanW0WcZ8!^aqmb^84mx%n_zqRZE zo2YNT7DWDs)bD?sR7mPkMzlo&gT=`||A^c(1M*`(YK1YR>B4ie${j-2STpRW7KBm0 znZ-JA(-h6@!t2}W9Q#A}&>owV_f7UNtSMoi_AYK>-^q!PJv8xiNB<4>$b4&y;+bAH z3c!F`Iw#0G<;$(Y=Bq}{z#XzfO;0Dn&`%bKKR2yn^+fsP=LNM?N31r`01v8q(UZd| z*`EA0_Q~KYYwNk|)Nug|KiJssZl(@LUWkLirPz($rr+EAjnvo8 zdnDBN2K?{8rD^QDYuX#G6;keGMLmRgU!@Qp#nZG+J%Xw#{+!qsBb)PMqIIAf8jYE4MKh8FFliid6$hN?B)kynf(Oo1g z;_8?CML+2FyQf7e<-mMjHmB1G5Cf?R+G7v+T(YGILT{0R3ldf~qjT z*caPz5`37FwvQrjrobg2S=&nsIewLvR8&d0hC`MdRw$KCl0KT^wtG(ak5}rS|N4_Q z@}ifkp2)XRNcEEB;bueDcZusq5@t#UbmqYpLiPqHnvgbRP_8wQydgiuT%=x-M7|Mi z*f3eYYRU!sv9o!Z@!L>;G?+b~QPBXTsS$qFjZ9(+H^ERtTwfvghFO#M=0CT3_JPH> ztdW_aCwJBb7A$-;XwTmM=lAyv>rW{sDlo~op)7K6s>&16DH^#n`IOxKzmfSro45BL zi4ml}?lOzP1HY}x>w935CQ7${aqWNGC%^?rQ&IBwzCA+qF_9i%*lQeWAyL@qvtz$_ilhL;2juiAwN`*UrIW7!RZy(6l^*@o-p;3jH-+79kSidl z43p0HlIWi}^2C3NH+mZ#0|)Ry{K@`3om~Q3+53lRz%QcR-}cebgU)JA?vX#8rbz1R zLuz}~FLoxyOIqqS|69Dw2 zu=LVFl1xu-KoD@`#$NtKxCXR35}|V>Um?e~ORYTdl#;(gycf#3NpO%uGB5gt=dKB@kSV5mk=BR!sP-II^Xk6JyXzAP6) zXRFDa;W%KD;V)jE|HZYJN^Bs}`KsJU3i3{V#~AXx_4&?k52FALv|jy6K9lEFl)x+< zE*~a$|C22WcCh-=67<~Ux(bOZB(*V&wZ`cu`ydlk8JK3vQl0j<_3_pr>Fivn-;ff& zxcnyr!6a|5rNJ=yH{5l~aH7I@SB1 zKc3zhBugkKU;jQEr^^Aa_f+@(2n?n?qE1MGHjnnSOp{1`=bWb+cgURjzXBZ% z5>;xnLjsyDD z%3YpN(T35}P`5Z|l)h|Gso26UKfhU1uq5GZP}SO6E8kYf>Z634=~r23Eb{?r6mO3^ z?oD$3A>m&k;mBq&1=~G}5eGOj8C1Q|?K{AX-SY7QkrhF#E=BsDlu7Ebz++p(*t=`3 zP#)`oT}vx<)5|a-ROP^CawP%F=&*D|sA+51Rx3xPWi$)Q*lU6B_`^@3N@YH$0v!RV z#%l_qtoi2#>{(l-n7yw8o>Xw-#&^h)%<=ZN64^-0*3#H;o2+cAlVEo56@AZzHtYuf zX};YSyPD#DTXua~q;i8|sa7?9MwETyX1Wl{U6)3D@x{sz9A&F}nse@dk;{1kHZZAj{aE*I;Y{2Mm2% zy;i5S9EJxDIXqT5U%V;{qt+6ct-b62PGx@TzEb2|u*4aI^)NZ6FsFWP)^PRco9S-r zv#o{v@uXODLy6sO8SPBdyMv{h4P6%Szud;JDuY0fi*`H|grH-pFeV@fuwXm5zM$qH zEHqGPH^#cKy|184fipY?{`kv9x9xGc+Ets95!o}@pYO!O#wpz9QX;4N4QHEKvUTr9 za#P;v{*W0Fvv3gUB(c8n%3-!o-{st(Noe16!-$k={}W+}?UhSWt8Itf)|cQUuflu% z;gW)GCDDWZ;Lb(hOfDF8Ur<>1$jO|*n&7VEwKJ_{(QE0sgZmH@c6Tm*c|tOx$tPA4D<8C;Py0a;sy@@1z{*I``FW`(7eUa8 z_uMM`xNuuLN%hiTr6S`X1kuq|YM-2SSD}LTy8L6XHfvw}m|NuvB|;hk&)m)2><^Ly;r+ zWpS`@dWXR<%msC*TF9r25pR<`mhBy5wq(zki*LzF7v8Hl|KK2ctC6JD$@q&HGP_eT z1sbBar#-+FESOIWj&Z!Ix8HGzT(1bDCR^}tixt`v9FKUYrC}H}~|Z2XgV8{+A~AkMClMk{04qf-PIn^-zm=HM2cHE`?sk<%BUi}}zp zN87Q0L@Co?am=C45!bq62W2~2T8dGE`PX`%sYIWac%1Mo&1)RX2u}=Tl!Z;nf+bj- zAStPrbJ|H`rLQ=yuz|CX*3uGTJ|b1?sC6u3){yx+Q#pP1EWE1I-ydTPHY? z=d`=MStTFt;}!dc*;tqOHax-ap3<-|t_{}8+j!Q-Z)zBV2wkV731f|S<*?GDC9Eg3 znxLvMU3K>s{4+bKy|mC(5H+*5Wu$YfV!>wLx!p%@rZbriyOOx3UAOhtILG}G_WmoJMZqb#~I`|ZLtaO3HeW6CY#X>&-N+n6U+pbVX`SyM zbk6|uIZPo31LzEhSYu)%=V?RFoN5e5Q~xB7~67oN00tb(AuZI?|s!oc;}*X zq{f?)-Jf+MtC;Kuk_vRXKkzwWHxlp6b17=s`KqC72?bb7hp#pIbwa<5NV+*Z@R)Ak zsT}hV1|cu{@GfR8I|o-d`gA~?%aC-5B?36mMYYh;H+b*gJcTbGX1cCIGdCnqX}}fM6#zL=8EBp>jtNWfBFHxcA~*|Y z-a}c<=;k%dmV0}=`k`5Otf*6vmDh1npwBOzAm`*b_j0^)rse`Ag*xh4y#;O=rOkcL zS>83L04L>HXHpZKL&j<9RtP}>-nMb00&F1f}jFhsY&Wsl76>NwgzsLPQL zQ%78AN3Zp9yD|=oMIBuf*ZMjwp-Q2Ag5_QqHFqVO7{Y!b&;(}#rykWSuJ!-|p+<+l zb~z~qQfW{!Y&(2S76!sE)D;e>Dy|LVZEczD3_9C>%j?TMfH2Ru??=CpV?9pVxQ@XI`|x!imA;Zd>34mx()G zQu4L6YVhm(P(qaOGD9X3WRVnE;euCoCnO1fP<>MRgW4GRaTrn^p>gfCypf3CS@+UP zK;th_Y||e!O%yTrkcIXVlZ}DlM~Ip;g%4|Ys`$mtgX;0vR zqN|i2cuZ~0Z1x+(X^hR})J(fj==d=?)YA8K8bC}Sh^)}1;@JfA%WM(Y z+ccu@pbyS|t?ouwK#cmlBzMBPO|i+r&k_B~VuDpjm-jd&i;{*jU6pQy_mukG*xAmA zmdkb8WfA966_!VdXl=R1x7QltO?(~)=Mr>gP5XLe> zxs^%kIhQ-_QAH~Az6^#T#_kbSWtIbZ7rIP|_||MUcOw7P8@Uh9DJXcviPj(f_@I|4 zWT&ojZ*25d;c!kxQ(C61Ne2*oo}lb3!ErDp$}xwZ-F5xTpm}7K{CreuLX&%dhUQoT zB^_5-MpJaoDWHLf`f_eA+K0tD^+jn%j*U8WXJQVGKS-ieh;wi-6wdD~*eBrfrcZF* zMp#=Ya%HT@3l9QjEgo&GcS>+&W));@*Bzl@_Oh61H|hQ$aTTTyUCL@JQE8z|7MJj} zEIX@x;^}@}HmBiiq1L{0B`s8^?Fdp)uANom1FbybJOE|T;C4o#&R^z-mYv&6tGU7e z*0E0NSxSZ+89m;|ZChpss_3T>J2BHn(g`_X7$fL(g$&B9exzIeU?m{9N-SGtTEX z(}!&<^q=dyYu|Y=crgM9g~^7%NNEPzL;X)6uXL7pV8<{K>=^CxC-MW;_zux=#&Bd! z4?8UdByz{#m+|DL!Sq5X`a3| zrYBr5d<<;$dM!0i3vXjYRmyZtbEt{W5gpSw3~kZacoSN#>Cg3!O=Z7Ei+&H1oSsq@ zuRRM#KJl@xSQ6CU8DY#5M7^`Qq50TX?F72~zM$1QCTjstz%83L#d&_A&&NMDqB+pk3Qni}V zldUVnX^q`zoSM3?Ox?)jKIR1wJw7kNnfX~LaaXk)H)3#C4@Iw#G6Dj$rbj27Fwa)) z@0ya(%=5!lw;GVV+*cbs#s|&c8$1+yL7B5kr%|m3t(IQc}mkS=dt5@eLVUa6E(EH&~A05B!5`d&z zM<3G*b<&4yiqCy&D=v4%=YaH%a&T2?sA{0K8QJh1u_w_@3ax(m#{Jv@w~4l+a{th^=`mS zdtpvLUB1nh`J+hAzHKek36M-7{1~xW!Ur4mu=jPIO)%VIxfWyIm!)ii!naVV?;6oD zm{X^O&_gVedPHFQqHLqk0d9sZ8}Aq{k*u~+&M>?I+U(LnKng>iiN0$8fQL^UgEHZR zl;~`0^pdrFKmoF)uom%gfU*jKGY>MOFaGb}k+f}>c!>t;HCr-U4fvez{| z^QN(Yhzf(oAkkKkYQfYHC%JO542ORC2t}9m1YHSv@V2|ue$s9+$bNDP-J z5KHJVeWo7&JjbAEWE}sjUZaTm?IiWveX%!InNHi)_>ONXx=l%IyTq+QV%d9crq7y* zaj8-;#*_?RtV~D_T4b4xgDMRu=a!-E1Us^XPp8p zccfUWwL^62t&_w&z;zQ;R-+yZHo_Zu1w!HaZaXm3PH|CR3ywkzl_^5bZQHm+g=0zL z^Ud^0C^IWwi3chgi3noSa%TQIoK^y?9A$G23j#JBTj-Z3rgy1ZstXBj9uRF8VTcdN z3{)(8*NTzcR>`TJIRk3H=#&*BV^*HiGG4UGDKjf`1O*D-G%^ztUY!!)@Q8^%^0>h0 zgz)4>cZx2&ZYl5ag{a}ekt+CCQ;LI>PV+r)_R}sNFLCE?j~0m}hY9>=-2hm^eqh3< zh{*y#ZN_6gaIwHjTUxFQUi!(K6~q}@_6YFJ&s)`A&u#aR2en5|Me0=k+sA|65 zkR&ybO`rAFICz1pQ=E8Hl|ezA-}w24M76T3;bcS58<@4WrilqeojQ1JN0!|acdygf zy-C|V{!wHo%1ihBQpu3WS;lK)$gw3MQwnL@kMvDq;1<+^crCF4yYZVlauqb=NJ44? zM-`L@TBmg5?Svn_|GASVoe!VmE7f}rtc1!m^X`vtNKE**0(H>2(?1{RzjqqwTCXlA z1MT66RHxX^t+JKbc&I_pl9R^`ThvoCjLmJTnZ?TJ(dB7dL8J4`Bsj0M==e0&VGVIn zvzaNQP2xs2xB%q=JNn3@LC)}V;*6Ev^QB{o>S$xs8BOOBBgv8B+R;N>bC-hey{SM~ zi+adJx7t07+&>kP#4vl0RF}n#P|q+O{}NnpBb}Mx$5d*lQ>Chijf|@FRjRUaeD_ zAGl3bMYyU~YPEY~(4$aL`vUq1|FAV&=7F6`8$iaLwRBu%ogddbLoWq2dEIRz*_5Em z-NIrCoj21jm7k-l)Lw9U$svK==;WNocK2;uP;1N7>9!fsnj7erhrb-dB$_ zxUoAOX{eAY-nzAXdq2FQQM;<1aS}m#ri9Z;#lW0 zlWo)Sxkoj_Y2}Lu7|WB7JNr#1#4-MyXz1E%BZxF|k8%v3o6;%dbBp@PJc#c&Y`o8q z6!r#Y>0&)8Ao7I&&{aRaDR_Zlk}5+2w;SL9p>~+vSO>O}R&UBkI+`hti`$Fqou;`} ztyG>g!tXN=S3getz}8bcSkRakhK`@|rW}>}i^5jxc|B&OuNq zrbR=`@LWf`hLm}~t^sUdW!z!-tEiD9p|JDi+SL8dXS2!vL=Igx0oTt`!tbs6vs!!L zn9_G)tWV#a9D!R_w=LoEW{e`PnBl#S4#pb8)MJlkX#=7~z@3+6F&AhaSk+v&y{-N6YyR?kCfgCJgm9tHap^CYO=Bw9!n|Ilyo>-k`)n+C+@BI#eI(xOk?UOh2Y zD{YDyvA~R^pnyKt{c!@s3%{vhpmn-1&I%}@Uk@7h>Xa}?8?@%n_4h#P)a@%HS+-p7 z4R+x(32V0s)lum63MW#lto|w0`>}&?XUwc;p5*XI%XOn?fjQWkI5r762p2RhVEj2 znSjqeLClT|v3t67vV{tZv>&>eesgFUEfIkM{GD5{T_O$W#hyVHTha-kSpQ)ZdcAWJ zdN)CW^-l6wMT;j%jO}6&TBKr0?_mg26c_hsyCX9^mFEWRPm-Z`Jb##a`5;B|iRE}U zWpRAby|^_ypiKZRy=2%-cg|squ+}d*dsVMg#jMR$QwiJ8{%F7ex^zKn7b7Wh#af?t zp(>Q6`Iz517SYALDiG;>{wzo-%bk8#!(fhnG%@1Rj1n5T%-@ffcp|#S<~A7!|Fkr? z-XQ_5u->(m+P1Sxi&dC- zhaDI}nRXh~-w7XW#JuS@fQj?ZjDDVwK-+L(?IBY3kegpR>t&uFB&reP+)q{aPUD!p z{>bgtU!Kp*rtj!9PwJtSB(MR)xuaRTURuF>eTc1T(9jz^;bWor(#-9N(~5r z4~XyVL?)P`F_Kf8foeI|fUX_6ba!@6!j`tQ_PUn!xCF@MdD2;0qMA%sYjv2FLK1yO zQb`b%709GnUNzx=Oxn9EFiOmpI8Zg|zsYjIO3QrL{W)$ZB?PZa8e z%MlmlQ#?Qw_kaGLJIvvbr4hoMyYlI|zsxM^Mc{-OnQnPQEA*E~SxuH0ak%L<4BE+B zGbxObQOxzlt18FJh3Cu|`utu^eMOSR``mG%zqFnpeYxe%%{bM}&Aw3J+-y8&CV?4o zgcsRX+0P6(u_(r01)16gW^MYD%r!Byi(bJvFJ}l=6umuJ=*6I~I!G&te=Q(rvT|G7 zS1nwd-!bXQkHGjhhvsh}01?QJXG|<`a~~y~ z7nFzdjyC$4C6yiI*%_}vQ@$NWMXWSkHzx0hnTH zv^4^1nsk~*ng!{b+pm;_et`!@sMf^1vX+iGxRgK3)U}i+V=B5{!t4@wN%qxiUAf>d zd3jF0`y*+XSzHuLbVOHpsj9A{f-M3;^;8A3PNfI~91O}HPN4Wa40 zOXYsXq3Ad^Mo6XV%q}wVsFmM2QL`XNC%vRs)we(JS(OXKc7ogyo?tP)D_1|-`O*Qt zlN{?clpZw^BtaB2P!gY>(u-72a#WVsEQ%c^zRH^YQtc^v+J1Wa^nTCbFqRSuzRAXs zW0ixM32jmQa>{=7)|ce;v~#vEoltoh#5r^7y|$A?PL`k*-9wpO*%mM|`;oEySWGC)y3Cixbcp0v1sX3^)as&R6z@{$ zvabO!QX2cfQw401cl-*6NN&pKhVD^FDC&Kdz025)%ONVY;JGg{eY%j5ww;!oDEV7I zYui0Wvh7YuJ1~-gsRqe^-CG(W0NV#D0?-6FYCcGMlhx~>2r|}+*CLUvMNj6`c4mel z;HtxfCXgoksVXHERJzC~b|GTWqifeWUl^7?^tLXTV<8#*4f?u7p46WOHm(SeefDA< zi!;-GF63&ja=52>p~kIBTPt>bHA{P~TUoN%%!UxZ>HI#;*V?lrUrPUr)K!`nN~xeca8MzF>>NTS-e`$t}lct zPt87Lz7FZTid=3TrbgqOwqhYusm!nt4IiWGh#D8`kYeQW%&eiuvC5RiiZSTJmsw?I zD`%B6I>7g*wZ_E-X{wb$c{L^j2*|1~KsPg~J1fQn{A7TON*mC;>DId^_f)sa z&;uZbgVhY!+cd|@X5%A`q34)KQD4W}_0NeE<$i$%a=cFeLY?uh9uJ<=`S7^AFTzu$ zi1r0FImqmXX1t?mVZg5*K4Hv9MoHX)^wSA2ni&t>yLil1e&bFZ8Os3x< z6i}n*`g36`no&F{?I@w}iiY4&;11UG%2zN+N^|V$w_bl}GWt_0 zofI>fp&WrD-zm@(07mBa=<=a`SGe*lVz~G){U-G+k6?t#ov8f3twe8xbJCS;*k329 zaA0UMxp3#<`^?EfSwCjC3K}GU=vjaySHP4q09NW2>PzxR*(|)JeRv{3Ek+g(XxL~P z=i8x2e&?pL__57HR=L@A0z*pWL@!}n?C;|Om4x}$u0&~{Gga{!Iv#)P6D=jf)*wO4 z!&rmsevzA1s|4E+nw}K1 z!gfUKfi$6D$a3fIy}zyx`u6w_%?02Bo?C#{t{7b;*E&foA?<)}f*6Jzwfc|Ghs+_q z*QzJwQ00sB=8$V5C+dK~W@y^d|0Gr?ZlW_ddN?6%{zwqdPPMVE^ z#No!-9cam&X^_Sm{v~8j@sgigvZhQDF-~7T{f(fd(n-pMrE(bVCoSgxOwwBo1XpIL zCtSslgNiKonMZOPi!Hrw1W{n`INJnfxi5A=Iocq^Zu!pN(?DhNl`flN-i^; zmIvld$6`#rGvdbw`P+Y0{ypaQBY4#O63KK^G@6!|J8(*-@Mnro8dx_$N^u!d6S}`Pg9Bf)RONx zVgKY|`UQ{=O_~MRN=xjayq^>yE^=LG>!I(i)I9D)Mi)uY$ggNs4j*LY>sVMq(Kb`YN8ccPu1#Mn_^s?~Hyu* zW>h}!1JQ5cq(TEZUnKiZ5K6^MJQA1&K2Yvt7d9`jc~fPRT2| z`#>^MX}xhNf2qRNWlMM*fkBIri+Q8S7eh97Is`vNt_E2ja_vpC_T3j%m!S)#_yizr>d?jxF+i(Y9 ziwL_*Q`_UqbTFPdbr_OqxH`b45nV?xCi*g7XrzVZZbRi$v6EW^Tl$rv&dgiPI3DC5 z!d~0ZdemK!-fsiSBM!jEWt&#(P@DvVuJmt{^20Y0sD*9@F%t|{F(FEu-zK&wxn=l& zXbLHX#9GQ0T5{O*zhw~=%X}o(^|v6MqV+6hKIt2G>1_q7$r4tcYWy`RpWumY+c%tU=_`d~o+-1(Wme6x*}R#NEjR7$|&6DE#8dI$(@%c){4eKBT!J>I)7`A&$#7p+0b-x;p}| zl@I%{3VLHkz*-4#RU`-oD)m}^+=`(^Nz&~oW#7rdi}&JZ;H2|(drSSc(Onzj4np+o z9-(Fcrq;q3^6h%SwBmiLu+pV#8^qWzn%;}~OA9;xCER|vZP?L_vf|!KaMm|H)ETQ) zIUb}=N4K}cISV6GIr@3vX{R50?he9s39IOnS zh07C&MdmEITM}zr}*5*3l60uhGvclLUY+{#sYRMZuq5d+K$AWI??Aw zbcg1J0poM8ewW=kE&&!MVMMeXUexRO*Tin?M%H-(z9;#!1MO`iWPV-0$n;GbpC>J- zD`N*)+@AU!FQY0nMXX6<48Z#Ir|~jmGPj=&S2ix{ec*DIj=5N~=k1JR_ROo6RQpFs z>UMc%>%C5CI{p2(H+4qrdp2c9tp0hxe9}5A>!2d~pIQJ5a`kfAbOL7U1zsD(FAoEY zay8MXh>Jwy+NfJg#eaQCAaJecl#M{&xqm(SVMi}hIC*XG2IB%&byd~scHm%TR_C%I z(yIXXY11k9Q@tGN#TYF|*(GzvS?rmA9xL0yeu-7xxh|_4=+yYHH!URkEj!XY1Unka zz0^H=bX1K6c^1}j%(wIZ*N1cGRfYet{w@2@(}z+Bw_uEE-%{d94HP3(fzKab7KEXX;^PIv$J<%0dc zxLieuN{=&gWcQznJ-~t-whf2>k)4TC6kP=G*f?_>{Z(6J{a0a|rMKyHR@nPV8to%)a-MG!V_5enZZ+%!SUS@n~~xtP#p6& zUd9BjI)Z*y8&PNq7-G9ouy4yo4_P{5CD^cK6Q5QcSnSEjDAa*%-6|I-@LV|63(qfs z7*95=HX34{mpCgAh@T{XKMP^4jMY?alt)^>N|5Qa2#_Q=^K9elB}>3dlM@X~28FyO zko8-eHd#x<1EuzTh9&*kZ zRY}<+C!A^PH$&SBdA*p#Q@Iv+hC{10dhCx2F;32Yb{v`rMvPDsyjFurgDWkyujsda zwKOcI>d9z@6BM+K-HcpvvNFN-@&5Cj#+~7^4YDSfL6T7!g;)6hQPu$>CopgAG+IE! zESMxaBs*rYSXp*Z121=$dDao1rblom*!RR}?PK3ws{=;5=^Cndzo8B7#$AGGyFO?&CYOQlFzXaW7jWV!A zmIgKpNY2O3^aygut`DC&7hvgFzcah>V0c~+l|3hcrdyTo5B8@GeajquSJy=WH#Gk? zir&C4NUo}6wB3oB=w2tiF4OOZNeANB&amzc~C_kN>@;Ja`)7T%aCPkT&C5~Dd=cmMR3A!LuqwKzhM8{ zxK`{UtU_yX^s{|`0dada7jOG=h?T(f#nooXt82%DxPni{-NZaaA35iejeuLfYS`f| zUoPssA*$71#Jg3==)x=s5487;%eBej6^+~+Bs%8~R3%O3m=aez4RIsKinV} z;oq5D6}WEv1Z9zF@0^okVZ7)jgqHBKWt_x{7!K%Y<#{t;ynIoc!!M{yslS(!Nt6A)sVyaPEP63}=UAYU5D3 z2jH-cfx~b0){{Fdlx&iXt|J1!RbKUoS4Zmx6UcFy-pa`_h z-6MIwzA()PKHH?td{g~KwK^K%oN^<-fpJ{uL0j_gXr%9PO~93Q#=u9Z-z8y=L`=_p znvI3?8>xm(zli;dvutp5X{qSfoE!Q9@5&_FFTs?-GBl$d^g*U2J*TrO-+TUu{MGHx z{|yH|OtEi+Q1l3wv6B1W$sps{>v zEKrbxUve(c7Q&th_f&A{grlOv;4%V(L_K;<6&PwSSKV-co(DB( zHD^_6DL@c``IKWfWLbLzzBV`aKsJ^sqyZY7Kq~F;2FKwpjd67E* zy!=k_$tZEC$!NW5mwP2{1# zK!bugsiclvWM9vCTqUYih_Ijn!0jKT)(AcFl#1p$12H?xmX9CNBqWc4bCvM_ypB=S z2xuB<oSK${j8N?~<^Hj>}##k{7>>ofD;{x)@X4Exo^iH29cQ})jt z@#V2ukMYmA`yMOw;;C0|tEewQ1-C#Ene;+m_$;i&kbY$e%1E${(;k^?wD#nX{$tfs zdfC#TJI#Mgvc35L-6Iw?CSy-)qi$muiRT{1B$e29yYs32q6dsMVx^r8HndpODH;2U zMzVLPsy$nCV2t`;QK3k?U_a=h4Q(gpEl+Jl3~=CW4Z+1i8qfWtT#GxHSHk=85GLk< z{9qL}Pt~(z*LxT95KWg+lG+ejEfSVnZ#7Igv$>JFZd4AIzP&H7)t{nQNNqiHv9k4q z0yJ(xjq*M?Sm(pbo-Gh_=!Qr2epF98yjIrKm{6a=3&{?HBCSiT#X=apg3Xzp(Hr3`3a%;TRsA0(9A< zZplPE;VCuSCu{|xA^ppN&m5YZf!;3%{^hl3zfb~ zE%^Hr&(F5SbR)n_#{AGj!k7mll(l>+!r8XeP@(t1vkMqr)3#b%{YCK0)P%v4Ds!yv3g^5C#4)*=E7sj~HQwOON*hOK8ik^JD()rWk z(rLYGHuz?)aZ^w!X@H$p6-zmHR&2_=$LN9-45(=K5l%3p3kL#zf1YKIU};M|Vy%=d zBa6AvVHX>bjM2bwE>WzcXj@1+QW+Nm-uJl|~}Zm!Qbo~^(wHq<4TK*SJn(fz?Wx+pKNjf>MG;j|@>6Tu(WHZu%9(uN_K zz_1CSqMdp_l2$LM*FG%?ExBDR;%GP?Rf2lFgfTl?E!PDDLvsqP2NhP|YzOcrM(6)k z;P~$X@4IKe+--gP%>3<-N3`sHjp^@YC!hRR&y$>k|ENhAlYc~+zRZ+59RK(DkGGUZ z-czP<>Ad|t=oIR&PY+hn?L+HH;s)iv+L}9)qsNsM!jau(a+^_l(rA4SmU^%U8_gMv z5%^TXo*j7I_%${DOV36?G?QSs49z520K0M&j(*+`SC%~Q2|H8#rn*iS3L9MQZGF|U z4x~-t=nl!&gvzzGHdIX30Wp%J@nbH5aRz z?o^O2fd6OnEl3T)t1(T|^%>x{rp$%CCL{>S_k0@aTIW84^Y55K6@&&t(!r4{H^n8+ zMQC6t&ZU8^%R)e8@)C2K7?)m}gn|UF2degx+e^#XE?wT8Cce)X_#Q*3F3~?tQHe{m z`0z3xo0yE=ABmMqfF{nn`)K|yvgBadi(#=Mwl)uzul%kU{}`%TJonfP%9jeHXl1x6 zjz{5U^g9m)7ayL=k7F@sn;3$|rD8 za}Pe3LIZ}kBzI~13_J34=c&J)vkk1Yq!!5bd2`mo%hFVP!A44fp65^_YkPxewqc5hIky>w3d8P^p)R6aqvn$cUU zPXTI7BsWy0dP`?@zr*c5k0KP*!Ci1NF=wSUP$3Pli>_Q^YiY3Xyk-bUkS?0?FWGfd zGd?T#)vO!1YZl4Bq@bt6aWTL+AlcM3qdx~+4lq3S-yp_S|6bLPMHpgr5L5e(38*DV z1ge`2+EYyvtYmNtMXWONhix4<;zpA=Pyo9qkPoH4+;-!UKuESKt?nACaqCe?3e;HKt9svX0h>8dZn9 z1f}}rNsj2zm%`TYslr2MSTk*Tn~5#9qO+_7{ll+6qJOv)$Gv;sWw09Z7h)?!>ABOO zL;!CMxz)4f0=-Zz6#YO8V7heTDLuCilxJ|DvCXx0fC3#Y-u1RzFYmd38g}T_h_+S` zmbay2((-zHcKx7IAY-fUrkqj9MhU=z{g4bG12q=p-eq#T3i~Z7D!cwmp>314UYl3E z%suM+1RE@~)M9dKZQ@YG^|_<@OKcfp*eW@_LDkXpSX(b+GhLYbz>_3KCqIeoe7sxo zISneIdJjf5C#*~rx}B~ASO&U_WOITnE9JZ@4h=N_&$nN|3J?z3g@lq-YU%){;xup> z?`Sc1#~(Imt%bfMF}-uh^Q<1ctFNvnxx{h^{euQiMk^~{UgL1+aw&3>2^q?+2l{LY zclwez%6fp~xP0!)sg!8M_*MGY@d#(K*8(bPru8&*e|6c&5B@w`qG@DEII{L5Y9(VPqRZEc@@zBVmt*(~y?dA#20J8@L==7DVt)6Te`u!F|Do5bI+D&VsIMGyV2LaD3i1@Ec2kd; z?HBu62$Hj1ddNGuu>j)_RiSi|%-C=h)uJJS9M2)Z zc~JB~zJg=0j6d)zhU6*6hIr?25)0&H#_8qH)$0tD!1#msONqKkBvb2Vi@b*=Z$)0^ zS&5jD;a$TTZTuo1uTs-BDp&|#-(NTaL)4-=!`!Oa_e#VAkhLSCA1Ljm4!-9vRAS_O z!Jl@|$Ec}|U;dLMm;g(Dm@E?9O^f5Xqm=e-S!NR~wJK_E$g@mdHd5my2=!v=+TL5M z?!4Z1?<~a0+X)}NN=mSZzm{NDf?FSW*jF3=(9EQW0dn%zyJxd2x5=*?*oIL-SgJ1T z#Qd|YiTN`4^ZT|Cwt`5lY{i+=Y)JGO`k-Uxb~_BpcdCmMvL%*p&oPN~w&vfnONttt z4>>@Nj&Hf;`~03AYUAT5g$GIA)?JdCa&P2^jDTL7Q>JHCaVqMm)<@8ZVvzMuo~Zeh ze`#1hs>htHmRwDV{N~X|@-{EMZ2DJm2 zy7|R!4o0+c828I@PXk9vRWc20(Qmr$+%Ii{@GocJs@u>P&l0JDedSR<3BazeJ7JHO4_jz06PMWUEW{Z`?%{6#*S(EvGB|`JD$q?ij zjqnfg*BJq3ey%Fu!;%Xw5iV$*7t{_m51F{?cHYr5!8YJp5_QJW3=^oZtk=GUsUST& zaK*sJs}GmwZ6d9*KRmqOC%)0bbtfvi@<>~LkiMPyEsOYZhIIg0E7R3jmgzq$Ifds* z59V5D%^4N$mkf$wl0p zjbHgRlHPmi`;?hA&4p=Y$=e(2IIHDVQOkxEMOW!1OW@?i02hEY^#?~hOxefiC-1dT z!p4lfuYszwNI#*L8CFnu95sWeKWAWt zQO0<_<9d6c+!~o6hT(EQHFeO?*W$B%YI9mg0z)d1dZcE&N=Rtz^l+~2-pEKwX5`E?7Y~~cm*=bY~h?>d%2+R7f;W$K29=2Bx}L<^)$^fPa`65$D2Nl!(dt>PPI|T z!Q0lIlWLU)b?(S>nYcUSlbmA%u3|{K4W6s%!2fwB%Ig6}|5F-}&H}%+T*=#);CxUz zJZTeGNxE~|A9 zw4lZC^FaOD_yni%rCb*1w(2&GkS*qhLxgst{sP82WBv4w{9vIE&Dz}Sam&5HY}7+s zyrYws6Y8;p!o{aFVLZf>E~WOMwwnnLjsY~eX!qVSiObPP29^O1AY1Gcz5mk zO^5utLAf4&6Hs)sKYWjiHgwR?Uz3o_O=!+!^}tnn5S+n^Ozh1N?6k}EaZUaeTCbM@ z7xNg!PV$aRyE_BbGsy;Difp@YA7RVA{a#K<5t1K}u^>z5HMIt=OlF_2{X zbxV1wqZ95@rn`yRT>r|QsUp%Eoi}JB8Zx*wh7U^r*jL)dzOpuH*I%;u$p^)2+_@3XI@=I|;Hncl@^m*??P9U zZ2dpJ-(?7v^ISfg1sV2_sUQ?m&tr?#Oc4*6Vwic@H7ma-G&0Xedo-&Urr`sIHif_( z6$!-hb37W1;svb+?N||8uWOTB)Tv(btD+8x}SAZG2iDn%rkGGs3?vTWW9zWuhh zRcf&plNFy_v{-TYnzL>QMp5d9<&iuH_uHZ!MV4jqz4@rJH0;}@gfnQiR9z$D8MUtX zPiMZEEb=OLDS~YLXo^TSPEVG|Gxdx5B4w59lt+$!X=HaSKh!fVd>YYc1JXlGd_3r5 z`}?&BhsGBh=@oBTBWzL7derKT`LWYJ{pzNdeFp9l|Q69?^k;>SGx^cqz^J9lMj3ciFS^V z_N%_FbB%9FXM9T<6x--%zW`{USAZT7p?_G_PXYsad~?}wG$4k&qb+9vz5t)y^Q&Jn zfMz>N4WwcbLJMkzx`}Y3{DJbaVW0*obb1P(`5SlFui9_0v&p&ETQc2L?XKnqPy+uU;*upX_Ysxj&y@ zsgObBO@WJ=ABPV_V)F00_IN2BOzf8yvx5R2`dcjm0Gd9N&F?X|*kcUR11`13-} zrI-aizQ))}WB(@F^`;-L!9rD#SY|j^`DeSrU<8e%yvPTCU`E5pix&N=*tNAOyULgg z>j&WHFWnt<#YZoqDoptv!Q?kCLa^78icFB^VbRVPrZQ^W%@Kpv;ytpgkVGZ)%>q?* zQ3iC zeS4F^d%V-G6tGJGFO7=0dV?k-2A zkc}tZcu6>s|D?TFnNauGJ+lB`>B|=Jcz|w-I7#?smhhRKSXO}#A2#C`Va?Ve#!XoA zl`_-bfu4Y|f^b_aC##~MqZ4)B1A+KO?C;v)^8UbZ%JIceU8M{&@zN}CIlCD#lOeH~ z`uk5uGVDBpH*N$(6D0bpXWq*E@FmYd#d2T4fw|F4++v&)i!g=BzA7W!Nq#iewmS#> z)JOi(l;~8n5O*%BNaQi@5)#xKqMXv9G$8OwAuyPr6~IcQ37!a5m^wCqPoN*zV(xt10- z;ZAh&I?Di|H-1xV&f_zYN_9OCk$gL44_&nbV(MTp(sV7hdUwq5Eh7<6Tc!EGL!d3w zC#Pf}rIDbJc$Qnd!GHc&JE;eGG=JVMo+YIF=&jk^`%oap+{zzE`w-H&4;Fg+A;TYErdy`UGKVK| zD<7~&>q~^n-dYE*hY4M?n+4W37CQI-AN@zZo-rKI+?6 z^7u^L4Ju8H^`+>nT`WFFqX=v3tr2^1Dz)d-l<~>sp4(|6)#U{9QwJ*o_A`{(S+oc4 zlU@)zy)P}qS*o{T=9GoRS3LK6Tuoh^{IiTL+K{p9rv#klm%0JW~~pl-uB*9JEfH@zFix*TbT7p}C{eS>?XS{|5x+bnVZIIhGl-``_h z#?nf)oRL&<5-t3um^dBjvPSR4v9w#YwF;T>1`>1vjfw&R#5*JgG~gd}Z*_kWr3z%| zDiiIxO3T(#2*?85bN^s#ev^3SSE zBs}Rx&)Xu?y5nHe@OpsDRz;}5ja7Ole&>y@&c?A+P8-s9fjI=qUmDy$&=qW*z*UuO zYcfh*rj?N$^^2Zz%xU@O?aqkD1ZH*2#_2%e=7AmW#56!E&{ zdO6~^IU{wm&o=RlskEvID-`8ysZ0qV7Fi6coim|y4E=1>Jwu$ zDIg0;|7j<_BB$^CsNQEg@UJSdeQAH4Ht9{n^)SBgr)4oJu{CHSRi<1E9F_82*ZtE@ z$*ha(z=4-v5B!=G|65K)Hll;4#FABFul0}kJ_`b%@!jSjRO{NdE4p{fw%3c7Bg&8M zw8K4djDw&?cLpn5MPG>Y4-4k95kw;G5Zq;Qe#deB@-HHKM&d%E7`k*w zPy6o7ZMFf<9w!gtyal9|u!}BeRA%P({H`aZ+db7fEJC00UOY8g41W?tK~k!R2Hz)B zYj=`I=l6=TIM@Jt$rq-bnyVSzvqW59^BZVZgE`eE7<0el5*TIMyDck6Jb!$PZBG?2 zC&Sg`QI7wbwBFUh`zA0n%IY_f%W_P46Cz|FqH7mlJ@~RQ_RU)9O|J&7r&X`%vRfuA zI4V4}Eh(xC#NI(yH!Lb(e{PnZezy*^HBUw6_BiGii3Zdq`+_f^{5NJ!qP4dsu^8vV zF4DZ$I!GeRUnr(Yjhn+Yv4;`vFBQm&H5Ar(@K(=vfJ93Pv9hjCRcAss{c@7u(E^j zOg(q$Fb8a#?tUdib;J%OD$JVOwtnjn+R44TVhU8h>Hsc+(Be>gcV~BpVdw;iRmeS? z4lI4TI^hyu9Ovx&9?F^D*YstjCj0hIQ-|JjNfVj@=bw0J@Sz%$ny&2BwU1In0zxY} z)$`I+A~Pi#3_lQn==Ejh)ida3C^*FbIst0_AV8q+nc_)A;_bH$r3lKFodN8eE@Up} zu$k0NKXo*h17Q+Qsw@Uki*>D!E5&|SCKJ=$=R)w>=O}G|C1p-i7heeNPOM+ zIH&HFjaO{$mAHuaFv=%zZpj+KRtw& z4A8*ZX`$L-!(40YM8A2Jwk^O6k-IO|d37!SAw7QTZmwN(``Tl{`at6KOvLcM)e2{| z;3Q~6FLi^*z0vUVyQ*TsI!Pg8fB-0<1gPCAADjSGhUAw;`)iIwvqecCv4|g7P#$%L zZ-`+@`;!i~+{oa2C{QyiE?s#pAiZVwVpzg{-x|;>kMQz+v`?~?&_5WQY~p~D zS?dOtZR3s|ympvM2F2lS>*ayTuliGoQd7g%*|pBQR-Sy;9w_n5Z{k|<+utkl(WP8&SbNtDnvd(u@ z@O1xvFaG>{g{JlgZZiL-Qe?`gqMZmgOzglrqmkUM8%$EaI@isJfH% zqH7DdCr#JgFko1oj)yhzZOb{jv2^WLmldf%OzS6j^z8Gn!fkZ}$+v{+%G_en+pb)= zY3IH;v(EY-R=fl%koj`Jef%p_oQCysL0hU|a-wFfW1?Sb2&bPL z6K8H8sw;=4G^P@v4yw%%V#I~kqdQCXH_tqbJW7<7%T*CxLxiD<03}*A$O) zYF{CeJzLj5&nk9U{{qJ~w6pU}^Rg_Y6((2Gmc#GgPMJGY zs^?T&P`B1{5hb>DJ&805A6V#M(L>vNx^^B-oXp?vb)dw>I%>HKdhd#JT3z^!Es@zX ztjZv?VLYLl$+XE0J-M=EhYuTDU;hgnyzE6&O%9pC z)$Ien9p)JaeR$mu4ExaXVu#N9C*l$B5aQlJTICNU11v5NK}4&xK=S4$K&#HIn8ZIB z{;b4)Mg7++A=Q=O5@Hic=|>#wM#5WzA#b{`&sYW>vVw&k=t{5obTu3d4FSSEqic3E znju%2o>$9Ek4807w1(Y;me>wnM@}lc<#SW6Vs39q<(x@|N2+mM`-AzGK>>yf=|}AC zM&euk(@{$R$(5MoJ$b6#bd}=@*fa*@KE_a>_YhXT)B4Sa_IvRQ^NeJ8b7a!Eu-fob z$fNt!{o5Rq;?K2h!jM1>{U-9L*{QYY);%U%o34yu1Hn*ZTQ__iH9>8kC~OCtf=n5r z($VI|8Rre1GlW$P2)roUNr9kV8UuCCA6P~X5;0HRN(UNlYzfG*wzFQmdDz!IHM{Ua zQH@|M`w!rv;@Adoq0NyAU!g`Ia6x=yu1&Ufy3Q3J-M3iOg2O91XMLy=93{(?^U^ruu26>7{ zEC&gFmU8LPqX%1J06ve?(}}C0+3(K~^n=P*w?ELDn ziUBwnKE|||`9CeQ4iJklXN8PmWH$EKluBXz&F)&cKtxggq)IPRZ9%wK_Q&obJo#kH zm=b)(o@hT$ZF#~mH8;sFHj$BR(CT;&Yw+*o=p?Wd6B}qlPLP^;Wj%EZF%Y z0VAjFnDdK3xr_;u#0d-CKto{D8OaVyvv|r=!aP{fOz8%^(`Sr>Q5!HZ^e zhs+x+Gi`j@&lN|k-i+Cg39*Wft{h-UXIl#5mXv+hzQG8X_kB$~>Ri%Jw-wIO(wx^; z=;fv9$9kLFN$DNrkMtCMJ)7AxtpV|NEg_N~-r>LlITMI>< zqoTj^?myU;HEwi;f>8%z3mMI8KQU*BJ*ZNopeZUv9{=;`65BZ1KEKk+oleQduQt91 z|5PD@rXR0)M*$%M6+~;m2q?=AoBurM%WoxGDhZ&=5W3H%_-|`Z_OO5v6U9Xmx~*-Vgqi<-O# zqY(+tX(I^5KMTXG(oN701)pE29+>)`bI*a6rQKt}8+0eG)5DiaN11k^TLdAh#DT!% zXWpkh|1dCJ5{w2WdO9UWhJO`3I}+I-)-i8Lv!Wi2&}G3bA<+`1Y3zaUF7G-EkA|7neopXWa6>q>61V*DYNx-p z-!>*f{Il8IfB;3r^s0o3E~#9-od(om4O6Rn{c{QsCyZSaP9^B9sle=4D}PHFrC`$n z*B-koOXY$WGT5NjJ>AuI%YQ;-4kC?Jl&O6F-_MG?i*YYZdJEAt-Vc71r*hOI_nmfi z30hsZH?AobH)6zFV3bsN``mQ%`xL_5m>;I~bliRQn@v+j0FzBJtPsELtpyQJu{&-B ze@XULdfTO|Oxoznjydgf8)HZ|5#DCI$o}f}kgfvfl!5^wF1+i;C{?Mt5Ekw-t(rZx zDbrm^jx{<0*t2q#=IE67JB1H;=*5$3iU;(8u3c6T*y^NZ+m23hlv<8Z*Zd>fq|i?> zDLXZ`1$L=6M||#?6_#}2w`#MTG~#2S=#hihdU=LPrhCftUHGZ_yW3sJsyo|D8t7M5 zBfm$MQt{VG~Z3f!Zj@ge}}}=XCl<`;*eF;&9j)Rtyi*QbzDJKue&oI z-6!**6Ebi5rAB3M6FS6Wrmh!89E1`R84XnILC(NK{@7r8M$$O77hfHd(e+#p%7s8u zAYMESc&mV81~BfRQ2zj}Pt4~jOcwA?RU+K4+T#Kp6>i9r1^b{c*Rgq-{is;C`g%Dq zB&Vokk0Z;P>!sk1FB#6*PG~hvJa)fa@&g+tk}SzV3z{vwFzCn-v~R?>a__zk%8(_`0&Q&O4>&CjWJ#PG()wv--wY@3V-eOT{sx` z-SHJ&C~u`^eg2kch`gp@&8HVH4$i}&Vd632?J4+7#7*9r?Dk1|*KA@m5JCd@kV8WK zsk}epvS_i!i-j1EKKbfiC2vL^dmnNTr!X16`lj+S4gUDaR2yD(d!pbRV?uTLOU%L< zKW``flW#00tP|<1yl~CW(gNIgnY(4%Hd8iXye(u)SYpMI93tkb?5l$j>d;un<^(@S zuU3v|b3imUb~U{3Q4YTItKF(==Bd@~b#qSwq-7F7oCk*MD)J~RA{TTOkDupH?nX{< zjZ*13ETExyH{3H!*2NR+Gikhu^8%m?fpF$_@fKY~`zV2Jlq*{x`uW)?rzEF_9iz43 zMB3mak9M(DN!ly3gFOr=MtPxTs=G!#ok1f8hQe5y@Sdw?LJm~6d4E-1ULTk^hSM?s zh{(OW;janMbC*c3yqZ?9bOPkV^3(EvEUC803NXacIeEHg{=#SWEOE9;xQq%^Ivc%+(zay^Y%3pth_Ki zw-HIxbwR~;X>)f5sXLnY7j@q793(F?=9fZ<>ys$Y%toycGnIT4o@+DcEevGgp*?ed zm=?GVY=ft6s!_edHhj~4E$1!n*b8oxfzw|3%ZcOKpGceLv%J^K48-3uCtTHas`#~; zlD6W+b0-3b&zAhaz&wtUM$*muJsR}LC2GL;rlklFQcz2``Z|A?{BPvJ0#ynHpingR z6m2~cSmGJh`T->kke`{`uYKRt`6X<`xLWL1h{GGjic#=@v2Y~)FkfI~O>Cw9 z2?P#T%kkWt-xAUL^7rpu|*mQ)=J7*o;y zXnc!*!(~(CenBPkC*CYqKM0|-Muf}$y- zXdpE5ia%v-AF^kdmhHrM#_$YA#3 zUXUBY9U}p50JUQj4;C1t7~@i8F|>`qh!=n47FAi2e@0Y};u%%!GSxY)TBJu%#%b)O zmpXQ8uXk~UR#StZbQ7*iy3YG6jq18U3s5!$$;KWD1JZ!j$V9foG|s^EdmxWe>&0gl z%lN$mvHa{jJ6Th#;sK}mY3aC!BFH?}DJ|T-tc2*inm37)j4Z97-)=2EZwkqBwq9+g zGaj0B4z3U0*{M6i%Rjm(cS6FHZ;jv?=yF{Xt9IZE7HVUhete`z_c2GqDZ47W8iJ$jI4`_%NPpC^sC1 zTa$mguI>8KE>CuS#Er27duj~>Sx_fTV#?s>GHyBL4@r>~_%1|YVfEaz-AdJS0MIP~ zIQ7Z9&*VtMIu)VkBh)Slf;Bctiw0Wu1tYGEc`~ei{K6$9@9UqZOwLZWv@3-I!_i0) zknA~ktNNtmYsUtyD(%Z#Cb)>U2mv>L5lemq3rd`q>Z&6DhFgg?Zmu8l5d<}HeMYIE zc1hIo7d zJ6fgR?M5)TX1K>~>P2L>j0PhZSU#aj@OtLqZixOR8FkBDBo4AV1apnXuo^)euY_Gy z5=t;kaCanw6X5@u1ZsXYw<4b~3`s@odYJmDkm8*jmb|C{9()at&RrVF@TpxEL|uB+ ztEOoYXm68s_$j^aa5tawm2^1oJi;s@wod>d1KI;cq{qO^1*#sx>sTViCht?fTs ziA~`iU}%JQ>bbWOm|i;YW{T?{fd!1ClN2x>l&paOI)J3%1BxoduWZL3IkoH$6*#q2 zN_rKMY6Y6bbY#$>0+LE~B#bs!;=+bOn`TdSh-4N$mEVw5@_WE4W zc|sIMZynCevqQ|kpYo_z1-qM@HP!_@)qiD52G#g7+9L#PA!$jc+^1mW5r|%7r!UE0 z(5eyt20%7Q0oF#rcqle-<|6@4)jYvp$|T=veXotoLsT=T;Wg-==2_8$0e|;R)~5^K zO8~SKVX8|#kdzE!a*t><8ud|L4G5ZpRSl6@#xR0$W55%r;TgWY!8Zx>m$RN2SirP< z8*bRh{oE`YU~x*lNa3B}8&GNc2u$&kbI67v9)yO>=)SeVK}9T}8NPc}p)A0|ph~?b zB9ePSsR!fQJhW<|vU!My#-EjlsCr!oE&@T3ig?{um-W0=XOngVP?c9Z!~(-`7YRhlxxUqIuAncQMbU8Sw^=pgUO+T1-QJHA;S5DdUx%fhugme;GXE&ZU=Nd-2CB`j8 zI4gAOvOMk6J8}KYpNpE#kH34L^QrK;E(u+=`Wu4pQp?kwq#o9N1Tg1z_|t0tI2D&1 zm$N{g-46x`vniTVgr>uSB?hfHh~Ja3h@J$}gTT0Yyh4`q^K10#};xm8D|1 zN;kQ%t3!FS+vQBLCWEt`mSwHSf|bMsHnIhQ{3$bj$_=_2EcJC|DOygqhTrsrgMigK zTV?>(k$`RiZJ4Ey!H@|+)m?#-?Nb$S@Us4<@UlTmAva>;Pa=~hzZUzHDUC}o=gTxPttdppNhQhy(E@xNX%wwIG(&b zF&6m??`F5slI+8j%UX@STN5x*@9Oj0@5U~T?KFWSw(UV$mgpN z{IVV}=ut$A244Xy6}wb54b@lK*7M&Di>xYpy0lMC(ldJ(m+UrRn=0Nd@Tl?y|5t6R zK)grvYtl~q41RxCPh}pZZ%%1-BGn;|bg-`>$}ys3oz536s?2WU)iN8jZ1GGU0!Ta= z#n(0s781Hx#3{uA4J)?kdyxtI&ewwZQGJ+$Fl$}}^y2V#=Mx$I-PKY7|KAF^P}8H8 zev^09VT(a5tc%&8$J${FYNqzXuD@+P1Emo}lPBz-&KFzd+NStFHlhOlfxFH!qEvNi689jE8Ge zC%QKYJoGoZXDt}*oxWBAE5v-uy|aQ~;^H_WNo7`v%e&0FxXCJu6<&o}Tub}6&aUAK zq|g*@*sEUcP|Vl~p>c+Hp?Te2y7_j{Q4-_e6^bd26L{%@D7aRiNLm_pmnEb~KWDv# z^6T-YvNP8bz;woDO365vV%TeXww;)kcOTQbw=>Qh@73uO2?@mR>i2<~<8ftUUkE_h zGN7c73)IO=4$FETV}rOgcPBzR8_M)^Kb?{OP`Fre<{Dunjyr(;^{YZ~CeYq0%*bT~ z#r{&C>nc6Z_71B0C^F_{EAx=KKt}n??njyTifYSWwmNsqclO@P?rw^L#KC*7Cy9Ga zuK$;(3-Y|xCX=#=99Ml87>a3{8}ia-f52tV{?#K7Bk&C2p8AVo8|vA5?+4Uvw!R8a zfnpMOoD)`qc!;nG7GM*v;?(3Z2wObWk_}9lj;mI8{(;q2Z3F|M`%`d`?w59e?!IL{ z{NRnY3G83yzJ??Oi4Mgb)HI&NN(dND-xe|=cP6{T1UTfxeE6|>VTiU*79oH0w}t(lmzF{@B8CG0eEY{s^K1tx7zk?yMKywI$dA;Sx-yd|)2 zIxi5Zg(F~z0%loXevT}`h@g*HB@PjM2aF$6z}$d)qz^Kg^@p+-)*Vueky-yee^vzy zu@UVw#7fV4&BwQ}&W5pTuImHUT;XU)jr<&OW~Ht6Y$Br>!duIXtnfDs2<#3bMg)F{ zlC}9E5)@#rAHm@Mkmc_{t3fuqm&{j|yvh@R7QX>`i(Yp2RLU6npzZ@}ggP8|Jp)VF zMC?3@UkpCbryG}%Rm@~gE12OwY_MAHwn}ZULp-LC|MG;?WTDCrD=v$SHe_{P&Z?w5 z-7MQ|cqlf^j0XBidG1;s7Z^4$%KTyd#P zs90jPZgjALP5O44V3(FC`jbEUW%fY%Za7nOXR;HDFMZzHk$5|S>Tl*`?1JfH#3e3g zGdQJZVYoh|vz4Z^-->W}Lb_s{_C?799Bqu|s!AQ0-z0E_2V7#H#z7&i^l;ge`&)GJDB0X=mduVKSF7Q;o@0=FYjM z*&s2(r@*JR4d9Y)X2coEC%%;hs?ooBmavBQttL~!r9aHuD9li-oiSVkpY`R^M@bnk zI;6FvT(cM)0W&5MO7lYfQ$^3jR*v};=Sk91m_5qL$_5;w zekSpDoF9pPa%RavwsQX~myMbgKB_MLE*#ZA+rhYtwRPIS(zAY@U z%*Eh@fwC@_4I_X)fqw)mlSt{}tcs#ebh|Hc$|>vT+M0f8HWY*$OxzVo=D*VxX{>qDm-0vYA4|CrxV0phA`xHS z+~vw0e)6g1%ign$^Z}f@?ST73cS)H!4%VmW+8$u$;I=ns91LBlfeuNwH9azVH}H^YWaRdyjBnZuZ*epQ zk57PlG@mWmq-}$LkKqM+ctx$?z7iJPN*NwnVvA_14-6O1(y_r_5{ra~RY&$ECBFwx zh_VFZyADoTZuBpM5T^$m%}MlgEzlf=`}wZ6~uEb~a1&pNyX{$f;3D>N`>`Zuer?ZoL|^vy39*D{brBki|ccABlX(^-xc z0}CcJuiC3%WR}hHRSfA^NO`}->xI0g$p4cw5@?q?H`z28Qa-!NM1vXiPjCSLTfR-J z$m+Yv^kU6VyL!4Z1v{H#4vc3GsVvQ{k}f<&JerJY``Ch|y`lA96T#_Yp`rXGTTwF! zGx$gDJs)oLkXGq^o2hXxz1-yGtq!DcNh3zfcss($GcP~OLegGW310jWB{F}0ChpXf z&%!-v0jYs5Jud*OsB|9jOyW!0vTN%|7)fdWP&?~V_n9KsG9NsyFvW#+P_O%jaO|8Q zpM_9)nzNI9wp)n{NEoo>P28PdkNTPNo|1WVp{4RN4yhfa3w-Qp)zBisNPUieaAEy2 z42-*CB}wW7IZ(+CQHinw1C9Vcjr3AIjQi{)5m+C}~}_NV4#A1go44%^=}+ zp_oE~g})hm&<6izKWZ|ND)8xfpTULR<5Rr>=^-jN4y-`ZMAxc>#T30Crn6YEkv^48 zGJqioX*5?R`zbg6m;xSo=~(4PieX#q%QQoteyOsENe3|~!?VB6J*#r-k{R=Em0=q$ z@7CdJ8RC^cNY`AXdZ%BzFNz+-L`Pm4HF}6j!S{KB-f;`=m6vYs7p|i!V611&@E?(m z{4Zqu;c(z{8M^@~z0GEGwJan{v7BA&kQ6QSk)`QSJ$YJ3i#(r%Q2E0+JA8DMiim?K zJO>McCKT(8-&Dc_d5yS1JaVEL@=O4uF=Y!0pNM`)q5jD?arZ@NmNsWe|0*;K2~Y3O zM2MEOSMK?FM=vXs;g}|o&EZU*Ccs;82+cOfk(qa-rvdvHPSw>D1#eRnE}n2y<|0N)zJC9#f0DL zWT*wv0YQ#V9b~+VpxnN4caS+|NtyoqP^MR16#-_Gl_LkS9-z;#*6}G1t{Eb@ANW&- zvipndK!a3@d6ga5Wa&kEe-Gu++jl?BP_?%v7Q;Ub;#`uO;G*b#YZfP^<2-`4ey8f0 zp5ZCY)LdqtEERN>tSw|P`NA(uJ=yo)?d~AGRZm`RZH_dY3orc`sVR2SJ0zUo9kvjo z_tlSlB-O#pR9#G*H*M?-tdpy9ml@`sL%FCRYCO{#yYxR)yg93t8JLkNHUZZZv3pZS zs%BbQnWMoGP6c$(!fFdE}dKr z7qrC~dk>*TDHF0{KnmggsPZ!|=rGP0FZDIl=$B0KvE6V~r&i=Eg%^@HYVNQML}nGj zFrm>dx$RZ!JoUSNgsh5?^S$IiJ;BO1mivW28+(y6!+tswfm=!6HI7v#K@P<9W|~~F#*&Dk z@ld#;E?Z{p$9!j$w%B>GI68hIQrDG<)c-S0{Y*9L+Hf<*FCp4l6VY|*I~Q~dEn*c) zm&^q_RE59eM>aU|D$iGU4vd0I&ROdJ4J;m{IdnF0vKqXo zt+#UqYkeqYndtDp!_|Wy)SbYc*di+MUdqU{6&rXneUmHg1y*c#7trS&Phxl8Ffp+> z8o?i-EsxO{VD?l@G`7W4w;4(TPY( zRxOwk&ttjglvKhi`(BRbY*;Q01k3{kKaaMsKam_$Q7{d-_#EW8#&A zO_%46xrq>(iU5l8^y0^H4RpqDGw{B~FXL3EImGz2-Tr9F z4^V%|g$1O(yP{;BlHjdYSrnDSoyY;1W4PfWbAXBRq*BU z2YpV~&hSqhuV|ykQET@gd$vX%t6mci4b7EAtX5pDTXNIP|6~jIUN2WX&mX8b4RNw* zpKpUbFEf;>{V~t{ZM*g9d+$xu$)*Mb)E#DNM_t`j&`dL`&!GE-^0zUzVru~^acKT? zPUEMLsP>^wpH9y_I}e`$dSzdxjcXk}XQg*H-E#V7w?}W;#^H zmclk1Y(pLUCu~d{V{ZSw^vA3|R3GYoy7!*v-se8gdG7P+_xruWFekJJQVd)ACQ!z+ zI(m)_JML^7tyEATw||5#^4FU#V*%9KwAFZDqCs(O2LU?x>R5r8X~{EJ6OVV_;;UlX zyx&ABXwD9y8T*i4v=?~GMlTC0UIj0g zcD?%%c$Deu3+arULvgC@EUeRbz-*WdrQ{^Tttl+g8$W1KGgTvjdGSHaQ2TD50JnAt)lTbB~nw%oGH+JYsA!^3=P0Kq62-f#7xra4Ier#TmM_AEE)sR%>#D0)v zHScdnQl#^vczs{#y{dBEiJnyHG#-67iF7UoNGt>U4rq4z3I(A5XU!& zG!0lZNt$qY+#WY%{DKDl0~J+Pq;tIIKFflOESWEXUqFhxR^)kYF_^C+TeLAm0#rQYun*xP%20et0fSBGZr%zW~U$_v8O z`h_A8;%aHNnDIM_c^`9-nETB?W#2$^-_2~Ww`@F9h{@$IW35v!x$O6Y-=Cna@BK=$ zN!-Cy54!}dT3Xp@dyPZ$aPEnTYRNrY^ylj*hrk=ioTQmORv`?*DxR%H(ke!FwJvW- zSWApXk8kLg`V(K8^gy0b%xuX-seF6xR8mZd6V6VbW5R_Y#-vEe_&q*DBps5#4*T5N z>mbW%JyjsPds(%T%meh&0k&Z6A7?zpny31zb=#vyBw*yi3fvI z`Nu$8$aDQ?Gxyi-0#Ow)?$DU`$aX8$|^BkJ){{*Qe08?3}O3@7&<+M6#mu zY0WnKK7RJ$X#-+&DemJIbYwh-YCJ0q&{XkAZ9eJ{a&vP}-Xr{;B0kRdCPT1YcJZ~@ ztKt=*n{<_@1p9UtN#x%~a^E_W6BO_3Z34H=#xyEwp=@&x0V5TZcWgu^yPvQhdAACShnIZV}?@|y#} zk5Us_jTLFzIJSlNR7Sw6@A@IsC@{DTlWT&(?BFwKGLJn}mC0Kqhpy0$;9)riH}O@H zj+8%@iU~QG|6z6N!~ne^`ym)hAA-U0t#yIuJU$P7 zwCiol(P;HpOaDG~JTZoGQsmIP{;*=fkZvZ?{9??dJ2$cju&_;Rb|5rXU|)GlYkDHW zV0<0i2riS?b~PF+4A({f7yuXWdTjsEEzGI1UX}&qM!DJR{~m{H!l8be64TQB-%Z^p z1&)5#N+g&n!y*=MM(ZI~L3G$`@uah#cu+XKQn}y`Pz+RMMU&e&1I0!?X)FnGz+oDf z3X@UF_cN>atf*Fc{2A@K9a3QztPeV|^?qz4)x2ia-L$#Zg7%!UziRBrlN!5{Sg81; zFK$MpGrou_T)hLTyi)lL?Ner5_tyRA7ZI8oIZdT-bx$jQ=FDVOLwvcbU5#`I)Qj6N zb9(AyBe$=bGj~M8$9>tksm{A@m1VjKw1{nDg!L$xKC@C=Q`hsOoQ6{F3O!XqV^W27 z>@|{C=Dx1;%)*>X2kwDZS1SfAz+~=;i@{L<(^W1O(~d*|C7wAK%yDo~u^~>umeDY) z+q1YEe0gob(wArEhuy z(>r7R*%EKfg|XF$wj=KFVeGy2i0Z7ou*~3*l-s>K)QLQ_;Kfc^JgZDPHn`4x<=REb o$&24n_68*(ROOT@oaHj_mf}F))c`X38F*~pg8!iX{k>oQ23?s?Qvd(} diff --git a/examples/codex/images/failing-workflow.png b/examples/codex/images/failing-workflow.png deleted file mode 100644 index 342f6811b3f4c7086c4ef285128806f64fb8d02c..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 252754 zcmeFZXH*njvo?$pL;*oW5G3jVl5e5%&@ejS$bNL6x;T@KhFUPupKJ z2_H@Ep4Xw?nL4e}oyurjL2RceLdzn@V#565y-xA8Q(e}M7&PkQ!?P+yH12ciAw~i- z2@1hTVt(+u*LOR72`-jb1_&gYuwOrUgT9u-gqmD8UN^f|;_{J-9~vV6R-}uCBk+Pb zdG%oJGIRZOj+RY_LpX$|WJ9kiR@c0r-66)=;vGY!V21Qwn5>h4x6E12Bm(dTL#Aroz zZs*jGNFIM_ea@%blO|5i#uoa7(jXR}BwwD2EpAn|tN5~msru~G2&3Pg4>h@?INGaQ zm?i74u)47pqi6GsJo7j1-u9+OPkDhs9{hgc0XR^E&@&1>Xx<}K>V+uhpkIX!#y#pA zE284%j%Ob>HqbbO&l4C0<~cgNsBxFupq(FKW*vc2d)SzimWTJH;&UxBPfW`B3mKKDN$X z!*^+8ozhimMlx?%3#hmgGLOm+0}aHLbf&3Jf;Uu2Ef=xA-tzE2baXW4uVyl5!9NJzR9(chCNmE6z%5lLBSlD{CBmDP`$I8!@I6@@wM$PN8Gt2Z_`#-6Ge zN%~RuLwpO@{kBj@&1X!M?nqu&9aW9X8a73`a`g)wo^Iry4o3yJ$kfB-P;}BKdeiXvQhH<+3v3({VKhj^I z!hfG1oo^oV@&z?ErW!RRD9~#7g89p!`77f0cZdTCRk5`FQ6C9yF{i}wexTofOE-K^ z`EBMmFUJQjG3ypK*0F}fsXXySsj(#_-e9AT$SM*jbWl3U{0PM-41M>kQjQZAZ1$8; z%D-PSiQ2rIEQfG182S_ENCD#xrN6LGp4+<}prxm9agq62I zqM+k=Igk%bq$iC?dNKY0Oq&yxGm~SQqmW}iYEVTs`EE<`{?q(e+U2Lt0`<|}3Ep9s z?=Jnbz?C_k>RHbvddYj`d+*ERtlZ8`JMb@(uE~RFEEP_|JK^`N(NWx}`>4d!)YO#J z5dFwhUAs8_Af#wbwe`vPY@to@RH4hb#`vQRCb&l-l}>Bnb+}Bn-pJ(}hR@!}d-^Qf zpl!NhVlqkLs#+CpW@9n2-e zZ87;im?Qan)5w$JwS4+frtzs_53Sv1Q3WwoFv)(xVO8v`#v+xe)X(yL?%04run;~so0Y3lG}ILD5&CN zI#IgU@oscoiW!xzdgR)y#Q|>N1028Bt#>lBWFL>MFyDb$)Ltzh+WB{`Iol)_4q^m+6|Y7o_{dL;T226L64CQp(5 zp9^d2Un_hzpIZH4U)%g91X+(XpVXQ(tdQNF+3Gw_+ZO72@ZBpiKC+n7V@7<&e@S&- zb=Pq&ZvF9&?(&c2)7MR%rwRMTU74gelRe`M6ILN(+jo%o5IQ+BMTAGT;+`;Z}QJD7E>6 z4^|R@yqyv{j%L(r;D^-$8>d=X( zVM|_%_nh_I9p6=7iPm&q75>@}wuAPAIDYhgDK|7Xd@ogYym$7|n5p?ca({fLD#T7x z!MSG?tEXzA>Z590VY4?qzeMvcwjy?%X2MHG_*Qw+1jz=apxBY-dx-8!UGpD-j*X6b zo(dRb56CKc7nO6&b3j!L_oVi-5Hd4Swb2=*R~on+bb@(|t4>?*et1<{EbmGEk8DkU75`E0jY1B69vyNKScio6cL}MA;%ELAf$1gkD;t5 zmMb*QQv*UFH8%^Z^E1*c*jYS~aqCCxm+N=Nu&uyWZdO!hYM0i&eu8X18j#GlOh2eS zS&tr;Jvzz#GicL#s z3?ib{Z0)4B(47iM!Caj`5>~(e->rI`9J40*zbG2xCu`M6% zKAp8K2F*kSkc>0uu@358+Qey$q-M1yfmo~<)x)uB?fvi2&N6Fd;U(#7y4H!E`gHmwaxv_<`-W@J6kL-#*U4J z2;4tzQ(wGlk4X}<7wq8IM22z?T5Z|%T5nh@+wTaIL%x)D7OxDeM2l@(^VpRNW561qb4LN>?xv1rrEfSkG}jE6(~IXNjs-N&wA(O_1V4M-m~I3 z9iejCmyJ#$D98lfLR3DVi$hD3mg8powzLPfhy1SfEK94UXZbnb)ZDRqy&FY6m1FCb z_;lbDJ|=#q7)<1d$CJ@Lr=K@|gN0EAUE?V#kM{^%@Z7B*MoJ9pH!+8N__ zdNnf3S6_aSyS4H)v(I?F6^TmjGq$@LxjHdgYrfvgJ1V`NO7+%1|c`=i5 zQs;^dFFXyp2?wU&n8wp@mqXVf=HAGO+#i7YlrTwfTFC4hu%R1NkTWS4mOO)WMF`#LVHfIjg6gME&HOFB54Q}eR2v$8)Gy+=(=E$nP&A*e1T^G|c&m&nsMuC9)PY-}DL z9;_Z*tPakWY#agt0&MJ@Y@D1dz!fYmUiPjgo-Fn*&;IV@U;Rj#yO=s#Il5Xo*i--N z*W|T>o2$svr@wCWAAf(3)7;bQKksDk^3P)d56JfG3>yb4JKKNs4Kx-0byQH*%G2Cd zPs+*;kQs0fQ4SsscHzGo{Et)rdCULORQErda&ic8|JSDf<*y7s9k0J8NS!QNt zC`LzTXbZ7#Kc%IKx-HK@h()t6n;>%|)E+g{xGj3p<}>d#Up>*cW^ZX}IR$NOpRk#O zD!=oq0lz1Bfp!by556*>{@57cj0-3l+Mirb@ZtptwOm`&Ti`LM|KRH*4gqLE<`2h0 zFpP-}bA0wE{r-#xkYN80&casS^`@W0~zACu>QXZQb@DF2@}CK4}hliYj(v2{f-#XXh2 z_eO~E*&kTjw=aH>U_TUxWq0|F=8oop&pJAGW*X^Fc4j6{NLchW!w6_3LNBgQ=KNNU zw}l|h&(*T!azY09ntWU$`cjfgcv@ zO{}NQTzsbPdFgW%qSf1Tr!ymyI!tig-SB)}M5y5GRV~ zJ7?5#Fj@tf_2H^G<88IO|HHWacNdGjds~G?Zjs`tTC9j*T@pP{Q&f#CVa=|mlQB$< z3WYFwE~_08J$6V5s&~7nB|qJPdOZU_1t;ve$!5PI;e9irWuV|DW`}+|`pj2jAaytz`OLLY;DMcriXlV5(@ zijvdvzC!7kJ7%&H@@%x5#9+hTzw7@Uc=+l~Wmj#?%OCxV+EF@IjoQjBznM#}0w1hYSM2E@sJIt~7;jZvLHCPuruE1uC z2&A?$BUqboWTjE8H$@~DD}|bM+L`t`TROM_Sw#Lk=LOzp0% zATuXu#V>VeO#)6=V_DtEYQ~M<(t|&|2tL|eMZBAC@>1{>^D|;KuBUS{ip<>=i$+Mc zXDfT`&L#>W{T@7%N_dPEs8>3#_k1aod8f2gbHae4ef-8SK6$a}d3p13QOCIGOy0%GZbZWboQO*Xcjj!) z?;=jSh{%fl9 zVOZcc`A{iCOPkmnNB@(*?7)hSNg#*h1-Sn`hizS;m^f_H?~0|GggI^o+EQaz9z0o~ zevRIfxobEjm_5JygwO2!gNoU1%WmY;D~;y)GWEj9gy%5h4IZmqE;;9`b7M}+1bOe`X1cVW#^(*iQ3#!Ex<%UgV1nJZkn|DBKCC)r z*Fy;VBxcfFqbNq?o~)DYCf0olrY_~noaCW2Q5lN?1Li6Bow!Di9WQvP?hZP`h4_Q6 zSq{xkjr^fECj*=UXNROQ+wg#mpxo%%8%>7zDTGhUDwAH7ccM|5gz6lX0P7~NE$_*V za#_h}mXo1Nqii!@(~!&Igpk?n;-gjHNQ$O(fk5t}i%-Nao$;APifStH^M|1N@AJj0 zqgO>ZbNqC`u;-sNw5k;`?e*P$(%v*4=Nln3p6G9uax>=oj*4UbH{0AogL4S()@ytTVX45~EJ>B6dMOJ3?6hf3`4gyr7Inl5+5 z(4z*ziF%og?WTd1-IU3fOk7o9VsMj&a8Xj(PW1{=DS4#zIM1&<`aov11M&LoG6Gd) z)@j|$tTV`5ix$r#cffxVj1ulkW}(R}d%5XEjtG(pGfMu3+>-==MiueqVT`GrO~1P1 z{QL>0^_FI{6l$5%7yhD8|70@4?DY#3 z?$*st#g>?3cgfr*rH4H*hUWSb3yUG+*T<1r0qDCcJv4fpmN4kaPVcn?e$U`chDf9* zzFjs%_c4ckkwBAd(-$_d|H)yuLE&6Nb?YUaLG3m?_=7|=);Arv-H7tB8OIyYw*{CfC%i)j&wS~t# zmtiBO(g%Em3Yj|;Am`_MnW?|CWZ6ut>a&7YZ+8N>3_{xcu=0|qm zLwmmzp@K*1*Iv|nEhXu*faB>BJX}^{H5p(%L-~W&6W*Ct`2qZr7akMA;PCtjEk6Jb zWUHo2o_{U`?I<`Mp(z#-^{hDU;_V)4Ui8LY+4W@5#?L~B>X27<+hNi(=mTU3j{D#n z@0T_)-$bpvoxp<$_(U`ptQ=gE3OL?aK64DL`lPIEwu# zt%}Cw66Ebz$FaI>9o(C9jUL8RUF=+nS-}PXdU#(innPMg9Y!{q=QDAAzWSGsK>eH_ zN!XytQ7HfBeAcTKM#xju|DL_gEtDh%uLct^`QJQVZj}gT!_e^1)uc9mwQ3u1s~h2P?pmRe5iJ zOr902u|83qefD}zCpO14=GJN~FXP^x_6AXOmnGhPZikn$n?6s|j2xN~E}hdseW)}1 zaFQf^revpS-`;*NeKt}k$_+qNJ6EEk=nt(|Vz2dwNWv(Oy-m2?0T(!0V&~z*A$eYx zV{@_ky@)`p)E|3un3vx7T<;wJOm|N7mtSnONh6p#s0-kc8*SC0PZ6+;vqd&&%rzMg z9mAe_amM6Dr}HzQheVI`g%kBNxmjxxw|7F{q@myO8xEFC$(lg5AP_R8G*sh#eLo)H zOK=rM^SoL>D1@WlA6Pf{a~B#bI9z{^N*mRE5+JL@c*JRciWc^06Vqq%d~J2WV6@6q zrQT;V+DO>sh}uQ>sZ9(>poUUHpRd?q2L6E$SMyu;^b^FPrbvdX?aU9`DZy&UHrAQh zt5rfp4S>RUF%B?J#D$iW%^L^T(E8G}^}}yp z_)6BS@LqWB_9KO&+(J`c=|Xq=8URr>)(JEd;cQm@J^;Xca~%qju5-%Wfpe@7;Q z4RJ!+B{&l}X0VjjE;T^~Z}9x}^^%E?NboZO=dE|Rl%li3yxDi|)KT~3_UYuAGY68NAge$4be zMNGITaU@#E28R?%E-P;qJ-BZu-*03sKk`D(DD>>rDW#1AQ=vL2UKGjBVa*9y+Pgm( znZ~K$zI&!!J|?}HLYs|{jF^I34$J)la6YKvTQjt3Z4)Rjo3F3Sh^fO1(yLEgPj(Co zCyu^3imlfygg02V6Bk zQt-pAC5aNN%ISxwtygW?55)zS* zr#0MnFSO~D(~Fb!po{`802gV*;l0t0%@2I7h2BJ)(FP5+X8UlyxmaHjN(|f$9ve2s z`3Ecz%@hukcP+EqDuBa6X#sjdJ)W+0VWj$zUky@7{b7|@+OGp>XZDMEBQc#G+b;xN zTgvFk=fP{Io199>pbqLnDF-Qw!wjML-KRa(s0~Dhly-?eJ(H+Gc5>Y~B(}Q@1^Pq> zOq}@#ODo+Y4zSzUO!KCy?{GY`T51V3C?LnTX1D1~|Y8a1JiV2_yQ$ zJy&G4aES3)CITlSHIk+vz@(J#=Yi!4$)@>w`1m<8l5|u)*?r)p^|_8>Y=;=VHO+6N z|FxQauOk2y&dMAFmbeF-2K=g^>t^}M>c{A2-GvirOtB9EAcUh_07iMwoOD8p3Vb z=+SJmi%ynQR93m4kOm}-k|ou~a5FqM03%}j?}Wd@N( zVeB@~5=TA9d$7xFl>115mgO%)J-HdQaW~tlap&L!)rz8l=0#Ij2MbRBHO`P)C0d2= zuB@^&qs2Gu0&c51KFvI?ukmy-$Uv3BlHtU?^fkpg&NDZ#N5bHb-~J)cVQU6 zW_bRHWX#yNqD|FelP5mo3Qk4|``K6j` zvp@5C4?MHP34@5W+dwp!V0dN5)e7JGtVcoWc>e9P*|-4%kBvhFB^b_5`75gL8+T-N z`AnlcoyW#!8|M|F{j(US<)#c?Zm5)*$!VHldtp&7I#+H91;+`ei0)o>owrSsf$IHk zTHqtWL~8M52n$4x+H9KWaaF50;bt`45x|<>r6|nZaHUN93(CUkJ_C1h9*Of$HbjmxcR0P$U3_b)hf$@9GydY zIl@S2J0t@@p#W~`U2uKxeBCC!{^LsAvx6+*UU1FQ{Tn9DHC^K;qH;bYx!Hb| zmNmk(3JVMmxAzI|W^2tD+jM<1CRO!D#K%*ms4iF(UAFs|>D8&RpMBwM(N{9$SL0Ru z3wdU;sk4{E3)Lo}EGzH%EMisgF*FUmw$w19<6|>@RNR2nyw#i#2CbOkS)t4{B$E7tt+-59GM<` zFW`PjPs(pw(ozsCKqm4zrc%D-R>X@yP;aTpp)$mP@hvw*S1gQVg0e!*#eN%4 zklzzdqz-fTMVb)$1%enaN1$IFU34EVQSRS|mCT*i^c#M8XOQaX*ri7p(MR&~+E21pud;9ucM8c>?G&?c-o)Q&PA?!?lDt4w zpe~&NIL*?3@Uad?^178;LzMD$ZGLd1-r)+}+0O6-wdQiq6G0QaK;1wP*P8Fsw3cfb zhc3HN;>}@pD*R2R@+R`3>GJYlAqLXo=TETn=|whWnq?16P4Khrvf3&T&fkyA=6(Kh zUZl{VI=rivv0}8v#+a*PVa6ql=hcHZu~OUJ)nx;{p$1;s>~xHUmp z_xa=wzWfT>$G>?A6;wUz@=f^G-c_jQvyqFcfXQrH{qnAeCkOj$qbp3KpKSc9bqF7^ zZ~SNoJEnmCw6gWR?8-EUzq2}5l`ATd?;KUnGkajlBTlFhK>9z5CjTv2?Kz8)p&u(H z4h;d%%E!w?-0e|PW&VKeaATikE6_>-5~!HUGSO*tzcTTqNpD87<#)*SK!&OCgEeEn zp4qFAzfu-%8!;%_Mzt?MB0}vOnrUwyp7DWu@aKaV#IaFne$Z~sC)JX8pf8GcM?24C z$zQ06QSl%eaVQJf4r+kJ-L%IoRhdt*R)-Qz(yXG%TJ4e?|g61JZF%<@Dkl?=obICh?gRCG^V<*7xxsD(Es5TXlx!X)_) zwVf%pF<@!h%acv{1;zy3&LfW~xRJP2Ir4FeRptZCPR8NSx?*mg3!V5d@i|P7PHBiZ zxsC;o0Ogc`$rheh`_IK;t=BCyhETku8!P`uJk8{6Wy^ypO0d>A-NEbg#x11B3s$2q zIprGPG8fPr`XVY#H~ThXTE$^q&A;I6mhiSg;b%%WPH;lE+VbAWQkBD;_}GY`ZM@oz zQ)!gDw~szLS5d|}WfFs)rg}??VwO}`37!tBm|Vn39YA^rTbc0`cIQ}umvQ>yF(5kp zKT4C0Q-@X627}wI0l}{wr$fySj?S#LGXYp%w0C`Vq5s?@XA;^Gn0nj_IOuT+Iz2lk zc%SchU$Dok4*g2o!o@1c0OTsD8BpAKqsMN>5hb_8V8+!~csVN>>W|t-p)G+wVJQ$AvXooVeZlBVXd~6zg zQq8l(1ehK_=xJ!=rDv6$kd6Nai`lDkW&OuA;Wej~!L=M@>f*FKNB!Fd9+6L{@;6pw z@R|=HmlMJWy>dMrmSq@?Az2Ix5!6-B2P8zspY&bdVPce!Fqt@*+`U|iit&($=L+ze4! zs#oM9CJkE?kTcb>9}mz$M$`N>J446_aFq2*u2kIZT08rh29}5ioKCkkbZSXN5wr(HuZuY6`Ixg~v(;sJ)ZxOD~QO>SCZ^!{Zw3htjT2r%s z67}SuXEd>DyckbwA2nf|Xr0&#cDKWRyt_WMIgxAGy5PLBaRcg1bC8?c*%pD!?Ew<3_{s7|3?EwZ^y$&yJ;{(##eXhmIBn zc^Kv60`Q@_92>`5e45IcS4vGG2ONlEl}C=cRkpd82b%&ml~Uo4)n@zRB}L)=uR9^% zOw;ur=ise4Yd#s3RB` zWRVU7y#okrW>EL95ZdcRxfGO+PzWtwSKGz2iIqxU8x*-?Hkf6;I?VoEbPmeb2rtnF zziuWN14JiUR0D+hnASg2bM^7Xx|D@I%RF_oP9*jv&T?NCW>KDeQQK|n6w8sUyLFzm zL0VuK5&WG|^mix?&EHa=x*uT4@T^=QRXLMuGU%?Gu9JynLaBiOB;1tX+jf)I>HKAo z*z#(rfqwT2txxyEx(54|q7b6-sUX4N0<~=UfYb7gOnX$+!4p1vd7yw0DSbFbMptp) z&Q$x{FD3>M827brkn32Ssof6$P=YjXrvS7RD9Qtv%GkdM9cmhF34!ygs z$If&dkc~}x!soE@DNV=NOAja&5FcILfjH=vH}kN&C_RD68rrzEHtnoknSbbyuW0dJ zDY`tL5*Rqb9}&Ea>j)#XZpWg%%Rg*J&XxH)Km7J#0K>|I`{8#C01ujMu?vv@&Z)dW^Y0>}mOFU=cIbT7`gsD6v3-)rMQb8Z zW!uOv4*p3hTCrB`yYV4Y`Ym-BW}x10`;O#; zLnFTSzt`pe8+iNYr@tdJFdTAY;rkSSsIEL5nKXXt%!s zi=R+Y2`TX#&-%}3|Nl4n-_=w1j9*=>AnWFOYtZlXA2yxOc~6({VTB(!%9FG0Adt&cK?oh#T+>-N1dNgVj|hxuus}u3zz;gB@lR0?`Z~tWZo{T zwLMcN8ODY^&Qx}LF8PP89f>^|fjMpQpyiJi4NwpfuB7f6!;X7Q z63iY(v+pW;vNg#RLCluOq*ErBryCfx&9y<^(0i6SpOf^a zqfrI*LUqiZ!yTdfJPf~|=YLm9y;5limkvZ6Okhi0D&6xj!KC@+eln3n;O{XMENmT- z>uT4}aek3D@&=~5o1q@Uyt~6?QN+YOZ<6e$uA|^(Q(s?dbCHhbC#oc~#dXI$gCTYj6i*x2%|C0FQimEI2r{9O zdQEMf!7jdN(PzOdI+m$?>%LnX)Wa@b$H;4?a_9;uuE1DLi4BC}>7(OcOXH-i;F0kp zE2Z(J^?mQp!gK+7lzsUAG!t8x$KU4yUv8y{$M#|i7g!P6~4}8 z-Ywg1ShqD=*vxG=+U+L7)Ry<+EL_XZzYgMN1K!P;XaSQTz40oL4U1qSQ7dj?w&{UaNgg>4)1GZB<|Q9%P#@E*Fa8$5Dr= zWUx1K3f}?KDP=wra^IP-iw7xFOP;$`3pmWx%hIW2IwTE!;FideiLz#G8~_kXL1YHI z5>RkhQ_J<;E>hpS5~egq1fkz)O;$7Y@zW_mrLWS*kUp(^-8}y}ob%s%!?v0i1ppmJ z-2wDnD?E*-*erv8MAzFwtX)1fN2T=^r^l>B+TnMU&GG=>j30yT)Z5z_65#p%$!b{Z zD-MIr#F|fH!FZ%^nDE&!>m25iE`!3u$bx$X@qSLo$`i3!vp{mrNVzQgau~*hYn2k% z{Yx&7Ok!?RGPN&AcnWuI)gN@5ms;L2);O_G&s7VAv=mja;@*3=Mh(8$xmsBaQfRRL z=DjO-B}j#QXE|j2^vU&R4G?uDn{9<&*9sy60DeHhs@qCerZU-OTls?0J&U%`ul;(}=3hU4 zS{(tYWGE63X0nub#W*Ad)upsLOvmD;Y)N|Ikxm;IX%$}n$X|jVcH*!#QO{9LpZu^3 zNyKWK637tpGn(}&e;dUHlo zirDe&2ITk&WB$-}-b5cz*S8ugX{nDo>Hig2g~sG}Qzf%P%Q=%J@4=W2>SaVPywml> zhEwDoJnKzxlDwkyr5`>njQO{d_v;vuCu<;RNU|JcLMxYLza!+a)7^4)=V-{x0hDOg z0}~m5X81gtf9`mGL@IF6$3_T-e#>C7A>Lkj<=O>V$c?4$Su!r%BL+==v5Mffo$hyE zbg@v1_x75rPShay`-zU$t4CeFVn4jdzMcU&WnrvVrJe!&)Z8*+-gimpvo9t?3aCT$zps|#b(ksKRk=zLF;9?+CZBw1@$VTa{%d;F0PD~HSSBlg z0{nz~xV*i_q4iNsvsgS2fpdkIOFd-%tytQ+TL#JqnCGz_yAcx77zEg<021oT$8``V zicSzt}`} zg-fYi+olGMME{=8xYu;=T^KdEDi}}t%2M$fL#?TpTo|1rf= zy@7grYngrz zAbdJM>l|&P__RNh%4?d0OUc>4L_P;$swpvS9?kBb3#vG0-a9#sf@}!6-du}>96OZhQA%JIIS?e4F3d3Hv6b+U-ZvpzHm1kp z4nAo};f;i_QYsuWpJlfH8piMR)WuMa3wpej7!*O=_n>ru%1^LRMea$RXJZA+nsrjlbm}Q zP{h-Lnb$}2;{kp;_>436>b}~~t{l>9$K`?$li(vOp4KdN2^Y=QB+)X%#vYA4(vWOF znQ$UKEc`7S=(iQM9+PhqXIDVps>*Dg^UQ3Y^UzcC;iQXMyu{gDG8SqR*(SS@kt*sE zO)l#F!!%8ibWEW$d?i;tH{2J0Aj`~S_O9GGkzO&>-`7uyzu z1MWOn0e#}#wK;IhXuUda?0?T=z-vEkyCaE8sxV&?^jb;Ou^lM7fG)JDm`!AeIB)&9 z>(Y#*^Qy{GuTT{)RIdyG$tMiIJSpJvIma7r!(_>NBHA!<I>~%=eGicbT(Md;^*xpn=u% zo+A)mZXjy}-PY1V#RY;s?SB}>?_L;kW7+zYA%ZA$DHnIXKm=APri_9E^|eZiQD>xw z#L%ProYnmm;6gx&*6|K<^>oyT z;^eJ=p!rTy5Mra&3+3$b5#f7mt9J!(vxXI7jC}02Kb-d*VtSPdqEPcuHLq)m63t)S zh>n(c*-U8Ol@%#@%28u*k0d_9Bq@lumP}A!|uqqu3_D=v5KBV#a&IS6$<= z_tn1H1AFQo6{h{#M3&yeLc{V;OMS-K0;om5(Nhj&#HtyjKn4Y4H${+})-2^8yw5`2 zf5>C4*@co>_yt`rq74T~N@eP&EIaS5)DWn$O&Ez!4^{)^^a8EMmL*^Joga%%Gi5KE z8Pf$%hTB-uX70CVH_4!iMvG0x<8r8G3I7WB=Cv zR13yQG|SDvAhvV+3mH!dJq5Dn?x`rNVQm15<8m#2_H^M|LB%OEd-TM*3joZs zGJ+7`8G4~QhX?r?dL<~h_rY?H*-_{xm>$mIX{zRMHgsq<`V({hwx+K6QTGDuq}d}~ z8=AmkRHfM9zPqBf`gu5C$D`07U;C>#Y}iv-fFiS|1YisOq9od=sje1lu`->_^_AWM zNC4a@YpuA;GvO$7JQ&;xuDu(lNb#z+Bqhf%B1B{fbLFEI#=lMR0BInCBkG=P48ep< zv0D%PHZYu7rja-~cs4V}HG48?$LW{8VOBG>7E#umUyGCPn_dBCB|cSuxNSXoRfb|S z{~pwJeN;{Vv#kvaVl6!YRMdr!I&ix>$E!WJ$5bF;iv`B}LmS%q(fb26*nAt)%XZuR ztbfwQ?!t*t)jwjKNr(S_*e72lqkMbGE=r;gNO{0%eYTrXN4{a5)u;eEOh&7t#ArGx zgeaWZcB+#ejO-Brz`Ituh9;OUl46^mJC3aCHN0!dtgWL*#;sLMBUh1TEMI995-In3 z>N|tnfq2hi)yk$_ng1;SE9W4Xq_2XNE?bnp-zvzU$WzV$vh~)xlN|+1n_RJg^C!ggh8)}Fm>=drYW4^~YA3T@`sEBu-<5FJ;kKJ%WWp~rRXCK#zAL5G z?Ap*zT=ZK#(aLMP(?`t44XyQ>2E47jY!JCj6y;{E!i4i9v@~H8+ZQIVEM)K@7qtFlUb(*wu7 zz%jhiD6{g?3qBui-BwCZq;TjvRFv$LSUm2XLl9OOe|Jen0c;nFzL^7IP!+)~pdq`R zqUnHXtxLAs!yw104gDkp>F3`q5=|a2FKZ5oRDFWMCAWBp_EP))#hX~oGR<$Mr5kp+ zuXBr>f9~N^uEaDD0HEyS1)E{fJ48e~@-6QsyfU5J0r{hL;DZ-Vj)T|@q@}{Hc4)IJu(EQ&KdIIh1#1gb1M@n{L)HJj*OkhC2yH*RH> zR!Hz&UW>DK-0FIk>skn_t)*IlTH7qQ-p&Umo=E<5^tb0?KC2|_lYU<+AH#>jsPJJV zrd`LA7Suqc(k6UEr1O1|4hubP1aShBPI(U!mA0IpsnQn|kqnfW(g^u>6b?4}*c1@` z>nm4HyK+ZwMj{7JrT%R;Za(D^HY9fW)#0i$>Z`vjmJ3ciKhG&Y4dJ*E`(Mm_86|yw ziaa7#O;vz2K?l6DM>KBJQzCFkZmewH``$>~x1m2fYTURaQQHm)P0%BO^D=1U5sqhz z%eA)_Yy%}@OTRPeSvFNTGN-F)V+PbWl_DL)q4~a(LHP9yH8;RE;jnM|;(XNoVB!%E35Zj=*c=pgIk4=!jdVM6zS4=lMPAKeN5>ri>J^ZUCU-tq22ZgK zmR{5jT2AZ$9A!|VWXS#X)(>@MT&r4*2E3&wd#bIPI{tY1V9Z!}tONr$KI$ndg2%=X zoj>Zy_=K<9Tu(MBz#rN80(U)5HZ~s`w1cQTZN4}f3wA&2$QG&j7+v_v-y60zcE7$GBvd5r~@SbD4#Bk6VU$W^$*A33b-H9Nopygrb4g>+ zG*!Fn*tEML_d+W8U;hhWqI7#iGw|Gv)}-x@WY_6E?*b!_gS8DbFPcSE>GqjZD?WLIEAFW+ds*oEKkI-tIkBO9WpF6l0^ zBhX^}G1L^F7$}S$nD=$}A#v89A-QZXqSMQoaVP{XOFKdz%;a6i8C=pRZd$Ifp`JV#Tn)9mpS5{P&f5itub>Jp5X;RuR2Qo zFE&9n-5-Y1nevGkHD6FIZgzxGNgJZ)gGCw<9b&RA>ZkJp;t4>8nGSQ7X-*5^+&+FeRVE~_V}=|)i;vcBF`g$w;7K%?RP89y2^a2 zoKeZE?&uJ=VaRCsYu1S4Auq!1n@ow++jn6fgx&PdammejC!3t?&i0oX&a9VstNX2F z@QO7WG}Q`q^WJr}7Aw0k{QKnfFFt%dby(XGz2dm*uhHQ^Jjl-ZVp{!im7f+H7Anu`TvuO`=g+CT}png6?mal*srJx^E~# zkVD}rI`SdPhT|o6$g2TuukBX*-SLQ0k3(WdZ>JI?Ikr?WSr{#}-F694YJhY^Z~p}s zEmBStD*{2)c`68_kF5?ntMlI^)wA7wCSE82=tK(D`|3kBa#&~n4ZsUH@3YZTQTjjZ zef3|ITle-6L<9sB1Vkxe=u)~v5N1dzX%OjdhDK091nExc?rs#NhLG-1I!C(U-T3Ge z-{&0ykBre-?iP;RxD)F!cs9CxNjuY%yqnVu-b=;UcRj<5lt8T-y6Ys9n%x>~@lOO_Y6MI;GkW_OICn@}F| zQ{xmriJ4!wvG<%e-|b_}u-i;eL|TK0zuw|f>>au`e$YjKgV*--4!y#u-urD@<}a%0 zYV>{$Z?kPw->nV_3~N{0C>O=l(+g-U~iH= z({iFIblh>uuYK!$^VP6@Cr^@y@B|#mO2n$w&&{Yv2b6|ZUcgAJ2XENa=f8g3b^}SY zY_)sr4f1#oH#H$L;qJ-ayrOz0FRfY&7C(-ir>hbDIhINVs)gD2RSX4xPfmPQq*xXW zW3I7BNBgK61Gxjr5fnSfal>tiHiONzDER8pRjo*+^x z@`S^yrlrzUcZQU_;fKv82Y}gEOtDjG`|h7RM+&h$*>f+Wd6#ZayOi|A)l^Wj3%<;tpZnno3?&#zhAAQBt-@ zlFe8rCd^r<8ff=?`c8;8B@)k3(GmG$J3vigKSjhy4VaSS(!1LtuW1r!)*6hpi+n8= z$uujisz77g*~xwwXsL-`umUmGzGpp_z5G7Pw;0C)b}QD`M67YNy!JfgMVf!;e?1-B zma{rRxUZepFP&jr!iU<|7(wK`pkIDmS5c?NRfR?-X5AL^R=I8mysxoHAywd%z)KA) zEZbPAHc(w4NRv4``do=7*#5>v_809vjXeu7-<%|xD3&gVa62{PhnALl6La>Mc7h7DUK_T`$3QMlW;lZ#8M0@m z#C#>uqF;*u#l@cbtesiq^D5-rL@ay9$0K_ZHwX1Hr$oX}X2#|Se(`y(NzIzl48vXM z62(GN!%6z8UxkHbp7A~$-6{S6-aGA~?S`rNwFK0QCdR)L+}=Is?(bfJ2}&2Bd+DB3 zmCMcfO?TfNiV~>V=-tsd^XtfAZ(5CC?2LV^!I@)h*@`>@tN6*NS67nC-|L|f^*K!a zm|r)(0NtS^q&s7HfN_8Erh_1i;bFYa22H(2Z?`4B_4HXnZ+ zjn;ev#YVDQc3w@3Ys;FDRy0RmvvPEGLQ7s zZ7f*mH`%hJ1DU>W85UxK?oUdq1IhWv>vG%dR!Rvh4pQ#6rm6N?k?lRwu)(Bs_A_|2 zcBjhXNMn`L!i@h@)OZNf#&~(e8x82ZLS```MEzpunrVN2I|_odi8KTj?FpRG8#RT#J9ks}nhj;-u1F7+Ax6d@=nYRl3OsIIdayAQ z2ODqVod#{95rV2@n!VJy-PmhQAtQ5b2h)REn*-J_^ySq-GuIbaxma|(=B}IpAMW2{m?x+}(ASmPu`uj& zw_>%En;)(TJpCF%wTh^`!OdyNH!fR$VpwCAYZ3;KF)SPsRmAOOI91Sgn(TF+({y{6 zZjdm)MW=dCqtvB^Z^DLtdV|BP{}GqPGN^M&Epf4w}?IttnMimo!q{YH&&j>rMsg`YN}_pKVr)b8b}Wz-fJU*AI+XJu(>HrI#DhMu2k2tHp=4*nX)vTrS%N ziiI+!F31LfFc!_p8FegMko|* z?&m8rN=MOUVrweOD9aQ098eUl0>t7lfI8^ zhTnKboQ5!~^h^Su#A$wMR~7j(nxYv@I}FkyAwb4opWn&r)->!nH9V#XEf<+gFx|i3 z&!TlI#^>P1Q?Hb~xdBmmJMX>Tntu-(VsO0v8pwiM$X1?z3S>ERa|pSr@kh5MpUro4 zV$R{68A47-&vJ^#Bv%T_pU@(VHy3>%=dYkQHrlYmUp)l_7e;Nrf6*JOx_GfZRi)dL zEX?Y4YTL2TXfP*`p9~hdWF=^@(oZlvbbN~IhYRhx^YHREdB61Q42&%+czmh+Fa0(U z*`K1Ms+kDD;G4UPA4bY;2aHTdt7{6Rz_l8VcB>>kQ=PqCS$qu^$6-yqHey|Dc|Nd( zq*hFaR@#|+PXr77Sgr&-YDPf*Po`Je4w1C`*@ZKx<;t@zboJ|ENV~njBl{=L1%>)ge)M`QrD6-5G#Ix!tVIldCZJjGhw5BP6 zr>^f3>zt$k^S0-r>!n4fL=u8W~-0Lu3)VOwY>B#QQD89a~Li>l)afxb*=pL#Tz<0) zpdcGmxwc!N?wCXL?#9YMXKv*Fiy-)k`LG8)GnjbEf};@(xu2*Dl7j@D)ZF_bPF6F| z`Q?#5>SasN3vY54BJ+mU{OEL8*zy8sM+@oM`F85mXrOxEotR$$ zJhxCQAQYq$Yq2d_b`SY5_L#C2Gr|T(TyfvtYMEAofkWp3(=2*e(;dmN$tYARcDQM( zQknzzvTpHS@1_(vazxKjS^c z-e-QcI*313{V*+k5S=<-mbZr=J3t>WIQ#aUTkp>q>od?ytWP};{I2SS z3F_%`dww&jY2U!S9>d;70qr=EJN52xT7FRCGCyKDV)pYYN*BfkbKPq)t@1G(( zeK34-VhTDT8zK&>s$N&D4NLDtSz6RpTmwptB$^h9z|x_uthRRn8t~nk>2Bi*u6#7u z>X0=A;L-ygIj@)3?3Y+WggiC`eGlj6O2jNbtZ4Y&F1Eqs7O)D2IIeO=6if(+5#}gn zhU~39bt5mebOBJKzd_Pk^XMO<3-TL}ztb({xNhcXRrnT6Niu)WQbj>l{y1=2+rZp5 zZvgz-FH2D5Nxci}yrms71@IuyGUPUh1wuoXTg%)FRITED0RxL#K$;Vu_B|*d%Epoi z)lCK(m0>rRt-@0}v59OhEv;bA}aNv;6v-kIY~6utV=)*A5Ae*4JnW+B9b`ju)37#em)7q-CjkL5XtK^t@@b*#A>~Jt&btuPiNZAfMDhwf zbxOYnOo(cX@Li%v81(sm`Odj9-JqH~lx>-ZH|ztkYC+wYIeGUf z;NFPS>99}Fl&e_-CQ}vR%Fb6_D2Nx_>RT(umG;|9Nq5;Cs$~+n1}Zn;CvfwF-2m4b zB{hol=B|~2tPByolcKF5#I9D2-RQnCbG7T97vRn~E!DiayT0+*>VyStY}{uk^Jxhj zKqPU1pYWM0=C zyOG&&wozBDr{}?xsLRPl7ugyjnvqt$%sDCtGgtNc-XraeHTwMsRDDgwY;3#Nq9x~!V?8L^oa~va!alyjs8@YmO#=Nvn?4c{m45vNGb}-< z0Q=U{Q6X`)Di+cV+q7#@RK>kIVCYMIB6GCm<22vBbhnWn!2k6sR>w4`hFfX#!{&S& zTky=<>#O>kEX^nGLR`Xn>5-_WGYw5z^)?-Khb7U7NP3pY&ZynIL|Os}0GF2|fSLs| zE!5dzb%gt5J6)K6LVw@HLGvMlC|vXQ<=Y0CZ<;-3=$**Pj}T3}C9d4Vy%kW` z;t;$C8pw1lN9h0wQf?`*IjvFi_<;o!|o6FJByT`tU4)I}Dw!&(9~709#HYkZz}8iNMfN|S~k<~{(;vtNW~Ncoufo!s@;WLsUwhm-V_U4Xy;1&&?YzgZ(i z$nK#Pe+OcL&RBNi$Jj?D@>sl{B}R!_jqaaI(9%Y(uXTc=hT23eX!s!!-by18aE9&! zY2-aFOIz$3XWR*W%eE1B3ehl7Bm3@XHNEzT5xPC1>ji7A+PG5rmY@^yM(W5VtCmg`ID;I9-D{VxXATM=IQSbShp$Zj*W}nsWMg6e* zo=a6Gbuv@LorkIXY(4Q!Ph1a3XRBGjfll@?rt94sd_rnQM}>_3-Vsh4E~{I>nMejC zS6oh*c7uf~+Q)k~ben`#_Brq(Iv88jC2yJXj}4%)|xV`lAK_C+?NzZ6kQX z=jJIcF5Q(wSV%`*&jp_KXr*lG7*V_(it~7dhRDEs!Z2luk8XCD70x5J8cxUDEi3hd zm6t4ngN~eG3g{3zv2!ImWY3?Zm5!3C+T8!z6qph|XwP97ccXFu76&P@7(_{JbiJ#& z+kv+-T^4Mqd6~ILs-lxSx6cg^22LAl;d((}Sn6$zbwM1a>?* zj^Id^8^1puZ)uHAH$*N%gwie2MY)*i@E)Wl6`N|fcCN)sQzwW>we<$l*tFwdJ>Mn3 za4yPE?2`(esv<)~<-zm0I)Y)!U*hKfA{1^`qCKL9;w^vYvMn=;?82vARrKC9dj3ju z@Vuud=GK=lT&X|O+eRJk`y(E)^+p+KNHzPc<%oB#E}5czEW zSe>2H8QvUkkQB;SXxlnf)^o-QOE_B~a%mD#n5dx!3c;rZS}Z`)YpR)r`3oZyn2)EY zIIQe`NBf#~4UoW_EC^Ni)xPt!qPFOvQwW^U{0U7l8s93|;gag0m#$bfNq3ODAMt_GLsq)4sPrirrsMrgWp z)_#9EjH+xpt%v6|ld`6;shIU+yF0FkjcT#G!*fad+h;pNy6tCmxVnmGYorZ{HDKRVihHfX3= zRv7oF16FEt5P^zqKYDin6I1~wM|_)d2ikWo^jsaRs7(i6og#h5EA;|`zzoZlC54?< z$N4F(u>>6dKU2aMT1MLoCv;^NG>+8Ps2>^fm=af2}&#=Kn zf`;PuBRQQmA|)U-fdmkBhl4&a*LC>k&fW|$KNcXx1vG$oWR>|0Oz32i%vnJ%@{&k} zjI(AG49L^UONBnq$bU?>qX*h1aeA1`(}r_&V~OY$)j->Ps;1TSb_GBPN(FX_^}1Yj zPs{V2BVN2#s{T+nf`MT9wDI_CJ0lr7%oN+5yHKHH(m3Fnwv{yLt7>zv7G_!Wk(Rh$ zn8ayTp+~7=ZrDDswwI_Ud2ZCct7d0u?|mekm44bAbI21K9$od70nQ#8Ybo^_)<7`@}wrEGq7XhsMfj{N6=t?Sm|XE^6l!FsO6HNx}T}y8j7G~ zRZ`rrM~Py8I&7nXUqLW^o>fh`=X^V_tv2;B79zrbPN4Ht{W=7iI*t|4i{jOzG=$Mj z(J^c}ks&9zR6JD6nLOcRwmP$V?OizV=F%tPotN9O-L-R365+WwA_Ahd!X7AyCa%5+Oq7|YygD)QWpbr z{9wfpMtbuMg}+&uTT}4-`j}rz(d7i2O;XLcmU7+=RS#X^Bp$JSa_2*2)~co9>Y*MU zbKKHJ0I#Jy@t-j9Fp9xTxvJDQq|b*GqF!t9+F^6N#R-!k{YjG*sf+xHVg0I#=w8zI zq1J9T9k(lMT<(E0@30r)bN+-7Mpb8WTbET?OJgNR#<%Z$VYDHxWqfUKx zx#iF!#SAI>TNTlfW_&K&#w(*W@%lHH!_n-iRSQYTHtInElsWx9!#pU`0=$;_`bEF= z`e+HtzM^IIl%yG5q@5uG`T+&GCODZs;5zpPkNWMnxRCDN>vGeB;*r}8-5fV3!^1yxh zamjrseh$OihG(KzE#7o0ty4D8snqUH$NNyyJ$~|}L{&0a!OM+Y z{zL%>5pH^&@;ChXf_EeO-mCL3jX(BV6H?Q(swv6cmzc_+37HQ-n`uAv_g4SP5tA+! zZxnXWG?NV%b8u@D;#O%qHwuN9E9Ypz;y48en~5HT{+8CeaurXFVhRN|2%$2f_IJGL zznPU;-c+w>xFKs#L;_8jQ!MMB6~WW4bjfQoi0TN#3t*wT7Pjn3ecj-jTy4+4XJHn)~u3!xJ;K2Kzgx4Tb8bzQ?~wohuBgigr(kKMV?0gbHkN| z45#-!Vh4y9o*s&IG2(d8)YqA9Tr9_WCE>OdlM?14BWMJ@(ZVv}^dDjw*V*D~7P?1t zFrW{YOwD7!`vr^FnskMWEuO+I)niX0dj)@rs<2;N2G$6J1M0 z>CcvqtiBQ{7C#f^zq4g*h6*d64RmICW8gQy9wI(+QD(}rvaeA=H5O64BcQc(Wo4!c z-oBf|Di2=3t48Iq8_N)zaqpQ-rkqy`y$-vRjLW8fi2VSl zXI@U!+oS<4_JHbPS-_DT2#*uN!@glUQ4?%2aSm_Jm3t^^%sutwaSy%j)6NJ!B^(kS zx+WG~YNW^a(9=dIvF3W0?Zqfs-`yxcbY8#H``Z?UKTu*tDYQ{xY|o#DN-kIYY#&@n~rcUe12O!hSVuS#t|lNSSraZ|n*` zDaV7wj!?U{4bGRX63GCpL`0$I59O(Z8@Ig+1*M1aZpx%oP23Wdp+bgxq~r2uMeZ;1L7l-9kNFu_h}y?2 zb}h1o_$%o+waL3UDJw5oRsHefu(hq$k|)y5lRwx!yAZX0cLWIAwr_t|K zN)(}2`@u2XXH$Ftb2XsaB3oD0@sO=`T3aApnD9Cxi0)VpZCMy+7!u1w7j~#%cRw0J zajDj!;*VKhD!z{@TgLr3UFYQ&8uQDOeqhOxUKl%5WEwgfxf0(wRV&1@q-$gP9n#Sb z_Mrp}tcV)X@e<2yd>{y^d&x!MpxLdN@p#9`ZagQev%Xg-y6Y}9mGEMHiZM8zCRQkh zR~nsawOk!tur*qj&~?4yMY#c_EqOt~9&S`QJATDl2l^%F1k1Q4jFG+NA<4`p4ys85e=hJzt%hUxfQ7P&NGROU zbBV)pIL#xAC8twHak~qT?Y-B`3!_K%YwBf2?>NlH1yb!};5AXqWSp{*gdl{2=JIlrR^s?sJVkk84Tf-tb`lL-;iYhs{Azvsf9N;}=`0(u`xDCwsYPV(v%BaL1nUb(65UbYiiY zZr^U|u-ob8ma1Y&oX-!VURZlQ)Be(N)vxPd?f`4SAiQ>}Xu!`OjH31xIO}qrS%(nx z7Ew&qIog(9S)XarWl*k%q{qPUTVX(aJGE8RaGvSF^#OBQjy&eP{E#0>7tZ#g-PcH0o{>4B z!<(4S?i)5|jR`k-+~c&1ncQpQ_X&XeI_jSKgim`N$nsv)V>Tj8{&;zfzJYD7 z!T?oDDw>5F44ZqnR!FdalE(PPNZ7Gv)(R0-kb2H46VDRQZHQqW&1yY|9J zZqh$zd8=%5Br9vs<1nQ-BfX8w?7^q>rL4|iVV0<5&vVZ8+6tElx6MYxQPNUZDAl^V z(?`indSYw`tn+s(yPYFC8v^N-JDL0TYs12i+1Xaa%$r2YRdr)&RJARWSqpr zjbYKGz3nq8avtK90!MlgX5VUsLmT~uH&M%qT!aC4Y15>!K;p}-reD%bAAd}D_gGL z>#1X#GN8j4sgLN%UzmhjTh?s9MNAonukk)_EK2S=Qcfb~7J>}CVaf7#BjqoTbVC+0 z^L8P%E;83=gMtg?dpUbyb^)Nl1L9kr7L%$J7GcP1+3 zK5f+4tXbolkzxO^!1nvjXc0=RPt+me-a*C>VD&!GdflrYbX86pe3!+u`OZjbsYDe; zIpH%0B$L9yT6|ZWsKHB^eGzUnCIQ2IL^5dK%SP<7WaH{{TaT|t6n!AfO0aty!QvQo zo9;<7zvjo?6GOE2aPFe?-9H*HUVM<74T4La`8=y1z^~CQt5`lTD?94h;vNoLk3O5WnqknCY4-7KlV`UsN;V~HOxCMp2E{PbX%=nC8hZxc zh;T&;e{G6}N(C>!d&AZtLroVfcdH|*B;|Pb29?q+#@H0Umk+G3D(&SMZcdi8M8KO# zro9sFl|Bq$Wgcq2aiBidoTxF6w%5WT=UcZx`qL)GeEW9SVGBSco+ZTD;%QVH$x;{% zMGtE;?SpNvNk=*npJW6_f7&Qaui1sSP}6zSX?&)GXtE+xvj^(8E}u4*k|ob z^zWE$efyly++o?5(kIJD_5spH^#)?+xhP}p4fL-aV+4=b z>ji8s8rYnlH5KPwSRP^KD3m>Icd(3Fu87Wh9cV264f7TMFqI!}=s1niRnp&Qn?Fgk zkjSgwK+G%NxA1myofiPVOxaa-$c2m; z(Dli&bJZKhDHHR#r(2iZp_0Uj9M5zis{KUBV-d%$QDFv}@~Y&L?wmz1X;umt_!)J( ztM>5Lxwkt@{JyvG7rhs&PO<3?A!HXsm6B`NRip=nK~V@He}GrdnL`qX)wqEA>+OY3 zz5&&IEdr;;uuR@a-ZJ5c1ezg1?dMoBUMmDEeRViG;#+2&2tm`>nBR+q8V`IytBHQF=cd&}=3zs4mB<_M|hKmGx+Q zLDtEIhNiF?bP*ctL>1UQIzP2LUO@vas(^nTy9MuGZ?V`F6xj1OFopjji+}CUfBXY* zvNxe&?O<&7-yHYyoKA!ScH<#E=AUGv|5k>88i3{C9W9q^{Kwb+9thn)yAJWcPxJV1 zPQ&qM0?VN!7p(l7<6hpR{NW9G_$i(4FHi1AdjUl#SPpDHt*!F+dWrwo*N3+$z-hYQt|5FI}KhOVXG0FRXp8x;4n}5GeBL9Oue+Ga6 zivAz$`S<(sAE1rmf3WAby}+0M@ZYe7?6Rqhpo57K>n9g^nOKj9ftygtdifEDdg23c;+8ze~)QP zZbDTkIzp)==Vwz|?AlM^^X+;VswLzB*l9O@*$D+R%2=skI1s3SZf;)csv}m=K5R-K zWH?7L+fZ44hK@rPmRNTY+Hiu?4mgx9&gDKH3yi$qhj@xLO!3>$)GLq^nT4|tV(4Xs zx7~iF4H(RzW^L6G&5C=4n)o{%jUf@Nlj$2a#aOMS-}G_~I7` zFa&SD6J|tAHqt{`$Z|nM8d%2yciisJz5;j=57CT!H&s2Uo@|*B0Y~i00 z1q@z)5QR)Qtv~dJz+$0qZ!^$kC2LR^5dRtDBXW#+cr6T^ow(EHcw(b@=09iM&hlU- zxGP@i2z{BU7}TbB{W|5td&OQCqDLMrM#f6lDS62xUpfBkM*iO=hd^fUK}(H$b$FeZ zDI|V{9Fa`I_C&_21n1}HG7Z7Rj!X645I|x_V#i;E3A!Bwa2eYh^*thXEE{9_{RXo& zh(IYY#-MCx4~@}(noTp6$WuIAi3HaMQOI}`vRPEUOfR4e$k!9-YwW_PY2^HThW_so z|AoPCQr^UamRet28LP6^Z4V6m=iOvW6KQlmg6K{Q#M+LqtqZxJ!Z_{$tqZ@wNspib zRY$R8Aogz?n%kvN?t4S(oGt(vNZu)RZJqI_p&96TgIZ+=ER&a@_7KB_0@m+E33egW z@m^JD<4eDEgOZPSc(T}^ks`%3zuyM;tB+7%=hGzvIim-$(f_1bL^AQ)qnP6pTsu2r zt(w?Yaa_tmO`5p!s?DZochyT9J~kju?DLz1eSUW}>!`hF2P{=M6` zzd>fPA$>-cIOIZ1j3bs#!#Z2*inimJPyg)gA78IthhWgQIRE@n?QEAZ6#2vr6&4gD z5=I67fSb(udz$5@H=rhNQ9Q)@wR={;rG10DqoTlM%DCb}4Tg@Hd#;6qLOk2N|Ln)e zffU|`E$}D3le+z{SpZvQMzX&y>wm_!%iAJ;`F}Xfn0`HNtQ7c} zhoo}YkcSsUJ4Ga%#4w!LKuBH(6t4-jIqsLs{qIy5DoW9EH7sAPqIGPLfKgStTQfJS zcuYp(mp6G%3%+QbNr|BY{OCSlB5@*NVl*KJsIn93^R551(O)0`*B5=G4!Ms4GxFq6 zHR?_xD(>a9%9RpW21!%YYj|h0+JnD+0IBt zf(QBd4qEY&_iA>&{MyUwpEST`ss;G628)D!=aTvT+hQrGu9HNIgc0Now=2!jpszP^ zpS;nH3LEae2RkawK#9`(=QF%{4X#eflfe8NC@@O<=fA!4SFQ#{i%g=zPJP7d0n<&{ zf3$wOa#mwK<9X!KeOJB2y~bjqQM)sOg=zs29hv$#KnjSj2i*1|evNTlxg;a`zFi}b zrGVYM{_^)X$l!r@^|i1=G(SkRDI}wddaGEg%&1^@gzv21`+!C;rRVz7P$=dA*^JkuFAh9F zzB>NT#r%KbL*MmuP?FI7BDuEsF5P`hvCkr_E>PDuTrFa)uz7cVyh1v6$|e9F;Vy*4 z=l%K$AEkArnbv->b56UqZjQYALT6>Lq=CQQphL0mJ^lt}g0e_wG~<$7nIIS?7`lK^ z9TlZlNPTsiMXQ}y$Zq}(lTQ7o#HTu!Mak@8RxP6xT7z9QfdfmnZEtk&=VW7|h zpYGPIY`?4^bD9r#u4EBKyd~&`4=x{N^mavqSx%-7W)NU} zNrucyza1zWbiim64T$ik_K`$q&wQ+&@=w@$t4qGdT#^EjZqnt{bQp*W0WAa`WU7e@-YAZd_ z7q3q3HhGt-Tc(o=G%q5s?{QV9y^W@0F}}^Rjx+GX-a|M+Oz@wvr2&_0IR8%oo5|IF z<`9Ff!;VtBg}cV_z;{qvl3CvF-hZu4BZ$zoj4S91gj5dDm`}UTsqo+Vp5sjx4(6$V zXzJKX#H=W8HdH&6xKzqm=ybxVu#pCT*hJihgfRaQy~yv_#CRdgUn` zsJLe!@9uGiTzeOH&EQcVarx^(M>E3vk{wTQ7K@TRVgdH-HsZ)&FmPgnM0 zlO4e_qejMe@dA{LPYbtuXtENWa&HiGh_KYHD;}-y9I-u{eKyv3hJ4}T@T~upt=-mV zBELY@qDw*0L-}{uu||@UZt|8ZqbZ5=U9kf}c0!VZ4>dBIcHi=EMn>yKWVHo_vo;=f zlET|at6*ofxCp@m!R4TVv)GdkroKO7XRVr>?{5!JAgj|(_E+euW}Xg^X8w4+0fqVX z%R4?N^SACeEIz1Va)X#XHF<&@k+iGPc7BFFKfacG&0>{ZzYUtfSakgkQHMxADy*>j zlh9J5Bf9KLgXlFxs!Cib7<`O>c$hQ5&RONNB07ND&LUznMQU;KIOx)TFpp zgjf!Wgb6jcLP6W4!C2|c;aZ0?~ zWqa;IFu$bhiGBT5@xb0dPgy4iKYjwI!C_xyuzc4G>cef9KO7cV=f=b5Sigz@?36!1IO3p7opLGf*)mM%_J7X0$3wfhjL+`-!4g4 zT8z0b$p(We47_&pHf+`ouJw8D`(y6# zBwy@RSYA~3)K~BnmMBX{F#L2-tW9uf73LSEkx5wZDH2>jL`E(pd9-%VZhb3{p6}?B zuh~g3M$y(uaJ7hS8+W{J;=72(F8fl?;$YRh2=PUaYUBN2o@U!sFB0?ZHr5dFm%i+lcvU0r(jcyc^Is=&RxKsF)kR=9rHEuK%Xj5o#M+Boh((D$@ z4R-O{e7Jy6mnPngLqE@?-UcIlkZd=oL{zk^eYxiHk}d;Io9rV97vzX9nR-!Z)&s~Rn{u6 zT;Z~AI*P-$d7g#N?S;Q?wls%-3DGEa{t^2AhFc&R5y7*jfZE*?HWC+}>6l(<>7J{5 z#vs9>eH>nY@j)i`=&Ph&d`QXXq53_-m8m>Qkv}AXgD5AC_3wRGmFFM?-x#MJva0?} z3je$i-En9uU)KVP`AD6b&EtEMwyL@+uq4k_7@UVE9!7@>fKkAnQhAnR)bK1sYY-U?}~|fM<7p>_os12Z;(G+h2hI z;7C*AxbE-ljm)%%Wk?&8bo^23yoEPIy%xWyHtrm!@9#*_0iwq|hxwegiw#ZirBl4s z;_GaT(LoocMF9hIBMi@t?NOpeyk~Ead5TiX#tAYApB9g5b2*b?lek=*;m*ffH-*2( zYLeyRq0JBpQ`Pb8<0w+<(N?vQFGi?D7DklF2Rj+Lfq0u?ec`=H*&9eo6mmmdAJPglx7IgZXY*pbNApSvw$mbEJ3V<6 zta6DY4~lS@4qa2q&8CJ)XEGh7b99k<#?Ea)q;@T`ZE@DH`9KJK$|E^nvub)uegni3 zPTqsbvc%aSd^sLb*WL9wx!}hX9jB~~5tSQ!wT3wN*r0_qi)Nq|B94DE1sxfUvk`n$ zhrj$I2lWB%0y5#H0>cwM@$Us>jb0rO zgGE2`Be@|Ov-%HI)3KRC!U_ye1!hS1h9|7!pM~pp4_D)dt#kwe$}ET^GbH5(7u9d5yg^>ulDqA(pvC!H`BKDlvb&_ZnIwSXjQptaKJgiIAL(@ySzvb~H zHokvtk*f@$SJJ(~E;(yrNohFWo}&15(jJw@^UXp9F)Lc2{*PRp z!u-d==Kj{g!hFVkiikS1X(bTAF`e|p3VV6hG)fs>V(0|BPWr|j6Ly{!-{1&ddm+b{ z5C;j%=?uhNA|UbV`_t9!;bTsg6YVXIWDdUYjY;|;tNgEWc`j?r{Y|>$a5cB{qb2ej z#XSjl!Zm{%q-@H>+o9oXMQO>#3L6MY3xjm;!%3>0&8Ab^LBVTGvz7 zDM>QED$Kg+WSWI?#;dM*Vpb?8r+6$N%x=VdrF&JeNkK`y z%BrDm?qCKU3^eB-{u$|hAecG%zrMvFd577whD30&FndD)6~@46os-nh5R>%zYB_xZ zvQhVTe_#-s`}gw+#rZ(?(wOUY0hheW&3{mZCFI~M)@A8${dmIv-!I;3n3w#3)qGd* zVrS>fNP0}{C)S(d(TSJv@gz4EYSoZa8^Nl{Fm@CZ!-;m6(?y(+RR^zh2h6;R)~K+; zCGNd`N$+^)Gyd>-f`4Q83cVqH7B>+THyMEx@JB?PsuN^cO0t*Be=5rRgL#S-|3OPa z0ZnS${U{$nc)&a6nJ@7@^v#A5)en088|&rj{$4am zRnY%6*QH(g2V-Re2*$r+tgeQMr?s;E;wW=|kON;Ke-AFaGyKKsDE)vyEieig$cQU1 zn9=0J8!I!$t!q}W$MxJ@ZVF>J4l3SxaOKL?M*z@ZaO(Yf<%Ge*AMv&sUoga= zyB7QTrhpO};6iS7B=W=GB6jbwX+#yNQ4{GnCpmbo_uXDH_7(9C7K){1?RM70B;=Eo zz+NTbdFU%flBFMmpD7c~+IaeurLj7-+9Wx8PRI8(?VDV$V$_p5=XtN(s<3_puN$pvd%gvL68Gc0-v&Pe-`M{5_6)R+1 zdlPgH=wvnFQ7;#d&q8T%5uA}M^2jM@)piXLj9UYf3uTS_^uCwH^fxWWPixdURnr&H zY3DW6j%eYcKvG_c^mEJUCEbp?GNI4{&2zgTg_nNF(0h3*U!Cv25P1b|iTL%;7Qas4 znojz`U`ri<0KLtoW` zPo0@cVU({U68EH1K_d!VN7eNP>zPaJ4br_pB?rJIr^BoES4!gEbPBj1zOgve4`;Xb zfBx8opOKQ&=#s2tYib6yQG;)T(FTneclOqH&vVS%q7=*9iSLJBj8|&!pO&f03EXc= z(D_~-veFj>EwW3SB8w8ED8rCRJdf*2;ult_Qzz+S~?wF3Jl62xNb1n&C1Mh9NKv4_ZG{+W@ z#s{)DJro+A&d%X{K_r5)ufUIK|fWDbd5O zqTdq798a^q$9s@qN8fY48cz=95I2QP$T<0ER81v(JMFqjcN35a+Ei^TF>bCZ{f$P` zG_Gy+Dh73zFEG&wAChGBvv>v@V>5=4f`6;DpmOekab7 zK6gJ#m=){J&PWXMsuz>Sf1oY&hf(vFhou%roiXTfnW{Bc?twl>b+y^g(K8{zLPtz0 zD27sr^k963)DYWrtfft>2Ylv!h1v9c;%!GS*QvBoOW9W-p^~>yKOAdK}vZG30%j_NH*94Lbwa(<3_KhEp5| zUhw&(EH=8W$TvwP)z@^FOQ!qe#A6kTU8jGLC}BeyE?0aze)Hc7h*AdxWhg&u0V1FT z0_Z|%IX2=O%ipL%KyTgGSFQ4{qK${E{PYg4qAt%(kTzuxg%Z)e5kSLw>O!A<&XS}q zXtdm2*4HLQ$0l4ciF0B2J~QEe?6e!R)r&X?h~gP|iAvsIYVqhpn~grm9DZt_W5u7H zM`JChhZD%u{(Z)~t<_vnv^PMqygqz*v5m=3NRD5q^8i%9>usn@Sw zU!Sbd)TmtB3^X^7E3H56$3o4_a-^tv;b(%73wFCbMOathPX@&+U_{&x2@}cN7c}rQ zr9?PYO>4=nPI^U?1|GA6RCpfKO0P$^uFEA~YC)<$UoUaqNcO-~t1X@XAtRL5JLE^P zy&VuF{icq$e+2MLw{F1X?Y`Ju>+TY)5@?iPfN<9oqv$Bnubt6?BLT+ibJADZk$A;b zl4kujN%?(q$w$id0ggFd`}vUioYS5)P3du+%1qjBt7#!w+30@H^D=l5iQsY3J*Jzm zvAa*+$KK|Qe}YpcJZ@MH7%#VDfRz>BH~WXKD~j)saJK2y@#4@XD~s4IbQ1Kj_4TC7 zCV^^!q@3VhdhrZ6a*eZh7iR}7yc^}qY=R!U9Re;}3*3p_N#A!L0C9tLGs`l(2^NrE zMaF60>rr71&&o1XuL`;D=7gM{D{n}kmER1Z-mPB4x|@}FH&N8aySCR4hwSYNyrYn+ z0sw?0tlj&sG+fBV$1R~7S!Y(6`}~z>$i&5Zm*>je2g{{dNqFZ~A>$s;JTECBDy$p# z1a+dAj;O~LQD$6`R4#KHxCv9vc52BlXe<;@-b4D$GY6yY71%Oe7(K4TX7Uq}xb=oB z1pGDay@apPye+*HVKZ4Xf^T)zJKjxL`b@}ZOUGeCMO*{&{OSvHTVAr`|7Z$-tAvOQbbBbL`sGRQEF%q z>6Dfd>FyXB5s;EDX(R^e?hZv7q;u#Vh8~)`Ip=)U-}5{7-shP=_OsdSSnpcv{j8=d z-9mEx@@Qo+y#ibtVvoEJ{Z@L@HPXm9IIX-s!hW2OQRdJ_X506MXC{tvW{BU_j^3_( z{(4(mVb~q2I_j~>jo>;KW}LgpUE`yhwwAWJ`SMAm@>Ynn7%gw!i)%CV{*E&TKg$4b zoV;?%v~qlAYpRY5B4ytqJ+uBUQ@42Dqui`x?glSH8Xv9+r@rRR*3)2AYW!vvFcBf^ z9o?g$yV!WPDL8Px=dvd0xH5zAjvccc=Z|q%9bl}PSJkzc$fsuIKZ*5Q)30dOw?B?c za3hIp=_jM6^gIn_VK=s&J4oPwgASWY9AT--nPM+DIT-RZLzP~i`>W!!>eXpM6CZZD zq`Gr>g-fHmuNsr|gpK8+6-r`0n^H_wq(o~uUu$5$7&%)*InyU!MH|&|7(R0t!Bxw4 z9ey@YyWn&56}duU@)KbRu8ld8pZs*Zn~?6U`5f3Hz02_gX9MI07orS}fk`LOt6BD6 zMTX1>Pz`K10+O~8K2ki_U>rTwtQXIk02Ed7ubSN(`$a1&LsJp0K2c?v)49%u3qA(^A(j`iJ|@PgRie1(`pfKyjdQz7e+O$&5_mX?yRktu`)Gn%*ld z?jjOGk*~HA@R=6HByLx!;ykCi26aR|CxBuc1$GeXK0XtAkhnaw0yvuSWWdpN6MR{E zD$*Tdjj#$$H4*kWB4+}|f=4y%){aS&6~m#(Coe?9DTKDTNwQZ>2NUOz4KdS>zQqur zLi-jk7=O4gXxgEARu+wkwrBfyxiM(5v@4uGZvRKDyCK*27Y^om@yewTY)ogMv zGAu?xAs*zP$g`|HC=Ye2{xHzM7B%02_OMl?9Y+9p>2!7(8la%!sykg3d7^vHdocxb+T zJ7dgM1EXDgQ%w?OpYsgr(*4?gthd z0^1#-?ipa8uP%JeK6$=e_pcwC4ujKF`ezqJ6QdQxcg(u7l%z?(9Kdc5NKJ;S7?jk4Cm^^ zy*Fe(znjm@=8THxq_a1Zt))Na(JF;p8*4~8f$2||Hp427;qY6%bQ4@gG`A+**lXE> z@$NnX#sSI-P~Q`tw|p-``w^nKoq?Bsh04%)XwiZp*T}hSZF*EY?Dv{I(=}PzZDIbV z8_s28w}uNz8Ml3bw}u&x)!?O-Hv}L{M(z4o6$P1_x3XSA6&C$TrK=cV#K9DaUa$YU zoX7f}ke&7KuC)v;x9X@eQO3#ENO%BU!vAMZXB*p|#I9zmz& zzZ$eXAk)=wpI*b%XjJKQCtJ4Fo>+(6sWfhK4_;f1Iq&VTgI2++zB4nwqYeH(g$HLx z-jGaQ60?Qw)~|-M2*o|f=EFnv<)3X$sHg^_+)FJ4NsBouNMHRyLq2W5g2KkOnJKl8 zaYAs6&V5=qU~B%CdFX}YH|Qi`m)($J~SzeI6_+9 z0Y&*7p^}vzei8!K9})2v10G1-nr?vCmgHpNTq@bCLcgjDh&Wx5sV2H>Uv2Hc`aMo( zR@b6LcBGR}6#*E4UJ|2^MyN~mOC);q9siW2zHoZ8;B*60ZaQ|{@rJ8v6mW~KP2H zC$??*DZ*{MNUwBe>&j~UTexanbu=V)(rcd&z3)B7M}+>zT#l4l)F-OL*=qE;1MMbl z!wUm|u@GWHZ+<298kSn&OtHAt?X7bj+Gv0(cFBh;usK(i%F%7YGF%^stw=k!JPk-} zIedmQMz+5x!(~vxTWQVbtO#Z`z|(ThdzJdx(?V2aPTCG-{8g8vJ2{aj1#WZT$xQsh zG5co`cPt=lGBOS%_>J*jQMFszCf%8mTJjMqcc%&t+3sJsi@Z`~`f|rtr_ND2isK&t zayvRhqiuO`c$zk?xSIQQ}bL_TDSZ ztzyYZ(}3d5Hf|jhUYQDxG`cuR{gr3xv)+Ht>*w^4QS%U(o7;Mx$+j6R8-IONP%%jZ zCEVmwq>V1Z8VX)KQ9+6$rTOQeTVusBG5xPszpsVmuFFf{;OY$BYx{cDz$YKL;feX+ zY3zU?Mu_^TIi$pVKn&*p%Sse>RoaP*HEiyU7vSt^m%AB~EId9&stXhFZ)RlZ2x%@m zZ|p|RoU5hBtTPYmAg3>~({4O9HNB?8rHBoEE&x-_e(Esa=QY(Ro^_E!Y}sbp%7=4$ zxm-2cU1ifCUR>-HyaMjGXFazM?tXX}uL8xG|NLZjM&9SCHB~UUbm50ZT-Rf_06qtLX|)*ftl;j( zsCWgAyvy1nQ#Y3-PSX}3H$}Z3Vgt94bub2iH0y*mc1~C6z!fBEN07dEcCaN{qhC9; zTE$N?c|0yOtnGxH->4CrrEJO(<7Ah`G_jDj8b`xzTQ@;yBmZJZ5c=$jWvbG+;B)fn z$3ksnpo*AAjQINHJ!xAPaTVX9bp!dfiw5k+WJUu-)8_A$HY)~Vt)?pKNL4chNhW*P zi~2?oo>RYa#FEQ?A#!mS@{vPx1Wjtk`zeDf0SWVV)k+mZs$f!g1#Wo{R&L@@w&cSv z)gAC6y9rXR_xXN#jg^{GoN@a(BpRINaTWZD>Cl$wql;lTzd=&HWbjTpg|Ik1FCLj? zGG_F85@Lq$Q2TT*#?1yl<(dvr;FZild2<^iN|au>?fT?;h*amlno)0>u5!5IiUay9 zMR)J60db}j@KoUEC(H$X+B?R*DaO|rl}_>9)=A)s3B0_fTQFdDSK8nV%&bX8_bl=Q zY82qaafD~H30ztxfvuu&nyd;jlVdIaycW!!TIab+B^rFeF z_tDv9udUp8-PypUGsO#e^;xIF?t2~)(K_n1rmqHt=Rbr+(}s2nLgsDCN#dg7fPBB1 zwQCKIsi^^vLGpg*TGlR)xsv|5c`s9W)0^^SOZy_4Ad?#!`?2G|29n`EHb>!$o+&Zh zl;$!EVn+3y`)6kp#0nBh$!b-yf$=D$b(XEE!P~!ITI-?#nV1HZUk~1#pWkdlzp0U}8gM-^M~qWUUmk2gGi}sYm8&zmZ~VtGddIhq!44?k zCTPM=B}HDj?h^Te)~x-7B4nl0!TcTQ43jeUMLJKoGdx<6>LUXuPk3g%YIll8e~}GT zr(W;`&LH4!*7ezrgNCF+K=_ia@u=QYVBK*Y>JwRKK5imml}oJEBp>rQNne~pX>#-{ z-<&qVLQT4R#Bi;+WbU`^;yO_ZbDfDt+n(3&DegyjIy8>J#TZ!4Dy@_#+mbhX6*B+ysIjSRG%Y*8_wnW?Fg82rtHpwJjUpS8_ z$~Mle*_n%4lBoag#5`pxQgtV=jcIMRcUx%FeCYK(z(YaRI?Idtg1evA;{^ zVXe)LZD#O2`HGguP@=X+yWzR~grJ2K8#v>2jd?Ei7qhA6zzVl2d2vTLzXi{gd0y-I zTBiF66jqA(;03;6mT4Lx(oz^hRh za{8DoyA=B)j?Kih_NH)cdfozanrhp+ApL0*SKUw5gD`ed*Ge2A(@MnnVBY1(zT%hB zpFozWV~spge$c17u7rd>TtO`vX7lP_v13x5Rj=^CQ`*a z^$O^Gg=@SXc9O)H;{fM!v|x|!JhzRXlN2KO$#SazvR9|O-w}Dh=4)K0#R|M!Jq^dU zQFo=CDXi0s6U^P$KZX#m?7Q^C4Qxt1Y8@l_vvRvk5F^_yz(`(<6xb0(wF(ktSN8F| z$8M!g;LNsn|7b4qY~E5~uF?#tmmohd#y^)FB}Ul@Hji>r+(DxH&xdZ536F zcwn;jZ0g zd_0EB=YVx{kguFVO^#QSWD?NX=z6`i8Sj3g<0hzDsj^&RA-=<)~PTH*PI1QhRm8&3Z8U z4P}7rQJe9MjKB<9puz+<+|)~8(46i&lnM9UfVMBG^ayu-BNEPA5;f7ZN^_iAodG|; z439~Dw0qb<8Q&A5sAxy0_>_Ov?JnJHqf2Aes2omAIw|QJUj6mWaYzu6nA21ai3~qc zXNE2S1`p+2!h%)w?0bgZlNT8cha|1=kx3Dz;I=xVSW)a)@DF-i+){ zO3c0PKqgud(27E*2+TFd-f zUDP;s2CwI2mb+UK(Wc94Byp&Y#XOeV17y0amwpnkXHPq&B$vPh1CI7KJ^kQ<>w~*!Wl^db^lQlSeL~ zLC@5d2}q!vzfZnsqugXG24G+-Iuz@0qhFKRfoW~qx<*L`q{c}`!vkbe2FxFwE@^L! z(TBZftl7aoxQmLwp}s{}ST*U0!KCGC`)!Tl_!~d$`XL3_?#37`W5Fv}o8r%qxE5K5 zIA2A){Z^8DPkeVW-YWVZYHxVVJtA+!HW66;CUIXKIP-bzg{D z!D_Bvd@^{vx|bs6(^Y9cA#p{F-s5MiH%6v^wB@3BO#+6{Dnff4_qb|jZDd-L^ zlE1=NDGwK9KD|d>*LAnI*eU{|5U^dy;7baY@mL3hWe1?QKy-LgRCt$?>szWyX#{O-pVFlnk)g?_yvaUZX4 zMYNlrBD_l3VZwS?ha*ST1`ztP>*TnfT}jrGRBx#Or+y`1G=BKB0<;WF>`C|5-JMl| z&sWUoVD54t-`?;{(KvgIJdr8~+`)9qvkayaO~~qZIJPdRfZeDCxPhFjVS=9^b9r+< zy48D z>wXlGB2Ro#-$i+d^}&yZ4#qLcDL0jKo9WGAjA*MHk6aeLx*qeMEkr8+D&_jqiY{yD zIM6GS&0rHpv-?V6#|6^|>O^iRxZ|Kj@8WB&g=8NQE6`In%B~R!0n%VlwA$)z zVJI+jud@m7BqXl_2cEnp<-PwnbYiG0rdb?D)(!UyJS$7O4F@HH!(K!1@*E%h_8 zj~ngrOl3FW?uLJp+2wP*^XD0PLa>*wFy;f>n>jGT zDi!8yjCMNQTRuE)Si2vB87>?swOIXQd`T__gES9&M>2ogwmaj+{fcPu#dpVjc z9kQWUq7*MzjmUgf?|;GrC#MgKdN?BC1B6yd!TaK-E)8BTQ)a?d&#F?T^9u3GqI=BBW3>(zM@28GE}FUilI6E{5-P zjVyZlTlI2>mqknija-bzeD&tl+A*M@?oimS#V;6}H14N0f*oC2c{^rQ!>e1!cUfOt zoSe&pb?A>N>%p~VpKe{Aa;&9RXH0HOiNU*EiV*|WE3=M0br)78aqZ~QVjlle%m8Hm z)U{%KkWe3bkGq6j}v+FX)YVnFw(ijGRc)_oI%KWk-WvO{>NSuo|ouW z;}7QCy-2^$Os$zar|4BU-ZEjiRfbgU0xrKGKzS`;UU(dMl_tiEwGPmJgxv^`jwYjS z*M_XYw9#P=>}rEEY?)OudCFQD@oGKiKEm~!ASh5o3`kpl&6)96!R|2SlizK~A_>qd z!+M@VQ+t8rLK&Mwz3`{J>`Nvlb=J=!LC(G)?VDMXaxswFQl;%-kdC{OvUnIWblKXZ zO*`*+(^Rh~o5X~tLc!fcOvIza#Vt8WJRC1;%Md*CBYUQ!noCu=aWz2*LUS#f0RNi~kgMNuOP{QYh; zrmI_AHgxg8&gFVnXEyb!#Pk5CQGofwi8yR2;dlqa`wJy}ftYFQ+n7DT=vhpTz zls=c;MV)VEhF@Pu%?n7&`%QfYEmAzp0eW@r)+0TVb&PFGH7a}QG+rhRqe{E~D{fAA zWz%>8>p=^Izj$RZPnYH7?qC}QUDbI%QM7)U-cOb->wPZXd|{tw1w?pKLG#tnxhNsQ zlg7*#ppidaf}VTXLuTc-3QTsHBXYWf*EFd)fH`&2pqVX2P5QDlS26F|NzIO+>4t69 zU~4v58S~cpaK@RwBEPuXZ?w`t_^%4^L{*XGCep(`kSXFC;(=R-EKb@@n?5Xn9lz2l zPK;*O(6G5HZ2{9%v4AOc%u&p6c>lsV_^ts_MJ3EBMJquJWR!mu`J4b@x^~)r04+six#`=GR_J-ru4Y zA7Ci_Uf1tPD79o~l<324H4D;pPvncm;%i9|+3|W;OA=*O#{C8a=!f~wM)h6^y6(P? zY`P1bM6@k#XRneY$KL~JOxoQ9RCZA?ArmSm*f-enoTzDj+LV26y4W`2M7i3Dk^d(> zsyzWvxh}0!9TfMctjkCXNYhaUy6fusf9jv%2bU1b%k(~nYx{eGt?mRU?5SH6!S({3 z;DVirH(xIhu}NSJytnGfr8~2R8D^^qEeWbQ1U8U`FO4I)qUT z%~fU5V(16Z84+K{A5p4D^stR%8p_g$7vQ&!4Yr~z2N!yzkF5MKBKol0AGD-9Ak?`B zuq!3>6#Wvgr8(RdbHFJE%D^ULdn_RQe6y_@!e7~WFTtzF~a7A2C1s;3jg*d%KREs4=Wy2ze>B!dJ4MX35OpY36 zuSym~awPHzb{{gvRIQEPu;11KW=CJ$$QMV0feKtS)x8=Eri2SlNn<26moF%v0D#DP z%3^nGFv47?yOpDLA_H~3Bh1@>V7Qza-%Or%nYcBsQ<=KhQgsJb2*+)fG_HO|$|@C4 zLF$j6ETKue4U*2ASSE4h)2I~}8G*lF>C|{9j zb*wnp+@uJIUN$$Z;pR6SuUROsUpLI2?Nmko#nmP2Hihb%PXs*E8w9JnOj2-200XWf zJ?|>nIANK{w-1DGmpWA*jJ?tH;yv=0J3LmW)ekpmf;gN58IJyOmpm|xf=xF5GI~vf zu!+rPHZ!Y2Z?X8MQGUQ3QFqr<*^S($WUvEWPpsjlo~FCbJr`i=Mqg$_i6ih7rJ{>m zdpwH#*nfHnU{NjH<_vD#mA2ZFLmx~2(QpGv zj#(8=-gHHDD5gM3S*mN4Ws9_vSkM|=;4XAvTV)LQNX&$W2aEMp<7=cNpE3w1) zEyp4~t|2>cE$vTDXGmZduTgeND}$1Ljgfa0tsYq(s(`)Ftgy-AU;=i(`beRa;XuxL zg$jZD+;}Thhs{8*c<6->UR^`d4(oa{%kz1go=T0?7{!=yz%ef7r+Yo!+~Fn*=K0W} z{c8H#{MA}mie>7v5t>M&Aah4;7|ZBr5sf?@cK>tY*(uCdZRQDp=Vp(uv7=k<^K{SX z@>FG*O+gV$vF0*{+s6Ej8(6RTLO)O7>P88#rePTLI(Pw^me8J`U9&(#8*vf~Q3^w%Pw*6@Vc5aUu@W$wC zy-iDvY()i{8plFe`T2cay1DSyu4V{}&B<(kaLPuvEOyTsn$->Owm_0E zb(%&1Gft~zsL)xt8hV>2#429_omWUswt^M4j1+4A+hR(c>t=-G_=29RK{top!|96n zc-gSb=w=?!MfQ}8ys!At&(GHajCj|jm470zG7ztP6_q-i95u@Jvl}DnbWV@kW@>m! zqJ@w6Vy4KmfAzp@+1uhoZI=TyI{;eg6)^yrydR#caPg-VxdS1*N8Ve10|ouXKP*BzE~YQFj-fg_^GyU$IA+ zE|q=h?4U{eB7K;$hNVB8<`rl(e6#3!Sc)jNOEeHhz7&hm9k#le9JzwajUKg1Suc zo;X_I=rGGg_qYa{;Glv`Ya8O5?o3U#r<=nQo8Sp-W(>>mD+c=mD!a{SO6-aJ zw53F(UeX4?M!oe)-yVMS#`wIs2nn5Yy~Oig-EiwYO@)kK9-txp^gn z(4l+Mqi)@}hoF?krRzgcPal2;o$`T{ipkR-**Ma*J3HvI=3NL6x4535Yt`bn65fuy z9@o!U&umyz1tCgnVNZT(H#hVIoi1cZT1}QUB@lQ#Q;dH9;F$SV>L5%{DOb=m{l8?Q9OShIwmb8Pp+Ph;7RAJXpn4 zloqsj?kAo;isR#w&cbQ+UQwYMC)^+8`3`~*p2c6aqxDK>Hud*q_pRywL?&7)O+{^E(6Kj=JAU>-A-q(O42wVkFQ(tm^> z`P8bm7MMwGEKA1ZzxM8h>7w#EY#<}Ta`r(e?CoTxn3cfW7ukB?o@mCb>CYppAHOr3 zZJ+2$G&kF_s4i~%`5pUl=a#Kz9}8bw)cD|TPh@{+C8_jWMLs}~p1%jx7~%?)POD%U zjdsae3M~P0M1`M*Z z88UsleyQhdz#nv}AY^a^GSsUpZJA7B z6IOZK;^+=>^FhB56ck^l-2+_B>i0I3 zfzFxkndK|3l;w1vpRANNSj0ycke<>6q^r4bEpIK~<1nmY79bN2bqzL>C3-p8-Jvk)~Hne~J6_A=4Md0fqh) zMn7}y-5q+3J){ql5VuVRP1nuKU^ds%I`yg^)iNvTkv{b*0+)4m*|(HRkKXd6et%G; z%}LlED_g0I#ZWkot~$36&+pG1KGi;lhpF*2m7KnMc9B^~h=LgP>2c;D<+FS*^(cq+ z(&5L8W75A zB-UqU6W9(4AUN#YVY1u+^=%xUX~9QARBOg_C92O|EwAw57TP+RiM1|9Htxt~Khei! z^G;rSUU@17P)_@4fitC$x-nuuH~04h1lsvZ@gI2x-5W*S*%r86UA7$TQ-x1F`jkjd z{dAfe>+-Baf4F$>s*y_es%>)*L;JPmkkri@2V6tdj(VEk$K+};4c;92zX(fv=Ely; z%<}5;w9>}!%j)@6gaU$8fWx`w%azM+J8hOM%RS};b(FT!q{>M{qxa}&B6rXX9;0|a zczZDSWO2)`W^=y4B__Z#b*3ugX!BP-Xe1g@M&@%p=K3Iy)dX+i;h@;l_Jh8ad|jJl z-7z_Wamiwxx&h!d!S!y-=S#d7u}*Vl&fTp6?T27yvw~9`)R%rA(qS|f%sM~q?2Ze+ zpM7<6aX&vIEqZ(5xA~4E+Xvi}I2V{4YY>9nE}rEre|Yf@$6h|2*CU+!s$9Ola}?I$ z=OGz|%YRD5Aok<%5i3U_Xa2Lx2e8`f^---_uSE~TUK8AJD_xY|tm^LfJG$BpUd_G; zl@_hP(P85$9+($5(%qkX|2>IvBk1v^#<=WR6L$kEv)94!1z-kr>GVDPI76a=cme%{ zAN&aIl<$o?P0_CQQ}`X9E&~z{7KyRBh1!d&pqUkGZXP0wZ zluw%{7$>-afgR6W8Z2;O4@B_l1=;d?vOfDD0(dJTLFK`Q9=|1? z3O^K^fdHG$FeDO`iMK9HuNc!}Cm!=m)1ds!sAA@+_SOc!S*08$Z*2m8uu=T>CE@wR zevE^lyzmQWrw?* z+R5W9gdDUzBWkI1oa}$5)6>&Rz4zITAgEXh55ZeGJ6wf&x5t)Mw5rFtzGt_d9<>Q> zO{wy8!oP!pA5@Iy>cY~X=s}l8|5hwTvo^&MsTql>b&C8x!|yxi2tAkAe@^XX;xP8S zrfU#aqj)Yuter(5^v#@`$&&lxS@L%{P%~u(!cQ`#Z>HQcy8ECf)oVV>eP>x77Icmssd@Y9hE4C!Y{ zbVh~$2=y|2a6PBSwI!6EPC7ut6+~ybfA~O{qxM2Kq^5t6rv@?m)pYPKgWu`(@)o%o z!qRT=rGVS#ix@g*nSGX+Hzgs?_(4vn_|al~4p#WLGX3wL0UOG7vbtamh7qZ zvMue%Km%XMhvPeA@eoFfrOtn)r^2toF15??fGw8C2xu4ilqh_n3egQHQ?yL0`ten| zV_LOYPP#5!_z&m)vuXYN%qui1P+Q{P=kU$ZVuSn>@S|%3b!&Lqx!%5rR{BEF!ZNG0 z;k`wU`^jdi+V9bGO>3(luR^F?nrh!@i$V7O=U*u8HCE3Tz0s++@BwMrdWv%dawmMu z)|Wx~h%Qp?Jy`f@Y_QEc#TY(fT*^rrv5J9O06F}|ihmz;-qNr0i01RvU6fm1x&8Sv zhye(5i2T0G@QYzcIH$F_weQ?7Cvzk*@$*<%67BL%iQCD?9J=9Prk{5Uni?wO=seV( z#4GvT0iAb(r@7rrXP>N`~Rs3u|=c^ zgG^1N_Gska3O8;Ev`%UJKC?PW9zCUQULR9*n)Bp47zhB9f z2yJ0{P5w<6_n)PqiYEl95>Mf}!~Ng6`9HJ}siX!Nr;=B7gW=D0Vm`N;;>IvY@b@$R z>p{@e?GjCOk$7PK`F^F4Ti)KX%i#FG3HN`_zn28Qksl=W_pO5%ZlC;6vl8de4gc$H z7H?^`w-}`UH)qWPC^(ob@(tP_M)>>BzmMms0PdKl^0a#L=RyOrTaK9_=(_&+TdMxs zff)Gq4QpsV{dJuPCmldUP9fI~zCT3#-^}(wr0^D|7xMH?{~`37=P!QneS5nCZymptW&Qm=|ND427T|~Bqt>r8fBuF~e)~4$5l_zV|3B9mKN$l$-6!pzJ{g%Ej-`F!gS6n4t%=yxOxrBa$_i*)YKvnb|{UR}w?t?s) z+^=ZZOPI6^Z2yq4m5gRj>aq~#_1ruA<|5VGkUTh_?9m`UW7EXb9?dSGqL=U>7 zX|rM(Rs6HDW6}T7o>-EocmW?)yno*R9I!l|rJ>T@6cl~$6)=`c?4hr&&knTCaed@C z8Jzd(%6+u-mDlU+{u$4c9a2G`>+_Fcl)^%8Zox!AUBUn08!9BdNaJ$=4cNZn8xuE-T7MJA&={}#c#mHod9`*?Y+7-bJAcy;bU?W-0+9eN|siJz)811<4qmCuAK?B;VI>*-4T zm@gaDsQ(dr0!b>7$CwRkY~JZ`pG+E?te0Pn^&>%kkf#Ut4^~KsP8h}!uaA?`M}RU6 z(Z06$If;UJHf(!AQ)2F$`T(sOxLJ?QXsHk`CEujf_H+}8SBtkFL5XewT^JdEPmm#C zy;7~9Y1M2?0il_D_TN`!NtB*bW7t2U5Ox!ljc1{OnR9IJFL$|o|COsKDOm5Evo%%_ zjY+3Ax&P`P6ez|GsGDcWJk@{gab;gb!mFc zH*aK5inbMOk>jk78GfMnAmT)8r~USs8zH`~$BoHF z>QD?rRjJ%Zd=9Ke?ZVnp!iBdCa0|Q&lWw`2nUjKoXVg80vgn9M7n7={=cC=Kb;{^8 zxT1u2(bYw$X0cSism3X!^O7x{a2M*mM#ou!yhZFbd5o6YtMJ(`#ddfKM!|}pim760 zX-i$`zuA2u3=qYYu&-U8{%BvS6M~>!(P&_N?64muUN&vC7aO?YVbtC*Zdj!McG;Je ztJ+n6o>Y!V1i{9U6`KUYuiT04Us}Rs-h@? z;54XI@tn5;{{vV;SBXUM=tJS_NC}qDS zl5Hl^cJDu0P>lVSsYrRa|7NO05vNDdrOe5e6CdR~5pGq96pi)7J74b?Mf#H|JfY|E zdYqd%i*fN;A(G0wbY>Zq(px8_Sg(b;SIg9@e&kLv`_1--eb&aV?hIpjQRZW^hz#QE z$ZkqWwaJ6th*+T81hKQnS$}OXr`MWQ)+Hw_ST)CimL@z7{F;h8r6c?&Fhu!*#du~_ z-xFc)`(yd<`jyJSX+P9GnbfeivT_(gdm6=c2M;;p+#2?@p|ur1JNFF(M1+%dPaeWHIql0!7zNex zV@I1TO?y&!Ij*jdFi!U!R}~>i69r^W2yrHqXS%rJA28|rSPIz59=pCNll{FFQsJv% zi%zZ!HQ4U=*-KxB5=GoRF>h4nGk2EROdOAO0WUnV_Nkp; znO=(vYQn!MwDw+5^NmJ>M$>t5m<||ObZLSLMWd&H&0->}C91zmH59Z+3*z-$Var!* zNV8{nuzTaD47G!~e^w%3UX=M4jX%7SS_-=NOQh8c6`a0TH1d}e7P+0!&uHJP&|1*M&z?D^wGsP|6mZ# z2={gAxKY0n30XlqcQT&!wHu9FXJ~G8DcV0fQIOazWPN63b?=WFH{kc(U;^!X90l3H z>RC-+%}d)QJ`kAKC-idd+|SMomQg-%HL0NsUZTx8wI}-W;Nmc0aAPNwBqS@u`cG6##F=XwS2Yw3^~SJ)jhKJ6M5o zF()MQT`#n~uD3sgGWIKhEaUsy-PML{2P#YlH~#7Q!;{_V@cD0s6lR1`ADjfd1uDwgA06Gp5*zvOw~MgYYZA=}B?V1-nn*9p(j=dw7A`lNuomW;uC^_TF3d_P(&F>YUgB1&~3=}(f2l&&}o zWNE(QI0Wq*8h^H+$_AK*$Z<6koX)tVTMNs676|E zGHGSa_aX_5D$a&%Tds{P#)dOO@>+eOR%Kgq@Iso{41u{OnS94j_?XsSgqflrE?E4F zIX)8kOf?47BX?eVD%N}gP=fM}cT86tW#&8_9mVZ*NB!l}d0mxQ8H0#18(m~@NF zqoC>o>1UEfgbnx3OlHxN1iPwc(tqG?hk9tFzX@La9Hx(s)-DBf(eYEQ@uta&i`V1I z^7e8k-H$BhG(RDRPuBH*`_VoDZ?T5**zfqTy%o4{Q93+)3-tZXvqhvycvY1j*T6^X zEb#&dQS$GHRzl@dS-xvA;mZC7`$z!qEeSEl{?;6~7?#F&1As7ilL>5k1eEnb%2+kD zrs)(Bs?^fHV&WU~O&*~80a!P~i66-yI)Xs@yQZCe<-#`tqQLCwoOatLaR7olyK-J# zUT(yfHn&sjXZk}Eqb~I$OqCs^SPHYU<&!0f_39krPJ;2mu{-bYc^Xi)c6CWY3xx3~ zg|dCFd8OCc-er6yWv`uW&hO}5A50%P!y(z+sdGF!9X90c6-fA^xdykcKFFLHo^a!y z3kW6I^qr(nFPsv)N{MAqdUjS;2JFvancCiGN_{%LHl1ui)(7|BJ^&`iQ{&;`9Uy&V z0v}L93p7fUykAcC+$|R1Y3Eqv?Ia#4B|p2Hkofho!!HXE)7*lI5*+Fg4w-ar)>R-+ zD-ClX3FrugA?WrLB(=2M(G`qb-KjftQn)m=yelB3QYzwIW4gY7=CjsETJN$9tqh4# z!b+NVFUU|K8W|WM5`sJ+zmB00@&sY_Ihg)+~y0I;y+?}&>`_JZ|FO$uxDZ>68QlH{xGiR2dK>p1! z_hftpE24og`Yq=3RA-Y_i0noaB(!ITru|qQxBoc4f{#fTTUu0Gv5H8ucay_3^wm`G zkF|NXTr66}_>}NVXR%9#xwuV4vqtQrfn@pM&Z->+th`9TO`lNd1{0ie`kE5FNv}_^ zG7Z1Ol*C5hbz*P9cZ?u9H%92zJKmr-B?!M%yW)w{1z%D$WeEBRXN=sOkMw)^8kC+9 zPqBB>YeUggS$vd-k*+Szs_pxZ5u4|!>r56bvdyU}0s?T%_1e~(2Xpi-^lS{0bhbTl zQ^^lf_Od$YuQq6>S?jQsVJe;O)DqsO1B)kI%;oW_*Bds@!7I-Q- zGSWDmT-x8>qh!5RC5hhiTHB1Fe}GiNmj3cO*@ay0?&gmNd~X*oNJEVcGSx&d8$z=g zq!uPvwe!@HA4m)WLyN^(wK-$#y-SQVxGp`rcFhF@1;ZS|)lV-6^}Yo$7Km5Sac1~X zsw7bWU0oTk%J`*rdN)}x^pV`|k1zVsl3rSdr>!`6?e1)$YG=HmWnQAF<^^UddxgO(T@*#_ zl>uu6GG3O96)VdB7~G?yt_5WZ ziO2MQ=h+QDeu3E+ewcZT1lxJ-v)a)s|7mazCfi=o*Kkf6)owJv%TC0rlho}Nu^bO% zfC`VSF^RDekI-`JoSg`uIEeX+xnM#JJg(fCi?k|YuxQ-(jnj2}oSAf|P$*7vl~!#s z_O>ItlfK{u3z!!kJ$nF@wKQM@VW*}sY(c^i<(V#@UiTO-4J=#qd47hfgtd?Z0pP=Jg!5qAxH0`ak6XCe8=&K}U7%#P_gwYL z@g*KhfU#nUftf&;)9t;`VvWV(^1BqneI(qeMJeCq9en$#u$WrCZEIXM$2bYqDqRYv zz+LyFpQ!eCS2M4sTIkjX5=9DLSE|(++dbJ}7}6><;<;_qVicOi5yAHRbKx9zra9$n zC)=YrOwK>p)S50_8cbqgi@JjGrKbJfzDHIf1{)C1;PRZ9pgS07m3MSAIgeRd@rtJ_ zvrb$u{McNOIJVM4>?x3Fwm3T&_u;uYf{NCw(Sw+TE1TcvbZ1>ZHa+Esx{(D+O618! zr+$0@qI;1!5*c9m?1USGkR$HTMK13REXhDX4ROLE)rQdLa0qxb39N`U=4EjPD|*@# zrLzq#{5=G933~xHUqnbX$dFQ1ga6VTOGp2j<65+K)iHTKR}ME{nWZSa^mOQ!$$lG?)Lat` zni~!Nv@gi+x704X#k&f5Pu2mNC{5Ki7fWtCZq;h%f3ubLh(M_bqL^>Q6ZO!3iAJzv zc4idi_XXSjzA!DerdUu^mkf=2b(t8X+N_ItzRTN#^h}*}>F6+P^1wiWnBd*RdJf{} zEbm-eknFvErGfyOp)OzIfQSmgam3Kei5pWzqkvLd?`Mz}-k#?(uK|;(iBzNGN-9yv zWYW!I2aR{pZMqiKO{?Iupe%d?1dVoCejZXw29wE5M^v^*VT>2sVX)(nobD%kdc_7Q zx27+yF~GKrMus|6e|vGr%Kd1~FQSVf4;Ap+k<1480s{r9MgMs`2f#s!y>IK^A_3I_ z3UcL9tGU^fwI^c`C@VrK1LGS%76%5Wz4YqaMv$2v7{l?TO}e&Pb@gWzqB>}qU*F!M z9ZO&Yz5t`xn-+9kZUq(jtWE7J@10wXZCNizU;)3QCQ&O?(Iw%W!_{$OjeL%jSzqiM zY(zZznR->yLEWhM(VSO=Fzh}>N@J+mQAH~l86zH4)OsrQpS>~o>|;}>#cIpp7yR76*X-sj#{adVoVrVg3Tb;ZixE>+ z0F6iZ$z?(zEi{C0%!#RzK0*hI@%A+Y`p@{YkQFVM+vbC=bPQszYuagRG_SiiKa`mRd^1DJ5n*t z*HckzV!9oYsw$=!;cKWLOt_Fx-*TZ|ek41BE7lw|q;)V~M75KPcgHTY^SE(htlIsl zO8#guYhcCuLzhM%MX`jeBn_|>u-e*FPX~&_$cU*0A{R zQ#>@JolY8?4dWP4A-v!~AF6J~n>FzgSlqJvX_NrJ40+Hd#<@nu3-$Rm%bQ>$`CPfq zll*}B5oI`mHxehzS&vSZCC7utn1;uU*T}m{iPW2B_htx9RJUHee$=0p0sD38iG62M8iWDXJoCD{=gOTHzjOZgSiCq!_R9(!`I*grn{F*S#qne!s9v2jts8GmWE3fXV%^G@t&<81Y zxpNC(Mz3UEB-f7&>4wR|Jd~<5Yp5DY8uU+?WdotzuIRA$(aZKzFN99?df#e4+?@$E zaiHpk>viPPGg&+cRH^oUzD*&~e=$>c>JX`FOHip&=Ni~s?hEXzA*DU(65D8PlF}^0 zT~arI`zo3Y<`v##34PuA@@dnxa*-qyZk6#w>g6JOw+tM#``lS1_=E2K+ubd$vT0Yq zM>-;K6%k{9UpA|!?<>U5>ANJr8lt%wBy9WPk$+nHz++!++>bNuu1&EIbEe#`0NmP9 z0kSI~FCNtS*qvMm2vN}&1}OQWl)A@mYAXY$F;VjP9mT2CFEo0Q29xp2YQYMI5^*oI z4sn~Z%9o`ht7GIm<)MQp$yyZ-yPKD04|YL1dEEx5X+?>wI2F})#PE~rt%MJ}Z!f;# zS8`HosE6R*hgjX8Z!P%E)R?_iIBIDznJCU7Xu9cnm2%2OT0eDr2tPapMXWFl?r%}2 zIK7HHdKT->t1Hey?I@Gob6sx`#A=GUS#D|7-fh<2l_D`4BtZ<{P+{~lH?QnoUmUBC zpNDk01@j))TB%E|d;Y37GUwA7>#F1??_$byUw<=E25xkQYQfGhRo{`7V=0*vyx3LJyOf$NR^u^eDi3}pbN1Zb%902rEseG`+_h=SGE zMRJFbTZLi?3QE=d%t|8t`9;5=n4dr)82xfaaT(2q%Lt!QF3To^bp{i3q*Wc%2;1Cb zFG}|?AZG@r=I)DWi++`@!hYmf<)Z~JZuOFOe6@idk4_+!sGhT&3KYA?K1DQ5g23G4 zO59CHv3dRVw@9NDwl{f1=xx+iX9YFgj!d-Lgi$Qhs%?#JeE5)BwI4Q zpl)!lZ@!G934|Cc8m;(R1d~iam<}#2?!=G!6msIgJgUoHN<(;66H;EZ5!U>9^4AxgD#z#Wjq$z#`Dew-|eJij*TpI)M zxRo0M#m|?EAiZ$ggnmFDmxaN?vL&%u$-PXU4BOp7R=4|fu4sdq)$+h0Hf}P`yiyP6 z$zIU_NhZ9#P{e450=4Y?ip#cEN8b+yuEAr`T<;J?Ac%ZG4c7|G-L=8;+xEw7#oRVO zk9&R6^&hApgR=VSK(m$qhF-FF&z+0Zdc+DFR+QbJ){KAu!-tA zT4ZO%0@dy*_Y|%9JiCIDV1sr_5QS=eBh%%O&$d7E9t*swBPrB##I&q_rJnu!%43)I zrcg!El^*9#fE+hZ%ZE;1bGNFbVRjkPdA8j9Dnpxh;?!Wf{P6wa=e~3>_M;rsjy8BGKS zCaKU@NR%P%qkUiWcp@#Fj%YVLKUSG=&3PK+H2QGW>`>l#+^BUy(DSu4j9#v!v(}_J z;jucBosHfV$yE-0Rzo#iV$29fr;aqt_LFdb@(*>cyEs*^U4*OWirj@MX`pG--~pH5 zXSK0{RZ-CuDY|G*Y=^#lh6%In7VTkv(Q>nbS4;Lw=3$r8_t^z*X7l+l=cp_EB03jI zPG?n?COMD@462qP&mhAkQ%!2Vo!ui}Y!avg+w_&04$Wu^P|Jl6U`qcqS*1jiD8IPC z9dcGB;5pyqVnI4cHS2LJ6p&ewFmeVbP4H=W#8lF^=owZ|iZs@!&}9C?m4&c)fm!`K z&2=5X^hSAro=VH+faerkODg2^CoJ&{auDu_FY<;^ zQ)4#w`%Y9F2{f!ov02i-g(Ez;_P)Wv#{hoTl!s)43plR6Z8v{-z;9?XRZLy<{LYXJ zTk!QY`-dniW;hE>IQ-k))OYO`-^vO#-dXx^#kqbsq|Gp^O4mTLY+P3!$kk5Kq72ds zz}csK%R5Cm*uU!WNUCX=TRGoQjaMvW3juABNJFkNPic9pU1cf=WWp3DSW29231 z1EY_yB3|NiMa59xOPPD$4k^J$kW1!U$Q3&rPET68wXa{?m9?Yv1B*c^h%Ud3SWrZ? z00O^;;En#h=jjxpjD-$rJ8oe*)cjrjIoCARO)QHUt~~mai?8hvgK?x#H+pFhYM#Vv z+%LGmW@;)m5>r#O4T_D|wMDJIvK-5Bmk5)}gFLdx`%5Vt_^LUM)B$3emN1sr-W+`b z;YUnpp_{Cgrs#^;v$W8X^O?z#l(GyFx#S^}yzCIN#TJvXeA1#Bm7zr7H+0Eu@mB^EQGiqXhKtdUmD~A?b4pwp@D>Dam>EhQHSq(ULSsX@+dSqS{W&>ujR=z znO?mW46f2Y{n};Wn+E|K4oE56ZV0S8+RYj$OzzOjKg~#%dW#H|cYG9S9W?5)kx6Z& zt^RJM!vz`APaTK1-ig)ofD0}a`*B<168}8I|AEf8Kk?Wq}9}I!zLH_YeH! z2+emLC(DAooUu5=U{T$BnlW~Ev^S>~hp{2JTb;pn)mImXS^dIE${NQ^nD3#X+rs%) z$p<83;ICC(zBf+W(+YjXzYfB6pPrL?ncqVCP_Iu_Wjy+Qtl@lDX;|Iw1+XZ4X6$zJ zT%;@D6lz3uletoNPR={A2Bs3$XDu+(Az=_+9g60^i~$zivLN5%0s0J<8Oy1Q^ru;V zZeURckO7Ob<3woppNkR~f{bLD3;^J%gLHh;0Br_+qm=VS#neQp!=Zcr;=3M7h|PPO zy;6FWga6UBOU1vQw_#T?Z)Jz^7RdzyLEld_Bs?#hu>=5H8d#m@6MUxdJzj`}sVhg(ohp=KF53FNm5+ zrQWoz>WgNhdhFom77OOqAwPW8Hks%m^7M1$f}3Yo4Q-eMl*>0=vTW2mcN1tz&lK27 zEn;mwtS=Y1DQ^};DB~R;#a&OzfRMxj?Ts27gU+2wJB*+EHZGmdbvxf&i*X1((K6^)Gpxb zRP&e#R%J531I6Pn2%xDCrpD(_nC%<`o}jb(2|EYatF!p34B?m22d~enS?=!;st$h& zkx!d$Am>k7%%vg+#YQ;oAg9<+uZ+Ut?;<#;n|ghrF|2#x>;iRrnY}TPP)}*TR+qnd z_!xD9O-a%Zn0&Gemv-V}tQHID4C_?=UgVld8=Q6=l5rSoNV==YZJjYJ87YUIG>_=_ z>^9OLn_}^jF{w=H3q)#Y0T7We_w>56iZ96Kb}ijDuQBuOT0g46a8KkLy4-~&bn96$ zjkfS-#B*`=H!tq(x4in@^)dDb(e@i&0B2ehWkggxYme^anj!bs{z_rY5t*hFqy3Y4 zuxD7uYcbiD;0L^crgfKwf$lmO2-?-z+8>)c~D`H|QRZ=Jst2f|(H-&89W=r?C_(1t(3SW@#1Lyi1{v(PXiq(CN-zd839Sj?3QC$gr1>I<;LZhQ8kf zKCSc^f6?+!iJcnIEy5kwIp>4*u|t7_g}mW3`IJ~O{o{~3%?$P&z(JA30H!Z$KX#GN zKOmEzpwlzLKpYL#@f)1s`3l9Euf0+g3lI{1!h!puX#{NyBr8nR3H9vBH-6p^aiT@c zHC%eSKdfWVP#KgTViFRTXt(IpypB#q0FSluY}VltRUNcY5%+#RH_83rb!Tjookc|) znDI3~yYErx?1AbMGikkVzQYdzJJ+_ZPuQkFi8{+6sF`%NJ}eWw`k6`Z=vhWm^UCI? zhy|vTnUT(Pgx}dRY-hl{L@5zbW}5b*)>yd^qpBo!gvxx*);{HTm>hPrh$DCN2Ro|$ zg(m+{g2&nIXHkkhs9FMOEa3x$Dan4-uQ^7C+AcZ#pEh zGDuMTq0!~exS*#gwg_P`Unee-PgYn}i`NDMZ6_v2-JqRbC#7FmTIy*D9o2%85me1T{8!ckr1!I{2H8=TS*QZVHQ~IK2C*sja~8fbJk<>Dt(d&;l<(Ee`?6C1PcM9nQ z-QmA8Jb-M98$Itu@0P6^;;~Ek`Vu%3!k!R^dW|x;(CLf-Rww;*Sa9)o{F{#`0W~00 zAfmsmgJhAC((ke$k|rB#Vo(?0uGQb(FJ+ZEnZ2OrU7q>drz_!BN$O39E|^V z@l5m1zpt60Q%;8fe7ojhl77%3!_`L<7wP%;Tk(G|aayRlCg7}4(Sn*qH(t0nz2XJ1Zqv1Jv z(l0he-?yvZ$|SLEgjNwce12#;ekp*)h}A(7$OmY~^_Zkh{iegT(hH4IYIFS&Wa5;X zR}>CG`5NyLz)69Vj<@yWT%nV>O3~y}SD%&jUvb;x4*+!GLRmed@j_??c)Z0ON{K49 zyKfs^&*k6_2*Rv$da<&1Ch};gn;h9FLzz$l7@%!yvgY#41$w5-1b@1Coj3WN78Ogf zOCnD|7$*+gnP$f7(=)1@%#mnp-92=h=d{@rpU#g6DHKk*INKEh8@O*o>sfroc5U1q zYD{Y+sV!@_juC<#F!xD9pLY3fESi7r@+xreMTmPt5?Qx&ODY#UlU~BuH>!mdbF=rKe9cmyq!Y7)0`p+JPycs84? zWX7*0lpte=SsPhR*f-1uBt+OWpVL?G+MI%3eX8@}d(hZHWm1WH9G?*0cC9bqli zw?AR~ojzRbhPM|cMjdlH9Qcih zSLqAzM4_fMw=>LqtU?Mvn(*cH#IKZ6<|G;M5T!vhSNH1mq(JGr9S}6xE$zjYMV^G(FdD zv`@?9Rkz}Ix0t#+6_$UExqOM>{$@n*G+TS56~gTtl(VZ`q8Ukxj~#FcJr6^q{OQS0 zusz;xq^b8i);0ws6?J@rm-{Eknd)-}=JfR4E|&2;%O}tH%gxvnmZ|I2 z2^vC698J-Y3}s%m3j5b@A`AGv6$>WiN_g(|*&onR1<*2%Q3|&zt_z{Cof#9|AN+ta z<+`BlV$d0K6#&}J0MhGm*UzGXD&X=!-*LglsnNUyEt+TN5~}mvb+T-6(SX=(FGYcn zamK#7N7~i#YGQT0^*HX%yo0Ai&|6|~kb&eUK95tPxCMB-A}d0N$L9GSIRP}~2^{yu zb695c5rGLS<-*pOaCgiK8h-?3^1k`ezoRAP$-u*LSQWJqiNmVqz8K7O72WmgB7sLH*!&?cWBx_Rg1NvW z+hRh+gLcq`?)WzGjWhv8g0B39#Utz1Uf*~ZJ-y&b#ZHwzXAst4hBVJZ&!^F8F<`AA zJ)8!7{?YQbVWir?IsZhKQoe>|uTiqX1I~K4i`Hgorpp<$Bg}h^6Py&VK@&?!n(U zDr7DuwMFuri|U2hwcss?dp-FDKW&#|cCi3f`b@o3qR(s02BSc&onk=`c-oMAt5oVyAQCXN@uea=(%sq9BfjS8>zVx< zQ4Bc7Odo2MlAYU{q{GA^q())2t}{LsKR#z=w`jj%Rsb+iT2%-)OEiYxu$mi^2PP^C z_K0+Xk50L5CN$#PNqkdS+@)2^*Rr&c1}t zRfw8D<#MEGaBt5=&HHFe@zQgo@AYqHEwHnF7Z$9!3kA@~M3=gf2JlT6%Mi?Nwz2D) z0?;Df#Y(QcIw~uDKA#_ACnI_|8!qVb`b$gpK%CHOmvqFyjyKG=@$Jn^mFzd*gsJ80 zOoTLB1Z*A52#I8#$S!F(+&Nr!SGa9P@cF$Vt2#B?N*IrzbH>;D{v@)X4f*4*JYxmD zcGVhk07cE}O^_)uyhEFF$dsKNsJ2{XCD+x`h*~C|ZF+LhEJRECyx)Kk0et0XF>{8l zKM)iB8OeU{z-a%5FxTaBp}cPNO(=)0PiiQ#HXUb;1v54Awwq<;ffRQ$_E@pJw$ftN zDr;XHy^p3mkiOA*udFh0BK65s9kG<7VQ?Q872frsNpa0Bp7|3Wq47jzV219A!)>J8 zy552#cJa5fKEPXCcS#i1jBh1p&Le3pjXTCgb*(2#wA%8zRs>yHgSq}WaiA9@6)^1q zNHeXb%g-A`UR-X2SW|?JA^MfZwiwtcf$_rz`^1rUbfCaDBML!_aYC;`-vWa6=` z@b9kTuK)~cPj`3s7+|*1j%=Re#!+t^+ng@UbH0^rn~bRjxC6upW2sH@O)9Q?E~%U5 zvczNkcKdP35rmXv>qvvh?r$#AV`viMXEsHjTt=6%Ky-7ok$rT(kN9`WLMywMD0W0C z_#lzhqL-i(m3+8dzjv+Cfw;$LEIm@znXg1Did9JhNPlZT_sZR*2Kp)O3?Ptyn^KjZ(f4Kx#mu$cRWBW7MI>QcsK=E8As1~ z03Vg>8|inF3bhW0UTjk1hy_G@4>^8a{`W#2jOOWkDWvujU)Sox-aG6*IU1NO4yXll z)9f7w8)2AscFPzz%yvNea(U?*$N=hhuI^N!S^%>ON9}qZ&G~RX9dPD3qO{m2(`^B| zB!k;!@qQGKCP!Vddf@V@dH)7s2$w3~-8SoocQlXYSzm%zaJX?>i>-( z#tzG@Oc4!f)b))VK)x$6u-!fXw0`U#9$?+3nR=WN6TODWeth5c>eQ!cHkRbHKS-C$ z68i~hHK+oGd=`aOVa@K_QRLllezlO$B7-{Zj;N!mw{n$wDo-B3YdT$!ZrR-#V{0qU zE5_A5dDKNXF_TQ^DpjoB!p}0gQ(*AfC4obmDIMw-xzA;7C%QkY^Ia89)mTY8cf*6> z9a65VNI#(U99?wNUdP%K95dEOy!Eowzr4$`HAJVQm4z&+J-gZ#_b)bcRn|-!xIMtL z(=4(rj`ZqlN|D1dV%vh|KEtl4T$B{n+==)5qOnx*ty{nDFRu`POELp7v`0C4D7yE^ zTUW|BY%QY`hTO#p+>DsJbHgYhxVW{iQQvJE4+f@)K3&%WwBa2e7`Fd|AXfJ0gevadT0I z^mppla5x{p_*8v(<+I^_oBZ9lw~?aDSw+c~82qU9nnL@VhK>=^lhyHC0|OW@Y%wFQ zxSikESj=Yw0Zh4*wcfoRF1KS}7G4c;X>A^c^(N=@h^rKW6*&$1w=}{5l_+0VWY^aa z-|BF=^ut~r@`e%a%H@D10oBdED1%vOSXETlPQ9Iwx<8siXxfzLY%)dU({1SDxn}fk zeV8x$SZk(M7z_?PIyr-e(w0*r0l=3uhfScj`Q4uTXs-?wh(2=YvB%<8xPGR}=q4!u zr9aMBGPDb%Vd(cJ1tJmLjGlSL_~GLDjYliwX*DP=eQlqwzhoEyZQR^hv3NN^&NMji z-aD<)$$7zk&c1%GI=NK&q^|Kp)^xiS$~04xiH3Pz>`jY%yhg;`TH``Or2s8oOmhfz zzF7CQi=(z(+mcm#yhrY$mK;>S+vke>DMWa+d1K1Fw)!|{T-tzGEtTd0n)E}UesT5w z$CN^`$F?@lCV$5FyFhoDoS zf&x^zcwgs$zph6+=h5*)Vb`WDfd8CkLh$6FctGvMj5N7QI+&}q| zHZ}?yFjHkDv{(F@CNhc5=?hTkp7JZNg*-MKK$9+2U-^N@>2xD|SrE1#BA;!g#=Zmv z>0*~xRLdl(#;E=Bb7~N*zvBiE?6)#r-7TWNXwKM1Y^^5ykg4&|@&k+rocq|6Mzofi za@CrLxjI5VNs;HOH3^G!tXbe?Ffze&xr}B07-55pKB4##NBDk8y1c) zd=-6u`U1hZQlZT9bt=4;qT-v57PlJW;CXwih1j zj%2nVwRmz`mAwQQP6Wu{OM*&1b+=2NEp?iahA2mLi3^cMx*ER>uu{V(racZ=1#4Li z5C%;nJ{vl_I6FfzH`{}Ctu1B-a=73R`;SnyNALWC3LEr0)T>O*69x=0BaTu{`eV%S zzep*NW(XriLPb#)H?%I_SoBqe?0h}YeV?To%f(i|HhcX!2rjw=;VH*TocFHu`#P+8 zHKYuY2an>sN_pTVmUqVZT%+t#3+4eV0!5TMo)i)wYQsYoi*Y42{8Dr2&P3Mx5jJTO zG0`@2M!>6D#9~xq3u=?~^XB||{;(^6R@`!4V@9syWjtD6Szr7DnHb(euuiAUT>9(j zv&p$>#{EHF9{VRG1T?zy>#9L>8$iJ)$*@opO~1jw5wAXwt#URSeU{a%S?zubIB;Lr zw@j10*DIQ}XWBXWAUacNB$1@r_?@fB%kUoH22~_~ibyt_j{Q{Y(BaBFSMi+46L)A& z2+I&I+T@7U(NVkqtj$45JL;&dOs#ENFI}deRU+c?aSkd;(E!_sF-JN}BHp`Zqtk58 z@#f1n7!DJfl@V1}>*E_nXOhD3wf-H+-U#RSb6q60>sRT!+h_p9=6f7z)-|wI)kgkp zldhDvnAVwW6taI74M*??9M-dX2qNRg6Z8ieNd}0ngi>!0*q_p1%G&_ywUGO}{*tlk zdlQqHuIKoLM-}5Wr-t&ntk7#n`K0jBZGO)lCy;Phyklt9mz7+oN#Yq@IDjPOaU&-z z7x&>M@z_K1e1@Q^O#~{FyMz6=g_7 zWb1z(CqR%9NVnv2yOH>j-$HH#0!O_6!l{v_BMwqO^pLKBDoOZ~b%H{^5jlx6f_nOc zQl1zU9u6+b9K^;g`1z&ob0=35RDY}aoby7Q=C%ky_6F-5sg$%Y(Y%Z!(hqJOamTqs zX9knFi?nejQHg(J9qy{8;l<%;DH4c~#G8<9c6ACVzv(+}c}&Y6H^u=QN)c2f(i`f1w$#7= zqFDLidyT!pgp*9wcWk>nw^^1pvG;NfN&t!tfjHm&hC=cp+4x8a(=jYv`LY;r5SqkL zO4AU+i*xNgHw=%PFBuwKcfE*Ss5Yaz(Wvb_K6;1jTyOKu5)VL}3U!CV4veG_Y3Ph$ zOi(E-#fJf4`quh#&+ksAnxDTeN4}BH1y{tBvmtQUKQ?O?pArHgM%kX;7wQbJnDlCY zXjbXw*=EtJowXYy`qTKG?M-y9IfE@CV8^?q<0N?R%cq&9hBmEQc$$2L0&5VJN$2bB z(?bQ$GGMWa5wT;)ffPD3-J$N5+PLIigI2e4{n3Ijb9f;_Uq(zc)cYGw_p>lCbcg;m zPm}5Oi-b3Wy!!ltRo}atG-oFi3JmQ058l_}U%K7-f&q{So+5MI+Se}o zMH^z3sNOcQ`8&iHU#c;i1ej!m)bb_{*(qAIcS)W&^V)8eibaS6;65OpCOKbkYdw4O zLfQ4|4s+gP&;%}xEKFldC(;>6u~l=I&HF|xvEgp_R}KO83T?TH=}1(bz0&ekd|}NM z;#>7>rtP8PMu>ScJY<8-KJ}i;`NJ4m2JV*J+i1!Ttng;6bR82mlEOU6ggukX#KHBe z%iMUBJS05&zOx^Cc8{!5ch$Qw-;?SvD$EmCO`R;h*91Me>O;%d4NPg0Y}e96zV&23kS;TXpmu50eN(^ZC}$AMNiC!iN2#Ib{fhGaV^()UGjn|d=KZE zs0OxA`P}O(jRd*aY*)J^>uGrEbHK@JYH0zY6V9`)4*>dH#5Jyk^!A6+Yk~5`uYObh zLS*<1@YrvN^9n9W$%}!yScwPlHTtd=($Ne(>j4e7y-@4`^e$F6|E?`T=Ein=t_as? z{p#EIgZQi47b&dAs|{~_*K7_(*8AJUl=1tB2637fYtgR~hlO;G{aPRZh1-7Rt;5U` z14LWV0Q2&~bK)~-2DloQZ*5J8kZ>ySM|OP&g;x))kUzr~ZVUJQ$OhfmZEk|EE?ktk z2(vL)rl>!sa!~Krt6u`%P22Mw=6d`4`a3s$1o?PIeQOhxg?Y17d^q1X5X}Zi_-vw^ zofi9=R`VU0=4$ep60~y$2HCoYEt@Cv9jd%nuQvhDs4o6Sx$EGjLu8?9#m`y4MG~w? zU#3z;E_Wrv*h*6;kV$o+Ij@MXt@>N!zpa?EK)xu92-C$n|CnM8@dxQ|3865FXT@rD z_~XS|ol%3RfN!_N5SGLF4%%$0vQvT9tbS}P%M9VzJFGWqFkF2FYP!S7W^2)7fKizJ zu&Cxl-akgtSKH1al;pQB8o2O{;Ijqt%FML^$_cKV8}TwLAm*6-1uQ@XP;mqT^oh>I zJX20Vn?t#5qObbezZ!@0wZ?~m&{$rRvMfu8qLJDoaRQXZDAZoFLr+{VAA-_6p&;S8 z1LeArh3Rb;H!+z!$i+J5D;M$E89%?iQiQQ14jgzBB-5iM zr<$$50t);z*W(q%CM!UC?_KK3;CRIbG-rkrVF4cfXH;3cH*?-bAbF zjrD?)EI)h<6B+UB^&^#3=mIu3f=x$-Q39);7+7Z0{KG(|cQSa;>6{Qo(H+_Du-jZSU-P*lyQ4W*yorqmiK8T?z?sJBVbU>b5^C3_#yPU5@f!5=UUin zHV_QX@YO!-U}({td>Db{7Qr1W;8&!K3g5o-hoKe;C5S!7XCcg&sm|2sCj3VOyu#1} zCZ?hssVr5KfW7A%;QDy>3BdH^$ChXN#WN=y3f(8rwxze9Jj zOA@gim{=A7Kv(pKH}5aFN^9a9lq?{7oAZK!+MlUVIg(QOtr^MLrgpc{+?uNJjCcI1 z-xBCkpARjF0^t1JnW~SM4OKQ@tP|ICMx-ABs?Uh{y;zGrb{7=EBSE)!8!eR8qy$J* zPfpM*+Ts2=-Ij%kvf;t6UIy_25dgCJGfS?}SK`MlZSil{m+?*k;*N>$c6 z5grKN_QO7Z9`qnZ({u*0U1bER;hr% z&EzNnK1gwp)D_YHP$i|KLI5e8U753=#`aMhn5*{;1HPyJVzyx;M$z{LA>ZHSYQ{bY zv%;PoNI2PSzFTQL0r(b-7tK7wSxTj``y#0tf&rVRU=~j?P3ZR9nS|@srcBL+p+b?E zQ(d#ssYRt}=cTpgVk!f5ITh-SPz>0$46j$F<}Rq9+s zhf@4ZUsNc&(upN7QILW0^=PE4{K@|3#N|69`k5!rI#YU_vo$=eGSif1!PKoey@SR9qpA&&s*n z14!us2I~A*{okO3e^!~W>Nk$PA^fQV8Sf(9FG;xYF8{2A8?1cxQL+;GQKY| zMS=aZ(j4GB^oRp>QT}EOqtlpx{);h8#}3r9CyK9A+|l0d!_~yqt6tlBd~v}-y_a|M zw?gxt?uoYJeic)VIB02Epv)SC_h>oymRPnD;H+s=IWoBOcL^QkM4P}(x16Pk#&WIVLOMG4GN z$Nui_Bk7`ie5t`Lw}a|qS@+7q(^m2ZqbVYxy^dTutTuZ+5K**d#yB1)9Vr}_1QLK^ z(dRg0n`hLP%P*`6gvSDl5?C|QI)i#$zd4P#EKS<(UcAn@GhL*QdY8sJ-)QHbJ*f5} zlhs1wt82MF{R*D@O|(+s0@3Qzk&M4MNrIDs^1rDQNq|E{Jr2mR`qQz0_+Ed1xbVu$ z5>Q0Gy~L#t9ZY_loXUHNGx9>n1dW4iu5;`Aj`zmewhgeVKp6o!Ij{f;s`gdLWvZY^ z=x9aGP;g_lHgon0VY6XJkpSADXgmXZmRP*^U~+>f9TRMIy{8{6801h=mqrp8A zDB`LUUAJ04=CDmKFQnW~GusTL_3N z@Gk~$S)7D4m|VIi&^we-u2{wc>ObyWO@R(^rw;KSjz8dqwNF{P;#@ud%+meOvlocR zN)ZY8f*O?#hqP0TY_Nt0@M`w$%SFQl#_A+(I8g&B9gl`GD)p|_BB^i(#epgBrq#p{ zO{FP-$s7Yf>i7-^Z4Vyz=3$aB#~&Y$k;y1t&Y}S zPcuT1>`kJLCle*Z$CselVl&9lJ~y>~aM09ndC3Y9VB7sMF@Ef-wP(4l!k<8+o#z>*;Yj6ZxnIIkv zGz#A{h^IeZ0q`iGz&PwCpuFGzeDI%t0?Z*&V8CNRxx5&;iMrL&k|@?T+s2jAS2!HpYVNn zfdYd6ao8pQyE5dz@9uaF;SI9dR6ouB^$lMz0?_Sa=~k#eHO>Bgg;M;CKf==K)9>QK zC`Z}P_|Ytk8L)r9{9o;R2DyMuB)Z}zZ}8_PI6+S$@T+$U|Fv~QOK%{d!FOwdN3VW2 zPXmQW6b-wX;V|Go-thNtPCht@^88QdPoQRlOPTvPuqUq z^y6v#*qB0)IWm^Qn1Uvgydbf8ePM4Hx0)7Y_b2cK^OQiU&lMGR#2#=cWqa z6GPO#|MC{#^;faLFCK3QxB!Flucn|Bz`#Xx>}SWTOvJq@$12Ssc?WC zqUbtp{<|C$pX6}ff4cDdrkB7?fe^Olf2xB0`;C0iK(F6F3o8C~4}fw)2Zo86SNQ)J zTNB_-%~%HRzgHZDxF=Dd(07<$ef;-poez3Z4pfE)8K;VoT0O{)j=W6yv6^v}Om8o*8Xk1$K;zm~t(76#CzFgr<^}#c*o~J1kjQjDAT>gC(zyow(*l-~E z*B4%Fbb#arA74EC&;I}2zNO8l_TNDrJN{}gpD7wJMv2YjhyQh_M8SJnI9(-uoWB>2 zJ0&1aY2KG#f6uD_SnjP*K=bY>|9t*F5!eDhWo!uldiAf7_{UGg0D)&tz4HD26$%dU zlyKRPFMo}|e|By74bbbG!aDR{KS9}iYVWl+#Is+L`8^iD=F1&0R|8W#hzO8>eH9=H zNJe*ba!vEU-lUrWbYSXA(do^vVQQ6z5%2+#?BJuJQC}X+$7&C0{gh7RHZkCJGbZ>y zCVed_(BuPGlLd_5ueXAr5+xB115Lq2%&~fdTF`+L1-bqII8l!HJc;02es5GjQ7ZCO zZ_thmpv;R}FTzD6-K?gUarAmb?E7tkWZR&JFR16>e!cGhb3SVTLaR-X@%b(KB~l0n z{)h@}+Frn;x_AXgr> zUz^u|it4ED=@>${f%?VFmfprDQVqReX`?-I*b|p-t<^#$J z85bFA$;F2jOHc1s`Tcla=so_3$6%2c@JQiHOSjgK4{oo?>ufH+l`-_bu5)9Z;w6a7 z`7P!Dn50CZaPQT_Kny@=_Z)F-g0+rZ>Se(+-A;*C6FEQ$9NA6iF-P~!-~aa$?j&1@ zC3H=#GL*>Rk&xU`$l+P48-W2AE?pZ;cCo%}mbLJe6j0dlh5A1R+m(Fjl&F_iIDcX!`C zdVu$OUe5R3`EQ7b`)@Y zbdA0eBIq~q4V~g%jidO1_Ipv857n=wxYaTLGCW@`-$q~E;SkD+dwLl4DTLkR1^RuYSBcYzvE7q5{8a4s8e+eTOzuuI)CY^r6T*qIpCmOadT-&X z{1{tJ@6TVpKi_TmW$zXw`##DNZ@RG57k(W;F`Dy5w@=h&bLNiG^4E*M;=Y~4x2B=| zoCz2~NiXZ19jW=%y61B@`Q+CBO}0uQKXXazW_3FF4?TUzO+D|#9$Bwvq1`9Lhe$fE zJhq!is`e#OT`S%Fz)B+0B3tq5R$s~T+w!AX*IV5zO#Ve}1ytn$)Lp|}eEax{!AyqT z)vxeNoONHfl_rCV=TMCJ3m4Pr7Lgb-pMgCP4U5n`nydASB#rha$|PZljfwqg#TSMB zOSULcJ01DLT&;-|Hc2I70k1@3dBph!_32kz6s~KC9UBn>Z4^U8V|7dkekP%@gEzxa zTSlRF!>^=atd9N;{FaJ4y+krpdFgdAhb0s`?GRr(+Wq>6TH5a*pZQ-#IUdYvKTyri zaoAx=R}|jPh`hKiV5R%ZTvXsZx(J!N$R0w{ttdBzs(oTZ{)-1&dG}WQ9V9AEFg8El zx)KyQWktx^>3XW&RYlnaG44=M;%HtaD}TdaBW8n$O;+^7RH2F8n%}P5wfl25!-+akGfv?&Ypq`$JlpI z?dQ8cauvw9RKLaAq#{bK*MA^f)|F~{i5hGC$v+5t&&f8$4&*Z2mq-eY8WqsvS1Ox>FajmHxW& zT4e7&5hgl<=9%w@{SU)RLDj!DLZfSrE?zpMby0Py@gt>x=k5Hc5@pxtYm>z{iVqrm zN9tzi$2CS{DWFcCfVV~1*~Z+b5 zSDRcTP_x9_?7JJrXz3)c8L8?pD=@pm^CVd&`<<+|O44+OM6w5TYo=R1N4o_|;=sg@ zMk}Kk{LtssHJsrv!1*Y8WS7#wDwn}oVj1)fbuBdp!*$)3k^zP0A~zwKD#??j)=Xeo z+cU-KtS5usU~F#jYNR)@0mPCHhvQ7X>_95~}gJro5@yl%k0s-=EiF4KW@6OgovXTw-2k zgUM+16DnjfRWK5-m0k?}hQ1$9?UiHm9ou^KJMKakgXJ*FEcXf^zHsDneU6Dvos_LM zOGXo+zExzupUFuLZ2i{|1-90XKfA%sgZn}teo zf6_+hm}e%K<>KCEFkm14c(U@RLB>DA6kwuIlt=~NLFylteiE|MAGLmhCQ~`TpTuJO zwo;AWg{p&QM%r5ZS-B`jclX zqk9w7*Aq0tJct6nN&6}n|0F*`-%N! z$JTFN=>z%m`DU~29t!5)iT$$B9?DSRJeH(MO-i@Hrh5L+c54A=eK^IwGm$UBCeL+X z?Gl0FirGx>cl7Nf5b))5f?hL5IBU`UlI;cm-D`JJWA{7>{vgQ$Ws;VG<`;&3wh{mTk~&=v#*2!VBF6v82!jbP0Nxw#(v zwYI0pa@(BiIGaTdRg^iFBNyxsJ20srqt1(lgzVN5NlJO2F^H3XR=SV74*6uE!JZ!; zaI2Sm%ti)+hO<# zn)=B@p^jW^t`lBToV6_pe}=J?je_6IgPm9JrGWQ+_p$w+*Ub#;LS`Qm<6#HE^sU%@ zcByB(CYoPd*nCa>;z92w!$QC}tz53-{pAX&1RGjC#ptM^k{|oE0gGU-j>$r2r& z@_ce}52OnH8Y4M}-i)1}F+4qwV*NxObR;dBUtyPM^YcPj=? zw#sFuyxNkTMn=rTC;}T>P(U4cui*~eqx`(EITj(4@tPZTQS-E%O6jfHz^TE)+L=R5 z#;3#{sHeR(2H@tQ?H@5^2zFU&dwt9o@AHJ~?bW;dLvl5H4kJhMEQT(32eatTM}Vg8 z8HZ->g?nDJf_@2sH(yJ-atg)1;{iQC7Q`tH7S1=TxPmS9;FhzAyo>-n8Ejb2cCUUM zjCqHX?AxZuN_Tei0)L&|-{eW+xF=sPmzL(a^~i%Yt;sgY^cxw)#thlfbA4_v%m-ruh^|JmF|5qlarcB3hEmSad2J z^Skg274pcEjn+wHb-}tPhH}>#g-m8zUnMIS#4IW9MhY`*<@$GXG~;efhcc^s+@^!e zSZ1R=>_jKhO~9~xR)oI%L!A3~468K@O|()mNT5mEG4oS*dc0!K^F5X=QrNiQrsm7z z@+AYh8w9MZ`E`A%1Nnx1q+yKOCIf~melZMS*u{{RYT5UrsO17(DDaf#PgG>@(8;tSum%2AYtRRtbqyLp%z-p&88JYVI626EVGr%DGL_imd{HG z(%kqliXlGhOtuRmNcWn3Q0rk8K`*eWT5m`yFI zGf72Ob>QNirTjelX?i*0Y&1crQsI46bk~fBQb{iz!niGi`-Ud%#!3QWD{sw7q`D`` z3Ut20M3^o2%lCgVNS=>$45U_U>o*ZZ<)~zbDi^|HrEj&Ri-er zuNbM1tKV-=O2G1_of*kuR>`PSP0*dMD0Jd*Wo7c^a$4$`PqPcBntCE;CzANI@tPH< zkO{|qp$tOKxz+F@Mek}t!ZukLioRaBsw4Ys)<@Uf;Ui)&bzwKBG$=F6L1|x@Y$Wm^ zx>P#hL&h?2`|S3uDyjno=GmD!*{|yuX1hwFml$;??oNlrop1p>N><79D7V;Fuc~Qk z|E!NT)LCIDuG1f%U7nBjFql@J)Oy&iYRE`^GN6_A4agv2&HSp!kMnl z@62abP^f-um6fOuen25jDc-8gc>|F|YuCBOaFdM-Z!MFh6%df{l#^}?5j*dYO4Mk zU@Qxy{c z&G%MS{SMsu`Pq6%F13^0J?B+5vD4jah{9|BAk*f5HQ;dzUQmn1Nc(qUj}?t(Fx5$D z8p>(bBhUC*XjLtSMn1_SfAL9u@g{+}qG*OtEb-6u0~Z1PgtdAadCkNAuD!4bM{p=g z*O!S_udVQ(V;<=k>I#P5nCtP6Fr4Vn-1i_#Ap_C*DEYI5ZrjbxP%v7l;k|M6tez$J z!ndHJTx}u$N>p=J+plD?Jc#{o-_}g>7i@_4g(;SnGI<&2@H*Qv^z~-jXD`|>>iAWc z6SZxn6F^Kqi`tYXaxX0R#me8B?Rc`W-7vKCPC)p_?m8|+8PJm%Zet7~O^I#Tw$*L| zu4c!Fye-O%e>w4<96V88fPewv=%ugkgEuhn-tsXT*I+vP*kEk{t!YH<`|az#y18ig z57>lEc775A3ii!7m0Eq?`l1eM0fhYnu1Sj`fC|Kb=R9&R&&XT8-!aVfQl-heUf^v)iZ_L zVdHJQXqXhz{vvY*x0v1xDgi2m)M+UYF9g9NF!O9g8ny@OyY&am0}c;4ZXmp(AIl5+ z;K$WEa#;Cwrr#duP_GTb4aD>V1Te@pPFdu zDUL~Rxp`c+xrkjZbt9?!+)GuRcw}zMw%d$`Bc|WkTb54&AM~4yluP4$x1PiXa>T%c zz@d>72{GW@9Hf@WwP-?zg|MVU!@0Sl8N|RC8=XzbR6B@ppil^~A_sf(56G`e*F{JL zL!;;nC#^fEQ?KEWCCjH~9Bi!96+4VX5q0PGjz}|?D<{A6B489YbdGxhct4bisDr5G zx>Egp-DZbwy7(t(V1A)NZI^9Xe2VddI9<~fYd!P)m4cZ zPFJLMaBW%Q6}I?(a``Yj$b1Rjn1`jvBv(f6ZT6@H0TzgPuJYmjseo7H8ugucdn(ii z5)~>$){9+o0;u%Fjt8qo(H+B&4Jsjc%rpu_%QTGwYd=m&lj15G871ZfV10te@8 z0ck2_!qy63CY{c&|J&%5>nh@Yz_WDWF@09U48ho_CgM*k6Y?YJ^2&nO_g`3Un)@No zPkvzDn5lIdPQ#r#f+EwRPFLC0p}OO`C$hRTQwGXUU=s6isPZCGaJ#I3)K+N#Dz)yy zIM=NwmpEnktiZbtN0q3_D(jqj6gOn2?UKGWrc8=JK$9QQN8lZPT!M9KY^gLc3+)>&*F~@e^i|{iODg>Dqb})JRJ%Kn zRSg<^!@jO__>_-J-1r(31E80II@zl1*2J z&#$0?*}ZQTjG;knc10Aw#!L!boU;S*W!MwP80L@^&)0tZ={5cZPj8l?e^3}w8$IaL z#dob*0vQ6ek(}=aFQYZ;zu>LawcaeIPNOwzKG+Hujo2v4X83`=FJYe!zoIje?_IjW zdnbft)4P)~g}ih-b0o`sVaW396}kA}DzvS_#yGf~q*DkUi*Ok}*`w2wi0qmO?cORbz;Vu}h6<)4YQe^dCU53yF{{rLN= za=wen+>izpp^1-==~Fr`E$o1Hiyw;vmJptIJM2nWZMb-kX2_A5DB^tW9+o;kE_?pO z6^(YQn*@}esuIq5x2HYBn46K*cn`=JY+wjv83%6qN zQ)#2dg8M-fNGc1%?De8(2`U0{RqgH-lQ`|4RT0nXOnXLtMn8SCvU(KVD~E+u{4=UXj`SG8`!|a4BBS*FeQlobn8@ zb8gzeP5weO8mzHreTs$E*!~GYN2lU+#qp7%a9{>0tJk8&<&9->F?{Bg?#{+>iVKcc z5MJSrV1nMnG?s5_mR2Y>YI);}&YmhHS{4(` zo^(laBpk*GWi%`w&gXZz$fEy(=ISlT%F>NcT45AG5FdOm`g)IPV+5CgiH0Mm@VbCm zmV$S@Odq<*@-XM@NI`*2@__|^;wSF^DG(%m`xNlu>-kl0xQTuOJ?)~L{~JdIMEkAb zB{hhUTDb>$VAA+pTil-AYC>Fc)uBY^!3Q~2y_4x~S!PscKvF^PkoohuAa zN#RK7lx%cB$Y~uLMe!@~oovkDn-9-ak?>OEPq!4A9<8?%l?kb9 z%iiG__7DEhod5jqTSEXP*y#@GJG)BdO(R^gqK%dclhJk@_wAQT9uqJM>-KmAQ=K~E z5zjj#t+=cI;xS?Y+ys*H&gAQi__hSAPBkR6ZbU9)Zdc~*M6!%H7O{IE)cNej z|9s^Z0RS97is;c#FGcEsqUFaQ*;Ubi2`7{6NcpQS08n19Vq8DF(m(dh$qfK}+>`De zo;_ad!=Hp?MfTGf2R2*phMPZN!2+$rPVe=fzaWGJJuwkoJo^PQnwJ^8vCa2tD!7|f z5srq#*m;p+#fehSEEH<~#c(~n4VIe!rr+q9NZh<-q~bl%z4*jr&)3ri(TLXptGslT zjOE_HpfaC+)k8ug1nc`D*6{z|;6Tg%O{w z?9Z6D>$;G8_wKC?MB`3&N7^E~d)uds)%6H4{-p@#hQt)oYqvz6tc=rRs-JfuH?Lxp z(^VW;Jb%Y`bSpp4-LwLu^wEi||7rK$l7RhLt@5oTKV7(NStM+P1x|hyKn9|l50X{4 z6GU5Tx_hR|{*;lk&AM_I+{=aaWGji28zx;rmGP@Sz*SQksmO#w*fLeUNMS!6Rfo^r zYVeIKh)YMT6E{w-CCbG@#<{|68&=(p>y)E7cLQlJDF$`+&Ej-KVOkjf+ebs z=h_xLd5W=6pr((vzF8}ucc)Hok;ovUXfL$WFQLc*FwIYgCsDsZ4ktf#b|V1zV1R_w z5sZWtoO=V*H8}WE|ao2(O}6} zuhRTSS+Nl8+}Lv$_VX`v0Klj2$9C7EPUYKe6R8M?R8y#@ity*xQIO!%w~kMVNltd- z3IwRH(?*^Z{Ljyx?3+{JeXO|O;T~hhXZfXnMNYj0Cj08ba&5=i^Z%#M1Da6n;B@yY zf)EoOn~`bkK{k2-*8No(Plppz`PLC>mRgo8rBBV$$)8Tu=p{OclP11o@44;a201^4 z$!y&9Jo#gF8W4ui&FQ6oeCyNzatM?51GJ>P3na}|PH_Yo@26?aovqn{3|QG{hr9m< zO+-WC!bBXHZX^*h=I5+__aC?tz@hVAxYT|S45{Kh+NYE1;0nI)4-pN0B&B|6-0Hsk zTsT0Bb~COi`iV%6rhT7HBN)&m-vi)Gc^RB#fMY6Zu*6Z-(pTRg3cvb;12U)Via`|Y zfm2`i%=vQl2e~n#UmId~A3#c73q{Y*EWX9PWJe6?ll=4+mzZ(kf$+iziG?~u(Iey` zO~s)Di&PvzcP2SW4TAf#>-+aYi^Vr4e)wl5ya|(H&|JIJFkxAh7&@v!EOMtQpww`Z z-jg_T#6B}BO3@q)k_T-qtu1lJW@~g#hiDkWLwnwveCaRpg@%7bHdGM%CH@M^R%n1k zIc&07-91mDE?iRj)l)EK!2%UO`c#QKFvw3K)czG1xXzU8x|@FaHFH>Vulr)dn37bt zounq7r@|fYMHZX&I2Zh`UF1;iErM~kR&^%uB9P)>u$eGQJxKBdfhP}jxpdRqop6-L z-6G#(uH{kJeYg=0(QaPI0aa|6A1M!^jjaj}Qu;~BfVz;y3aFlco3W~2wRsR79iQFz zaEYEx-J#__T<^|)vnyP**1`?#SnSP~-dcL{X?`9s%z`p=ch43{X4#&{ZKzDE3XXn{ zuC0H+IlsrA;b88ppZ*d^v*LK@n)%fQC&42M4X zL&yH(zgRfPP7=Ga1n+f`eKk*lY1_XKNU5nI4w=(9cc5x3__~$G4YdFe>FZSueXgn;1QjUF};ZLRU@- z5(7rFrr^v~WIF9b$YI9+vEaJ0OFTfUBy8@-q<%qW7(rEifdN{Io|oX=-!Di+R(s4}5&6P%2}yIlKI5B<>xAI>~5jJJ$r{iXA|X&}Fl;G20FU{WuF%m8D1PEv6<*2i8S8 zY29m$9pWN2(+t935?JLzX1V(w%j#Hd!CvxyG*dolvCZPg(xW{+WRcW$b#CZrXIvam zrAG`hI#@Eqi)#7IiRITefc#zzjJ!@Ch-&iViDHcE>R8k=7%HZn z+gB;Il0@JwC#ErK*>#Vy4dGnI#G~%eZtfQQ8#P!wFGz%xpr?eOc?P@I@q(58xoS{> zy8;6~E!DrbAoePqncA}D?-CV0G$~V#!d=J@0^%O6#*gFy+|em=I$7L>3fVE*+Zj@T z-O@I%YO{3rn&#XWyc&{?7WPhV!%HZaZ!Azzc;M#Fx!Q4v0=7xbph!B9TB8O`JH{Cp zxoMSjMeDa)@4e?mU2ixj%2nFycZN`FsC~c#Ik>^mKBdd}I5;E8)PN#2E*=)ppL3N) zD3-9RL@bPbXil#=xrWuC^C_s|O|rc9=j>73EAqghCA(GQ9Snw>AWJ7rqkG^I_rXqpOp=U%>NWr-rO2j|RXy`E zDfHEqsbDR$-Sk_Vte+`ga!6XKOKPck{Dw_q%NCSG62{RM6%vwvQ9 zUl^n>SzGaKFE@Q|x zUNYbg#$vMz61+D4VUsAOYmORcI3s0Vfk}@noW;=iDoGz208UitznRv#Q~D8S{!Eb^oum9!o+7)_bGY`sb+xIGwople)KGrMr)< zqG*-Vd?9FMPw;b72Pu*XK0%})YVVFs*Y$DVW7M_j0Gi_ia*(e?fSm0NOCF;31D)KH zHv~DhEnHw&ahQP|jiepJ1YcbcW5~hgwyMd`uRKWn)qLCb(lAt4@6IOLLB-LYq7M-_ z^x|cV?!`R4uE4#`xtaMuUv4O<A5t;2%6^{%C z(4cvT5{oy+R#f{x!K&4*;iQ^%MJW%&8v&;Feq}$6?u8s7sHQKECjmhvlQ2L%0 zLjou@Km|J%E-UG3^C3?!SY%2R(~;zz_o4*=hO4=#bk318r^p2z)GRLm@>I?3D9-Kz zS<{CR+=|Csse04npm4g$D!DDGNM`(7pME|Bk(K;=hSL+wKh*7r0w|d~m11k?r6m;d z6tS*wS1KC+Q2WDsLMD3~cn{>KMz=@+dDcK`$ zc^C|j$cC0ZyBf^zKgA_JkQjAwxX9K1?Va!CP0QY;3BQ}{pzIT~)R(Z1)(%a|vrnq+ z*^+&lZ=gU$SL*j&GtJZw0%(7wigd+>TFMnX?0*8r*NOB40W;YIk4f)EEv|rvbY&?G z$bwxkZAukD33~k{2p#G<_`5tpY!Hpa`qhOoy7oQX0`;PM>ap2Dq9FH?W|6li0w|l) zaX45qj(Ky+4-f_whVN<6Z&Pcb%I<9}-P8S!xo^DK&wSWCwZ+$L4{EA7GJtxlSi|PX zdjEDTFJXe$2(GEm2}Wl|9TPYCZOv*N8swYvHmG+cVf{z}wI_MN_ibu6*RBjtKjQ64 zPIu61iO|iXdr|meZ-XGIfx7eyvPc-fp4XKDU9j(DBxEamhIHMn&u~8Gy`dfOrZwnX zUB{$Ko9@Ip_#Ie$1&6}Q%ztfc17{d)fpJ30Y*PUUe{9VzpP$Csw6tcgo8%tzpbLtO z^WH__ypMU4;fBxM6tX8k-MF}7@1(R3HGEMZn1h$9ZDj+%Qtg1@T|VC~p?oo$9Bsm# zN~)OS9YA!+cTh^@3DO=ur+&S%Ickr%FG=QVycP6(swMJ)3;LZaKz}d?9-}cW1oU6T z_#14K_i>R2@pnH!R)6EQeMV4Af`N{PnDDJdZqiMBrnVm4lBncpaX+iovDutMSI(7K zIEv&b2|%vHD*Lf(%Dcpy@Vh2DMx#Tkh?-xy*hV-*1`6I7mFzV(0gijM1Zw!p$q1D{eOz^sWh-SJnhb;}ga4@ooK+$QTYa3(U zcQ3{G1ao8vO1UY4D^?cy`!?gb2UmWYJ4Q&4CM_dIf4(%2tOCxvxPrElEf z(4EVI2N~f)P7$;Irbd!J_ZwLr9PP_Lfajzc z)Q2&aq^aD*SChce@~dV`Q;*i}n4(>C2;VB!?WJsN61u9M8gZ>yKz};+9n&0G@spysvf(kk&YBF~k`bQ*D-3^5e1{Rr9?4`k(*U%|cUQF;t?VE1NkgVe^gq zh3vz!1eu}gNW-S^a{dOFfTdPuG^c@F5F;gQ1FIjERm=5LtCh z0=j77q08RsY>m&UA(Dj>JjpL&N+*6C70WLgZNdHc^(9RK^obqoCOqq_vfl*ojyCfj z30Mt;oDOz1Mh{v)dD;9>F4-1}ql{6!a|6Kt@LAI`qdC(^-G^3dkL|~Eirsg|?t!$W zTDJt|^j9}L$OD;26^+qP8VJXX@-sOYHUfMn$`qF8wk|kxA=pfDXXgLLQ$F^}#a&4K z5qjZ9mjO6J?{&ZmXg*#)Fv@C-VjTGhO&h`6Ulu_^o_!6JH}2XZfs*(DD0#12sTntS zx;ezj`SSc+3zc_(iDrY_o;{63VPVF#7MQt2#jU9=0o{F~yfL5kSD3t<^-b0fHGiE( z3X!lh13T^EU)>k~HR)x9lwmgF+YmV);>{P;L%NWupKck#W0(MEG>W=T#KGnWe01`< z3TOlY0s5Xm8%IPA)6SfEqx6MZ-A0B@pmn<6{4gK_h^0K6a8gO$+`GO#j0$d~La; z1{zZc$t1tsG*7ZwcM}S&CIQ$Ii7GImhr;2&YAW?{gtIf>20Bg(r(%8@0#IJNTF4e! z&j&B`=RHH?&tL4F%@MBq8=#;MlW{aeh5fSm1^Sh%OjxjyYOuB-sW(?Hxi7_p1Ef)c(6;xw?~Kqs@-oXsxRQ`HbjIFziApHWOo<# zt0u-cffqU`y6TmLaGH(50GNpkHvOEaQb}KMXk}@rWUW`-cUIp|#hN1A;W7|IEpOem z*%fiZs3w(AX4{PGd-u(7{B@)$g~2)gJ28f_ia-q)T-kBGsQ@?uePtr}Om%r)Q?R#c zy(eAyPmUYEMe#8dh@#Ii%F5ok`6&lFEOj1i%X`?ip!@C>_B2~r`f^=23&;TYgUbJm z;y?l0YE3R7yB$BW%(;(A*xA9X)ALe<3v1wnKcIHySXhRqK8}%DG z%Dv7sx-P*0VDfO8 z@`~t%Of=XLeSRLB)*HiH!=r>dMg?F@}Om9JAU$IW)FJOe^C@L%;FExyb70xVK`^# z9+^AmkG5Iot@fC*{gGraX0EH1iFuqHxR|M`YqdcK>|ZxAtu zC}!AW2QQg4iDpeJ0QkwP7ib$xm^|3MhA}SaO9%yx3l>S&XpoV+j=E=r2Ylpj^kaHX zah5BmjP+r^teU}YY(0?tBwACC-r-Sa4E+RkR|{Q*`9iZ>=?qSd$rR?|h{MgE zjbLiS@-W2^Op;v%WR3|y*;*$Bu1yIY*;%J&>vp*hTCXWi0*Kr!me5oqh}D8eXN^~W zzOv3#;TtMLX6N)me4u<)*APj~TR)wsXNw+&gdnnGXh(RjWir#u}HgRG; z)3*UvuYxImhb?tkF0cLx!(7L!n%wUNOUvM~&5uTo%R_{NQWeLCGh}!??xxz8zO*aI zbv(ESB8>phTM#oKE1>}nwgAh*t84*IF!i9NQ6#9|P^6iVZK5%(BgbJ}OFi!+c?!(Y zRcyY(53oJW=j_ECed@`z``Vjr_YI5YRkatw1~1QC!(6K=xwJ1(ARz@&b9gGJJO% z6#)N9S?)EyrMVHha_pCpFB-<8nFA5qGrC5Q7mEz0)84-O*f126Z!Afp@z}7SJw_du z{AX74Dp~IH0Z3@RAhz%wGcjl`a}f3$xH65**0xZ{T-qcj4a+2JpVp zD|CR9Tl6D`Y$yE5C{eRwf2e&NfX1$a*$6^&oq8J2XSdSIpQz0NlnHaU?Hl6Yk4d)WESRAJ9 zN|E)a^v0k2xmrb>y)K-7U;)@&=O<;JvdHJe!^5tfYsf-z05GgNApLde=G+#oQ7Lm` ziSTF3L1C_ISM&nIa`AFG|hAvMcM% z3VfQvS^Yn3&Lh*$GIFj}*AefxkC26=esexK)vGiIpDH2aRH7`E#I{6Ja;fo5ZzSaU zOGnyY(ct92 zJ6XBpvljukq*hDVa;tc{K}VUQ_5eB`Km zAb~l*b0AwkS9a>$0yL=>5+x~_IazmyRnq=Y#<2VWAT02jW8xynJbJOVJrd%T{suZr z`VI8m``vyZWo1_}-;?Qyt`DJy(^?-DO%vunziNnWKBCu?L9l%rrge!J&<9|tScc@A z3%Vuf@3-Rc*LdSmESD;`C5W5BleQ&4)JqZRw8#4=MhUO~F#5n6Eotw$jti|HDzOBn zGs^Ev&7DU#%{M2kW$w-l?p|9{6>@VB&kK6~!WHBS#O8yGQx#-^NgSD|X$nW&vuCfO zUSNxuhNlK7f#iM0k*Vw$K?Uoq_qHf-7pAPPIr!3_BjZM{w+x$+mrRP@e5536b?l#|1^|`H;zTASI;v zR{(FqT1#7;$-^J4B4!|WO=jKx!vB30D~ewEjuh8Sr>uCn{X78BJ_s;0s7~|0vl2oI z8`vV!1m(!s{8srqHFt>HPxQzBmtT_pDsg-Tb$sicIv=7ZP>`<=j;nE`lS`6NDYjgM z%ywv7_1+6sxe0MkAPMgI@6p6V`!`DgKo|W?n^Kv<+=GBb9+8bFfD@BWZJ{LxfJVHxo#Vrpb8Z(6DU@HPr#z& z!R{HkpnIOvcEoneCv~@(1=Yt{N68C6enS!4$X6EKm*aKT;d@%^)6IxC@~S>j1g5yY zmmCf^13NPYo^I=pK!w0*eYx$ip@l9{rwRrM8?~Wa3aRtujrFO{#$QxKSl2+|u)5M0 z|C=0{5H`=F;`7S&Z4a^1d9c@q$Zp+#g@Y~a`?qJhLW)(0l1(dwjg3u)=G|2Ux*%u( z3^oABzwn0*9SYkm!nHY?J_tDEtySL=FFbrXkiYD=KHVa(50=c2urF6}G7IW(w2RAT zHS>wjG^{;gGE6(lIEnEZAiNFIYzPS#ux}2a^1X@GOs&ftNFfqj?^tup=nJYRppQR! z5&(p2>@n(&4$*-65Fj=u)1*pYNY;gK)8?jfIc^UfYt%MJu2uPR*LxEj3oj`ZQ50nJ zhAAXG6!5r>VK3`M>VEAi&N-g-t?JckA;0R|5>D9WO3Yg>ae%<(5Bw}bvGAJ@ z;IUr-$zb2dREZcA2K%|&sc$OovK&ZSOe!icp0WpF(ORqxa0se zMO--C`>_n7C`PSDhSb5Rt`sU$t(L&R2P6gkR}n96EB%)O)~~(zZ!Bz1y;u-ru3{pR zx@*YcQ)jaDlS>>6kz53vQuQG;da?rGMk(*on*@ruse{psYy9FbGAu5|;3a$i1^3#myOod?m{NdUN{ii3w7w@+#TSWi&Z`26Cn z8{j!$y1gr=Ou8v*A>I`N)WO^lT!TSOK(qMz@S^VoS>E7 z->o#KmX_`ETt1Q33?%GmgNBpGS`EQ09|WDMiZU(;bvgAw>OnwXd_DJO#B6(B$-!Df zCdPUaFVd9HP&tEk4MLO2dYmveJ)f#Msuc0@4>#hqTbAaeNG`!dubu}K{x4XXY$?)K zXK*}J2^^zncx)Hmqhc!n4uOdU9@(sED(hg7L|3f}h3aB2wK(_TmkjU_{9+QTi}lF* zhTai=S8AQ#HxdO+b9(J1j$Mrzqtxf=p=(_}`Up}UJLPF8XnBv5f8>|}xGg&IEkWDw z1NpsCU23f=K=(4)ryb4g8C!1XEGRZP(-sF;_F2fRLwxs4V*ogZcjpCSU+It-he_?x zm=jCs)cXKy=2uq{EV}%Vh^ra#{Tk%i)&hok3Kp8p2;{v zq!Yz@*2BS2jVMu!g#kSmei{1?En1CZCucC=WjHO_Y_YwLy5t%HaP-3zm<)9{>FFYz zX>i$V|Nos+9 z$;MA(Y@$VTu+Vj&BPr8tVdfqz;<532?ib(&E224RChO%)3YU4+Df|KB)GT}6_vcQ$ z0z~~G_%zNMtne%Jm;jA|*YR2C+dI@oXm_8r2+6dBN1yR*`((T0Ptx4`mtYUkp(l&< zf2;prPH&OymRQm5uBYicHP@$~&|qAI9Q;UYV(;aJ$5=n*J)dU64~+#O{C0|5g}z+X z1VaNZT;%>@keS~kpJx&Vj>}e^aE_@qfV8@*dsr^t2?%a52>*_T{)lLh|Gf@2;{gB6 z=J1FAdT=5(h0iUaI@T0Op&;s|LYlofjx+}XZ$X(U2dDBC|W{9ykv{@nlDonK}d zLEf(oL#r-;D!>DWS~#W>b#gG;vEC@sg`pZx2tpDFv;iY7HRxgNn@S@Y(^k``vZ;WiQ-|G%g5=JZ->Ct zWf_h*=yb;hiIk=67`6VDylEg-pnvKA`O1aY5C89MjCs=im~g)ZQ5US;ww3Xl03^hI?V2>~jjA~sFl3SqEJTu#NH%tYa{i#C zzOChQAmCg|7720ww=nBz1QL2Q?TGr-?zZ->^GM67k}mxJjQRb$W@vvOs({Wi44=84-B62WHDTQWNjPapV^p=_IGGo5bJcscL8 zPrq(02ZL7Q{czR{Mo&<5!EMi#BS}pg)~n5L`uB7Jc$(a!L824QbdyHOejucFY=jF{ z3iNbT4s^^srt#&Dl2Wmo6@Yz&WS5D=XVR&blOg-Z0Ahd=Xdk9&)N|!x09hm;>aVq9 z5e9>c#wDJek)-}9*-k6qMDae6?+iixRt{ASG!dl=`G+oi{pQ>ruXD}SA1GHhjXi+C z{fLRuiNLcu;Q~N8iEmC|L?L(3 zLpo%Kt+T7|UB%`F1gIdgmzuN?55*hDeZVk3-zH;OAv4L(uD1D{qAY$g1I18@3uNJ0 z1i^de;PX)OeZ4hD0I=jr85=p?p@O_0wxtpZB6DNn`6`(G_V%s<7SB^G)JqBE0uQA4 z6GWEpw+rZ?#J(jJkVH-qT80=chq>Ewl+;n)>l52rC~sbV#h|U#%*NPKM>10DyB!Iz z3&HQ-zvmx#PI3N1EFsTq0WoJS8(+tPedBmG2Nhc5J+x_06gPwgF(gcL)r6npdjUg{ zAb?$eeaL+4v{Zpm0rq;j`%_g(q~_+_GKJX3PN^?Kk`I*Pvw-Tz&vj=^7WsmSpy&ML z-^hNE=JzMCWv6(67)O|G*`2w7e+WweZ&SMz?F#Q;HCFl8TRt`Y$)dk(oZKp=p0ZA;J84^-=E~;O-2st^Wt7G}7b3c+S_&3(+98H5SlN`RQ~mkCE?0&)27&zVqh>A3P(S zY0Z{Cp62d&1*rUguf@}`YquN2$yd2`JLs?>*|U}OZ~pw=$qxc#8ZbkTM4I^jewMF_ zKw8J3B7jI2zHu7+oqYWCK}jE6dfZR>pmUeicTP9v@7sE-50ki^Y=HvNHNylCsiu-k!$~;}q`<+W1HZ$U2`b(7IlxSGIOkJ$hdUgteTQ%zbwzvB{1@&3})k4nXM0`QDop5`dF z%zacU5lT?93#9a(6Y*$mz28F2X~D~&B}y`F^nJwD=SNmc?cDZtqE@;5H$za*3T0@4 z!JyBkZEl4U;O03d!wkMpxj0aIQ}f}I;~kM2@hB)yO=zvp-Gxqv+AH51!wMuEM)DK^ zjl|~xgzEr8_v6$7Rg->~q$IKPiy5HDk5iCJB5MEqED5%_a5Rp>Ipi zY{=`Uh_UtdV)t`>Rnf2qiUo(1ZQcUrGRq$yX$CuJwfBlQYG(qDjEf)aptw*Kruj6~ z$@W~uXAk`7GUXSqAa2vHyEoCTxBd@%Ul|rv_r3iX2r3o^ z0#Y(`NlBL=9nuXdDJk8lh#(R}iIjA=fHZ>AAq~PXO83wML;d%t4A0~9_q~4a`{Dic z``~qrhdF1?IeYK5*SgnQ_bsT8g!(jEQmGkEk75~pOOtIIOu4sHG=1~Wjl53sU>`u5 zJ7bVhFRxc#Fc8u2*GP*6CuA94*U6Y*reJ3oUtg`*-!V3m-ZwUjJ{d$N2@=nSPY>gv z^*9M!O!j>pyZoqy5tl~WO@^-hD(7sO7!G;hT`*S%67#bfNg$lsm#oY}Anb-2xbq_o zbLB5zR`wWwdjJ&wVl(3IH3k#X2@4CO+*dC^aS2i7d+ysA&#yO|1r~?(4ms?Xs2vP# zrl&|nXzDuE1PxCEcCrbpGf2b~yO_OQ?q}?Ze)LXtEV?-(*GPyb*+?vhI z!MZ@~yG)83Zac#)F%@u0I(s^wCu`GcY8uCdi{nFi;Tg%d>{-+$@i|vgXO^q6C&Rj$ z=1{`>t`b{-`h*({XknO9GeGj~?*LjSRmJ&_j}>a>G{56O!`%u#*ts^-95S9T zK-8Q9k-{50wxF`_e>PM#z)&HG){jwBZ%?d65{UJ4(SRGyLKP0C_)8TYKbN3>c!58K zM9+8$DDsp_rRlmza#!&p2~Se+Qw8yI$YjwO$%>m33`T*2n-vaAFRIKIw-fUQtg7fD zNSxK9CWs&kl|gQIU3@(JP>Bw@v^rVOt!3mDXw!oZ&tBjdxrGH@QF(v`)AqU{>EJ)O1@{) zseKgI9$PF__L`Ji^xnODcEeX7g)i>rFmF3;({QgK9|dDezPKuF3WP_wVVLN{9ab==R+EpeVBG>y?xmD<*q6T$66hO0U$6fJ|6ybg2$>Fc^ zZ#RSLfxOY=4CBaID3{PS#w)|Pfj=h>#0Pj_qC(SGkLOY7Nry{NE{6-Wx;Zba;Dr|N z+fbWHRt8h^r05ABNX#;{7^W-?T}ISW_@4KTo_KoAaR890r*ejY|Q&$%N2(~860b2D9f zAuLzZTf37FXXjZGE3RlbC#Y|X?ogwA|1%j%0rch z?{IhRa+aY~7K8+S`F<&qajkBmxef0U6Iv1}n(SepSQ;p4Xe6T4a(9HM8Mnnq#n*>d zPC)4l4Q53+vSxsE6oYYB?4UcbF!be19F6yO<$11?-~Avt4?T1o_5ee03|b!ve(|6n z-JtF5{U|U=p8s)@YzKk03D-C;hT&M>u14F{GAnrG*uG!uXX&U7O%Bd_$fXD-#>7N& z7kp?7+e~XNQ5j{qHM~|0za^eAyF3k;Lxz2a2=FiUqQeHK!^M_pipm{1M=Y}t(ap2; zA!@oVh2GcB(UVcj$3d;s@*)^(7rz7x>h3EzZ>EY(XKxYkM&Rx{rbAg%Nx}!L*!v$r z>DBUDbMkMrUZjwj_;3rb7Q$0!#J}F}s9vbFlKyyd$Byg1??bV`gL|E^0@Pqlg+F>G z@RW}UfMVzVrbBT&_rWKxE_4TQbm~L0qr+Z@=YeHSQk|{;ER8nv#5b+W9`je)eM!7v z=IZtg0u#JRLl>Bs2mB$)Txv6QJtB)?LI|zq)RR>F6?WkOtBmeI%OnK* z1w(6J!&Gjm6?A(HW~iwRFhqB(sy~Aa=!h!|UZ**GL6&?lUq`JBW!k@YJyW}|Y9bOK zihqo6R?zxRHN`gqF`?s%E84-yxX)uh|DHi>wR&%LP?wG7%3D_D0DL9}c|jEsscVe4 z3>m@f;MBSs5ir$u{|PDUel-fI@eGo1agKSeC-r(sZ!_rLq1=NPrb&d}9#G5N z);9o_InK3y{rFz~@u`j!ozI!bVx>MhTqh2WKsxKut$D__qjS83?xF8jIb`OKq2A@6%m%16hofiw>%=QCMCZ@kKR)pLCpW-I;D3qf zoCs!^_f)`<+B}AJ%n{6{^y}mmd~kCIMbLJP?m+<>Mj>Lr6FHuhzl}VO2{`9O^_~m{ z(vr{yNBkE$27kZ&@tGg=z*ms;VxE}0_I5(Evv3k$`tz#`p#g6f&o??4GiPvUz-9J` zkN^Msr*9+R6gOA*o5%Qo8aT3N${qbjeBdXO$(N)t5H~K5o*lH3j%1LyCnCbXVT7on zx;>tkqg||v-jC-8T(kqOJJ;^!CEsqR_7xlN8fW&Q}K}@{`X{Z z(yOBf#NwAu1NmMx*C=t`XBz~p64NpSNfyOsQ*;#JEc%@%`S$_|dfR5@&unif13q_} zFk`pWrkk-0(K3X*GEi#yJcO$!m-0nYmVvGh^MY#9y5=jfOvmyS66hYp=jYb=FthGo z_c3;T1*Dh(ZDr5=5)(@h-~HBWc&cOxd}>HqS}oMao4a2VyS0R&6NjADcsmX|gzGvKex3#|TwS&f)q_a(#mo9#x1K!dS#d5V#W=+O(d!^<@qr9v zmbmOwW@_v!m_;nLj-7*U5VPOANWz!HZz`awp~>O-nc^-ebPWHP5E!YH-YK<|Zq{nX z%)t(q4*8Vw?r5Qs+yLI!qX)Ve`LW;UhnJTh79OR9;``ZCP-2@wUPa|M*X(1XkrgPXJQKRbEw2_%YHHpj z9~*uipBd0}6F{t%wLw)4gY^YK59ur&L?TFTDvs89kz%5F7{(zzP1EpTe|fZ=;+c40 z=tPtJ{1%{al`M`ToU{3A_k@y5qll(Ur`$y)h+sb-{(*dFZ88#&?%qEekm0vX=rD*Y zVy_X`=dS$>0@#D1lAGitOiVHW(Xu#eN^phTygNp)%?n5<`iuC$|GW&g(}bX>C2w6j z^&%*o-}Tb$NR_i~OVhsV>{`4#wqadJ7ctCWI;AW5Il1QhCG%y|o*IkunaYW=%ZClY zF$O&w%=X+W^Wm%oR%#PE6jSxS($l-j-+gDyq?1uyqR#nB)KM{Uo_mxNR9#^HQk=S( z#qGRy`HAVR^T|F0Y|{*1#*CD(=j*-)9lyyN_*jLu{ruaC!)mKrdh7UE+ z%L=o;YCKzGdQqt<_)fwrMu{Tx&xAWJA32jw3J=hOW!6(_sFb;~A~Imwdh~j{K)u`F z=JCv-FlX`N?@-yS_)Guy_I#dSw8OkAXSbL9&*;&+wLUE7dpdB+efK z!x@#NI^UKFrGog1`Dm4jTr8KNhvTf5E|gd8qC05Ya~)mZBC=Q2DV>wV zR3_dew5@eGW=~l)+|w~NhWaPJ;9Bx}EFgM>;-Az6i5ekh!LUCPn-zEqC4t7vh6cl3 zE6?V&e!=FC&-`SQs9wiMxpeX6BKQd)&Q&hgpX{L8rX9uHSyW2**0P-itv4P>M$n}? zlj?lW=ifE_QlMa@1n`eGUW>OeNa@#?Kl4pw;5SVXLJCP5lond%dln-agYr<@geAK# zCkO>*Pl2!{WC9P ze^i>8dt1RMAPgntvT=t@>w94*Ji z@UAAAy9n&91uCqyDHcRBy<4SCE`Sl#IVr)WOXmiR+<0G!G~$HK+9ZLF@daoszoRaX z{esn29G}Anf{mxjTLgDTM>{4uICUfD8!ah<=s`VXazZ8i)a$?6Eh%u4^U@RCWOO~J z?&}1}gQPTU*I;F(8|Azyf~7T2C_XgVLQvl3KP+@KwwIq$fHisnxBuUZ^JtlFzZ8OE zOJ=EMLjb4%0geHce>&g**F+#}jd&A1WxI3`L{KiQaUqA8X@x*x@M`C|KUsO=Y z){@VA7!Gqqwt-kyl$hAAcZhPbbTwf@3XrlAudb`6>&hXN5vsOynj0rF6%(tJ^VBLJyVm`ZP30qsYYjlqV?^cmpwn$zr%6G6X#_ z$gFRfWRh8Zw1(puiE;>4@q+g*m9I4LqsaX2n?j_-9wfQP+?ZYusxO*?w?(Z-&1;#3 z7#xgS8Lo2D6T=?w>m)f{nd?uj9%e2jYXQ$P)DeU1sZ9U5Zrv{@ri1wD)Mw_ofHt=jmX4^dHU7kO=x25cW<=t^CSTuRgpG%~GXh zwkmOfNw$@AguU}7kptf7en4;)wQ?Yln&kQRGLs!F$n_tL|Z1A4H z(M*2|W!G=G5bWP;)P_nZouvbh)Ha}J*teZ&H0WS;m#r!I2g|D)u(&24ev=0k7$RxyU%aa+9Xl7N_q}T`N!NNG5=-{x+Vaxs z6nz->_+D=~_hstT`q+ptRK;wdEs*>C2GS>tGK^X5LTEUx(i^ziA=#Q`na4B}@c>!N z&UyS){iqrvIH(1J4R$kw{7C3%b$H&hj5r>wM&a(7 zS2-tqA!`|Y71=`Sx!jGQ2aV0ZIr!vOrWrMD<-qFb@0>g=y(aovyu;!D^4an3udTi; zELwm+@TowI6o*}N{#Hw&^VNQyQY}wgMHVIXXz?$_()MRI-@ZMv^}8S*Q|1UMcSYve zA$W&y%`*^Gz;Zr%gf%&|devsVk+JE)h85UMf33{7?)B%U!~9-P7%vlif3+#N&j1Qf zSVvwER2EMbEq1Pb4v${zZOk>Qba7lw@=blTZ@j_8#5Ua?rDEp6K0*pMVrzCeanHI|a4pX? z7hdg|X?T?8Yu5CBuBW*qd(hqv}6FH$M827EHsr)nuBu8rB`udJfDbc5M+%(KiimjPwu+Z#k`${FnV z^?k^)`dmAprK8pUF9K?RwJ9!(q=y*ZV(i4#j%AS4w&BId@n!DnAfk3yfp$ zePRq`>y)IjjhNdavegVwi&&`D*`Rj{7tTjAMIlu4B1XzAdEKex<73xQ^vr~|%;U<4 z55wE*s?oLEyg*vKDp@K=RUC+C8iPDnbni^N1IG(attz@U)f^T_kIYR2`zKpc`Y_|( z$R$>XdOID}%4!)BwsyBtq@T!ZV%eaDT~t?pBt8+JSt*PoXgu2LmUVIKkoPBh61s1_ z^<2wT6(x7|Fk97y0olNKB*+6!_Ds=3`L%AKWi-XxEO#`6z-g4bvxQN|Fuol;uqWx`GipOK8ry! zPJq%Fh|rCG!W($M?AM zzaXRF?y}GPDXW5Zna~+sMDAG%3_`Gn9%(oeb;Ey;%bTIk4G9a-G0rO zkelciB@%6Q{0B4$xJQ2c|KC6TzXNd%_YI|vW~x^{8f@v0c!a?fjuy}H?}>u1d}qdZ z=|W`PVCF&yIY=?D|6gx|Ocy-Q_~%BEqu=@Kg?#?S1 zeWcs>k^0dQ%nx8r-UD9nb3qL5_LsLPx}MLZ&+WK^lA6b8ze_pT+~^xspi=>*W6)&!g8djy9>f z=6ugFw{RJq30PH_Ao%l1cwmi7j%%z_9({%E8vylB?Hk#<(Q(I*kb9dVS*~KH(x%uLcWfop_wP22ID;=8{33l*6{$d?| zN2B=f;e8netfN!m>T!FQkQ1BbZl)Ir$K6807!I z7zQYx1na|Dn&Dzc&-ZLhpay|GOHW+tW^MV=i#c)nSIu2B?(w1DKJ96dNj`&AzW%F4 zh(ZFdw?Es=6|l6Q=zeiDF^43;HuCz$tHeKA>eDG`0OAIN#+8X6NcC&7v-p_TA? z5eM5F-1y@JJ)j4r7gf$_Xvb%y`Z=`p7a$qi2W$E0mb{+>+b{Wn-0JrW9hu+DLUl1Gfy!9M$4hsD%7*oC*-m^Ix4re+?2Rav zf(w!Ol+(hRL&#);r(7!>bjgtC(_mIvU-0@wEjs?t76yZmBW$XA zVOF3q3()COl3pTaTE}OI1cU*+?iKfZ*Vb0+f>RzO-br_9>Nq%sohlE%WVjxwcS~g& z(L|b04jQe|-6e1A2kNKV)HWPz_vX#aN&r)>7@9=!H4#S$D-e`pQOQPS0Nt43%!CRB z#P-H|D64zpB8^2?UoIn1wTuMPgyK(MTJ(&2VfTUuP)w}IX@Y+^qhusY z@}veHh9wW=apMF4gl|q=d9X*{_((3EOE>ss(55q%+hAaWcCGG| zZ|vK?i&PzO6UG>iv2kUq>bbtcYMD3epDjWlOo(Qy#h@aS?|cr_h4qTRUadDe-J2Bh z;7@I}pv*=7tP_U>G~|n)J@U$+zIl%L)Tk8~WgaU2BGe7FLj(Btat^A#Zc^A_2fHOX z5Ned5T%+$;2UwmzPfuf}>Q0TWa6Fjq&{9ma;jqOHGs>^N-A|fLg?0YgtBI&4W+p+i?Y8#Xj17%oSxw>xGVYdb;| z4u^WHjjf`QoA>rx6)8!>8($N(iWo|@ozPzd?T2h*#gb*2Ix5O&X8k~`!q3%) z2FfAtB4b+qXwZiq)MX1px0l5pUkIg98YJ^Cft7S9nSGlhlYHHj5DDh^!i0ebY^xYl zh>SkTsNUTO8@8?Kv2TlFj0ECQ_aoJEpbiMTzNx!brQypua7P!W@nxMSUkd!k=0u5A zifx;2oV|E)z};itBAwlQ4?67mxlB4my)bFIn$CR3FuQ6)imkYlL!!u}MR9IVmlLL& zvT~~E+d?^LrsFC06cv5`-ekBDH*8~JL`&CmNu@Xr(-ZU%6uv+1fA?p}VK2m4h$ff} z;$JennnAIq0oZfo6ss#r9eKvYI5b7*x=VPg$w)W>csL4D^CUEyioy4m&K1q2+mpab z8Y)IwvaYv@VhQwd01eu?K~FEHOGK=Iy1?{ZTb>}5Y&6v=>t8aL<^ZeDvu*Xp~j8dIbBC96J-)X!y7i*byEdi*%$<& z5D(@E?$9WjHKSFk2=|bkK~}aTeU`R&Za{H>A^=Mrhy6mU!xNc0NM{`jP^koy3%K^m zE$UpI`AAXsrowuSYTV2Q3sCx=ir8FGjoCcjw6ncxu_S?>Pkk!A%@QEAVW-8Q5FQjE zckSs~Vp;gT@oF^`sQ&4eePRls5rzVa7^wyqdst8-Ay1)004>?$3(!wn9t*q^np<`u z1`DN2J(HhC$n$z%dZ055rDeH|Euq0{ZzKg3E;>Wm$p9_KZ)pyCipp+-;gKilx^l0vDh9! zxkY%YdHvz?vBwbA4@VG4RbB-(XUn&PHeWK{8ejTB065u}Uo+4U&Fvnb%;s>xe`$Uo zIjVivC2{#K5IxOQELVIV9LwXbdDnahqE%+Wc8^lyy=CEjekHZF$PI~G^~6eYIC#T`q>rwB7;1XMEF=Z+T^sGA?6V=BHEtEb z3&H%ZTT-wY-7o!MJ}d-5=x$9cSAZfou;1`~J29dv0S;&ZBCaId52{sa0%UvcmFlGVin(bZBdv+l zDxT-1RiH1^Ng+*=Ynf__tN+yQtY&m<{GeMdpsor``4?4Y_~5rbqmf z=%8A^1$;G|-4c%gFe9`q4pp>_CW;Uf+xBy*sp`!2q(_8&d#F=-@PIcUgIKh$KW{0f zXKG*0W?_I4upP$c{n&`rm5MK70mT(@G;RenAqK9p*mvEBBGn3}`OmDrMf1IvS!qu1?8GXH6BW=a_a& zBx80j+d#5)ef=o5ZY;_85n6Lnu@3NdFIYl8pPLGs6!VHae-!8AK)hF8Ru&-<`Yx@D zQg{;dUD|(t17u%LgSe!SMC0nn>Rse1E77*;Z7AcGl!GG$SajncHD=XmU~`a>f;iH$&=9--`IS-Z(_BOv4PQf5I$A%Gy4x zUhNoseeCm$0ef`+h-kO~U2}J# z5`MV{|JutK8sdZ8o4hF7OtaInw+7)Lajq6qaz}(F$fe0@v%@$clY-5tvs(oWtx9GM zx67gv!Qp2}lRm2KoUh#iDt@*4EAUcFOaLFBJ}5VQ?@}DMx7~XteMZOXmiXF-0o&lC z;F9a~;w=cb^?d9od(c{20&iuCJP`rPzWo)Z_)6#YQHcbKB7ZRu>Lst z%$jA>7L@$i?xA!bs3dFl5+-eX@mi1zQ0n-WqE{Z4J#nsSEE`Y>cEEXmo-PtumVO$DHv9YE<-N`k`SJEaP9EPkp% zVo5vfiKY^>!HrG|phDt6SZS(O8D{4+W5JIt?&*Q?T)NMq2Y7Z{vYLXZ>8og2QQt49 z4ZHxZpt7?U<)|Jx+hVo8kET8fcYi$Hi9iG_SiqdE2DGo1-h<{HIdlN*jDf{nsB ze8}b{Vpd+_w8U@*JMX=YS+<#&mgUxW6Gxj#TOP|MRC{Q$cPelboGZ<)(&c@OmhG9EU^SitW(Y5}IGx5}B?wDWBlzi^!D8>b|XxBx)6 zC>cOAPrafru{-;=#m& zOLXQ%t4$?~cg9u2&iu0a|0;kD*#OZc)Z@aglSMCWE~@V?qOX$$m}FQui3$phvNQ&49YHnZ>ac=0Qb*sGd z$(S6-k0IrwL<5KNvOY)*_=+=xE&$~p@K?axD+@Z^Ao+Jn2=FHB4#2Q1*1l-R;6eXx zR~`ijjJ5RNv;onU?;Aciaqlrf^y#BVe|QN@w=H#Qv7^F7oKFB>hfop!lx4vzh+T9# z_mGekQ=nFN7hKA`SQ+=)(PDZEUh5~u3m5(Ym3rx+5!6?P26#t@z4*a+4C%++#PB~{ z23R)qjop!p{_A@G#(qb@r3yO~SumoPb-4fpFJosveu3yu9_ack=EzV<%;7D-!}K`V z{&D!xJqXEwn}6Jsc;V>Yf6TdpT5!4Dd~!AnSY=KDE_GM>vk+$EF9Roz_I&v#NV@F{ zPOMcIqdLMU{yUII>mAI{{|@BOm+{{f`QH`!U(O<} zzp`LS^;~i^QvnmQYy%Ymje_l`UgcfiWS3Yf;+^UC2u8TVY zAOS7)a2gw%y5leCw``CC!n<}2LQYOTJ=ib!f>_Yw-r7`~*?_{)#)m}jawA_bAm$%f zfmW{;=m4(D-jPRm*P$PbPoM6K06L2MBe38YPOFwffm?#keoChz<(R)A<(|zs z(gGTXqz>HBf;Y?)o3Gt>FE!7yOrXI}oB6&c`$2}%_);y{(l2eCa;C#7@~o53^i(hG zuA*M#^V@CCs}S6Edu|S^5djfHy5C@}p4F3q4Hf^FFGJ3}K0v2fq+6qysn@4Fp`ZzO zPT*kvyV*j6(~m)0qy~GrhgsuG%wDbrMkDX`C?*p6;T;@MK5PRDT_>$xNSgLL%WaWz zWvICxzN(32O_iQQDPH1F8hKkaB=G{5CGA_S)D$rUGz z8@6%X^7ohDf3G+L5>5;N zzpT$W*z5hSh}bdia5WS5y+U}`LsE@@SB98?e@)4(c(4r=sBY{&^#XbiTEWFK(TY78 z-_I-Nm?%()DrzF~#)CilPtmJNRyy*`b7%k6DJ`oT9&nvB6rh|?Ht~xJl@Z~zY6@l9 zPrsBh54~5YUb(dGdraMkssb1o~kLaBwl1#h(w8j+22RCD#+J;^m_!fd9frD*}>=fCUyRB6)mF;A|%5ue@My75bc-ZBnj3`X_QG;%= z;vDo+Hi5}=$#ECw+W>UxpPWi7vR`o6ABLA&g#uXy-V5LR z=6V%VO=Z)0?AKwOuTn()$K^J!I3lD%AE(;s0Kp2w{fd6MrJxwJ#pu>=0{RYYP#$|z zG*jhdM@&=qt*J7U-_7gl52AP!DS_r`6zBt>cSSCR;m-9+1D1As9c>@Tpw;LP1|y_= z1YN{lf ztP(q?S;o;=fR`7=)796z39GP+R7{m>@guzu4y2{$E6}+JhfN!uAb{^b?Z$J?mv)4y z`HDm`YcYU~1?PgD2)m$zIV>mG&R_cJZGrWiUtoTH{Xxtt`l2;pB!{qt$vTse;!ZIoIND@2*6ag`; zg?lxVpi)_>9mfEBp<|vu>*b#kIG0xnz}|Q1*&2o%iR>(T#hu_u_-CajotjA};0=TmVT^xKNjU*J5)VEcY7o+PH&6U=` zY03BiZVA>$dnV4POzKU2K(z7P&rf=tQ?o5Cp-f#6p{!at$e_t3KNR=L->s*~3BMsa@E_$QFhEPJ!36xFh!>mlJWm??ru$eTiO^iR*6fC)c`2vx@+xIV*@8T2 zmGY7-T}Ov94FU@F{W(@vW=MG8F6fVB_K&IX?2S`naQ3(+RqUlhKS7X zLz*H(0q2c$s?_xaW6+&E3hNU7St|hpIP7Hxb!w^M2=yGZ3HN zpAx?}T@JK;oEMgw-+lSOLXj%Bl-^;L?fEk6LHXN@{#?HXf6Ecl}0S_`ZvUaW;! z<946W=ZvP5jv_Ftl`K;M5(h@I3AX{lIgR{9U-b~pW#FJ&ouOa*!M*ZU+@9V`)OZka zg;JUy=|C;p-frc?Y*hu)w4;QCIT}VOm2RrT>?P=7^yWfWag5^xaEXmqXuah43s2|E zr&Vq9+FXS7zLRx!c^QyixwY&V4+zjB{pghk;AX?)_H!`$wQXXRqeOgqq1(aS(@f-Eu@Ma9 z>QmNtL%ZDe`^}wG+$IH0-w;2hQ9mW*cX2VO)YuK=Er}m0JDGeOw58Oku$$ghd6Rm- z?XJzCQdgNARh?J4~`g4Tj&?jgw z%AFJz{2bJl^K>;Hy52|NCf&o&R%KxGb7XvZ)^-upS65V-%yn4Ajp=SAV)pZC+ z)x7C*s){_w`8p@E1X#M%^vsxm%Zcu+GF+OU%97GUyS|9qS8i#rpVU@+wIox z=t-3qU79Q6;YpV$RIJExARI2SjbRc;HdH*jUs-A;6@{cYMF#T03mv)<9`3hAn7gYR z2!X^%W%_&9^yFn#AnOv%ZBb;<@s^{asl5U;X7jOhu}sv202eQBmB z$3#ZV(7k(>dP2Wy8U@f4E!g;!K_UTNOps;W+@Km~lSOsMj=o=k#=k4OeX$JKm@wfXMm`nKT7OZxzei%^|uQ^Cp43FYIA0!^a&-bqRRLp(089(`|cIIx&LC* zY=ea*2qvmr_nnjC+vPQ@y6*IX3u&=FsMh-kRIKSVaX0tI!Uy zPAOnetAT5bnT~t~l`msgAT0C;j50R`HRCNwYZQ7AG-coaVwm<%=IF7?@rL9T0?BUG051#i*x8nyT69HJ)F_79^+lMIR6 z52s=Y6f+-Vft7oaWKSzTYd@z(?NP88=->u?+|59<@Uy)*u9@ETrw@d&9S4VtXHp(V zG}*29%JY+7Z4B-U&t&qF}^eOVxX0T9!ze&XwAB0+Dd#BlTd7aOw!_dUKA zD7@vcN^L529n41;T3RoD>&}%neln6)!mvNBH8(Wp4x33Qt%(BlmY=x92gQEO#yt3| zIgia&L&a@x?8o!>evwj_q7m;`hBkzgmm=tS0n26D^mIqON2-S&NQ*uf7oD}SnQ&^D zYzb@H;YF*&;2$EhhT#?T@fvf_d@m4j`hTTV&?zzHr_>81HVj%D$knnz93IHSu_Z$Q zpJ9*rE9vi`qB!Zmv3x?xa1Via_l~TIvDSj-!LK-l-^TuV29H!YI+=ddLxeaSuDrBD zlT{r&;n_y7jpg2*?G-Ry(v5qMCMn75r56q!4)L6*RLE@=so8Th0X{vVq7|nvUrWUg zy)B&e&gXRXjg{NC-__rTJ9cUqvsMlhLl+G=hxa?zL*i_v!@C&x9<8@=A0hldm(=pL zr~DGoh~fK1Ho;s!0v5bBD2DkltGIDl=~J}XlB`Q6Sv=$H-{HV=GBflJ4+Ymxa6wU zCN2(F!g;2&YVEz_v?&T4U@lJvKdyViUUfjD&Cc8h@h+JGz>4mP$^NY^2#kS{`e|(t z*eDoK-GoLLc|R{=`=R;hPpc%%2`hF@Vz^t6JKr$eWHuN+NEIkQn0kdGAqud`2WYmY z=_Tm0nl#_hF@n9l^m^b3!0x|Fb%dWl*$pzuOXrhx z&F<#_*2H!?&;8AGzEuN}g=a>tLZ8dr?PL4N{D~%%mzZ@?@jz!qZK*Xc+yaPJ>%)-U z6n>InjjO0VP>zveT^#A6$B302c}lSjFS_{6?D^MXSGF(Vf%*NBGW^Y^jt=d|iv7c6c{PDV6B;C}6c%nq47oRv$ew^Kx7PH8 z<`S%P29pE0QFG8C*qfylC7l|xneGZCl zzk8Zwgr|!rC6d0e;$3c$hz{g?zK9FW*dn@XNkvPW;E(!vcy>Yrr=*%|%h^9$PaZK+ zI~1kO(gs(YG#W0jjHrGyrq(N$J(u2}W&}}S3){Aw7YV$0R>z>@t39e`#7Z`-^^(9l zWHBCu*1*lAXlh~;b8lV0cV4^g0=KPU z%S`Q^6ST@-cCdqM^jn&(-Fj~nP4r&VQN|j$TItZ)Y~j&1RI8?nwdc#{_6#zANV72D z28q&j$%OMPQTyQ%Gg)=L)fb#9yO4kK*5tVpn}_XN#QX^U&BH<2g~2;lALlvQ*s)Ta zOz(<~u&l;z@$7Pq!e*MPOd=G>FWb5!AA<6lo|1*TH+u;La~IU~(1Ny8kJ~hxrseJkLmxIh+4`v6 zpw;osW@}NSR(jx#;GH{d8Ls`f4ADBLbYW3CAjyi$Hh*f*5x-rQGk5#rGd+o19{DWG zZQ}HrI{Y6&>uH-rVBkhqzp%0r`D;R_2`9Qf8!@)Xl=p;Dm7g*2(957}Ukrrg+1DrD z=6nWWGwz5?Nu;ndD}$*j59Dcv<{84t{ORtPP1{jCb@!|&+PZ4&xhpI6`O{bVId{)T z5q7>@p%YljPx(vhPYp2R%f#_1WCIZC%Mkld+~Kwu8DLA4YQYWLJ|DoP(-!#D0bA06KuK$`Ox8jt%iyd= zshvU{lY?8us~pYhfeMn@2CmI#;T|B+^zn_a=rbpmA8@_*xGgd?jM_64$WFIn(+d>M zYg9NTej1(7xF{ZH2N;v*o$bdL)_R=eo5iv2sCs;mv1HIz*C{G}=mYr})Vuin++!jp^T_1T*7_40UVP`Y z+1r-4>EKW!(ytJfY8ut|9d}OAD{#|hX_$%qeB5+qvsal{kcBW3xUzX zv84pP)I^Cd{i*^!01hltb%$kCpxb+!5>Wb|7S8ZyHwf(!nlis3^Us33v3?u(z^s|` z$@GqwPF&1-r&mkFicD zXC#g9kHLFm3_Au1IkMU&CG+3<;k&Cyvy2kOZ1Dr$jrA1s=$^Udb$p0|Z>iOAxTOQ% zZpY?4C!kP6j>ymL3kgHz9aEYMzEck=TSw;`YGAx6 z_PeLe&ff0td>*C8e(zgJE>#)Pp*Vqn&3pK2-TWKpbuyXO37 z4$zmUxLE?eBXw9)ooh3V-4(2yP1l>2l3=6#I&qRep3?d)rx^pq3rY>-TV4&~bmv+W zid`^+A?h3u9K=;7xIsRr9k z>>&;t-u8p!r{{nPx5ZnzkiivEJyh}lu&|lmE;dScC!Gaz(|za*Im!+=8IiTM_wW+t zo^TMEXo#=xI;%YJRZ%@4zKwwAm)V)a2P)t-#fR&s6w@tZ1?*>t%l?PGw~VT)>l%h_ z6a)+qY19KqBS^Q1q@;kfTy#q}hzJOR;GrAo2I)pg>Fzkv-EnB%b(Di~-`9A?`~3LE z_kF`ZjLqI>uRYhCbIn?F&b6Sx&<(H92VD3mxX5DL)~R zC@{6hzz)MZ%O!{-A2r*R7Xv4R(=o_!Z&C)t|E2V+-c@T2prp~_LG%V5wqAb!tmB^O z>?rJkkl8`fO$MWB<=y$!zFuio`Kd|~${-pVnsDaJtQ*o8I!zYv%TUcQZ~Wi3uw0Z$ z{T5syC>1**Pu>gf`;-Hd*FO={ zOf=6{cM^f>!Uo{KJKye-gIRR>sOEmr%p+?2Au0Uu5gO(oDc}RCZnM-5ns>6nd|9#9 z!;sj1F zWCP}XpY`nP3kT@Srp+FDJJ_l(zI$@tF^pC*yZ z>3dS5J3Gew?ij&^I<}A2fGOILH^*r4Rxxk`n4~!ym z(sd{vvb3aic^hN~EJ`u$UfQh0tJQX++Id75f> zS<@A)Btu@y;>KoRhj&$|->|o4xOrC!J_U$~dW~8kkJq?@ZOOxLfyFTIHJ-jQXo?n? z_!KWi#o(REV3i_6f)4z#@FJegoY(>x4@cWJF3CG6&H~li8oFll>~$v*&wf`3z7H-c z%2fhtVw2I>rq(dFk&r7ZPFee{N88~6UcyS+LNt8}@P5W`7ebAcU=@VF3m`$eyU2cyD?MP*dH7rMeZ2Pg=$qM$6qJD z5tK3IPcyWc@{VRoztY$7f*JG`s>SB4KZGaxK;TF=@9kCm$xy8x3Ze0=V_;%RO{n^} zVLDJ87cpdQE(|CbS7sz9JJYc^l!3#l)ne1^Qx(pqtbMIMN`0SfenI%|yeY%&A94%1 zyBNv){56DT&{DMTI*WKsPa3JIr9tt(&}9og5}Gt>dhPXNJm3lYOFU@ zVs-2CSU~J)ljkAA+}oUkM^vb^+|MzXVJ;pHv6VKeV1$O`_RhX5pcCd7e5;5mV??Q( zs}W9KHeP3A)FK2YYHRG&T3L|Uo(fsi>K7zQLZepQ{y?obZcALrYUhV_z>z!+PK9;E z?8-T04Ev}W8FoXhvH+(c)e_%l!0SEk7*uKMiwJgEroi}Fus$48y-&M&d|w|zH!p*> zx=Xmi-C_{*dD}#&I6j{JPdAMX(Nsg+{_fm9zu#kUQpTa(zzvz1G`ektvu9wRtLyQ5 zXA`cpeICi-16-O?Hkk(wqwoR{;fA$pG64uavjeRXMs8=qN(q_e-FOBWm*_Il%P z5Abz!0YfG)NKUFLCxC+Rp37?pm3&6f8*x2Ge&gqbozKZ)-wx#U8Dy|;O@8}Ax5ZW} znw>e&;n5P_+vwwlX1}FIn68>D#}2Xh)S3q8t#H^)*D#++T(pGxqLFmnk1SPsQZ(qI=;9uNS(@Fl4pM zSzO4sg5NUOk)OyI2%IUx8~a~ZDi75UxTs0a+gayVQ5H8eW>f$>R}^aZyNZdRxE2J? z|A~*0|B1f=M8nilH`lD_1S19<2%Fhg#)Y?82X8;LHErTDZt2;-sG^>c^NUxmMVYMe zjJC|DyWKSSkRhmWt#B8Fh8UV1NRO~g8C_u-CUCs8*E4tI9qo$uaiu8}&?3Xa^8 zewy16whS+`XHbe#zm3tsn-YRmQ&-jwJLBYu6>=;>xeq0$P&B8#_3U6QLQsW1$!+!4 z;obG&7zpg6%4UDXurHSo!ZH@d8xK9KErfO@Nr;Mt=7pk9mu&<+E6Iv2dZ22r1lr`^ zr<1UG8cnHG;7zSaU+)wOghbh^;>qS?Rr7oh14vfhhA;A@P8eSjo-W^y0!K3ruHRSQ zxr8{xsRhr z-3o66$LLxMjYz)r%7w*lwH3Y`r?gn_Y=RzjH{$LR~CoMB~-({U8x0U z4hx6`77eAr0k)pi>R*YRBF==GJ4e9vXNcve19!-YgYkUh)#e-1Uc4ay`SgA$4Zr8* zfV^$ch2rze)7DtLpApCIEQgn!Q6N2 zehhqx4@%}Up|$x{T@@aRLM|TRXNFDu5L0XHU_hF-ARRCmz*apGQL4(qgV3G%rtPN1 zdwy{jFJU?r?iU0M@`r!n=vX3KvZ0-1j>{4MM!NaH4u~<;hsC>~AJ|xJg6s z?>?Yd_`n2^j}24KbpF>2zkM<6%fJN+NWyPxfn{DlT3naT(H_nazG{b#1yNWV6PX*{ z!d2N`8)TaL*oY|tJBreh3uD%U#65|pQMh?uDCc|d`lEl(XjmZe*UBTya7#m2%C;^I zn;AqlI;W2~-QfQIx1nC(HkU4371A)-ax9sFwo+rvx8A!(*{;EYZGX@38Op zog{42#kC+p<3s)G6ELRy zviD&LATwk^YxwKeV~)puHE@;B4}QzwyruxqcXz>-0S~F=Nd!UtMMH;g{rvFWK4+83 zb8-i(MbY*hv=1S*OJA-fnQ8A*Diwy6@>NbJf@V|>=W!YB9tYJ8*I#y|RxT2PEntOQ z!Gi3rI3H7KHOk&#GIt;BW}tvM?3eQU#=2F3IeE+0R(epUJ8`g1!;=1P8n-I*wMn@xhAANE_xF=;kjwAef}q#V6_X1{ z+@5jq14TAha4jX#wJf&%WWXWF!V|t-iB%9>nxw{({3&oub}2@r)>4V3T~VKkb%x>d z(oV&5q>Pu4Dy4APwtgvRy?V;Y1QQFy80yP6y~H9m-TNGCy+%l>g6Ou`aqbJnqFOR& zGwrsAjMNO1Rj{k-_oeC)HG&#J63YdafScq_V86$VPKhNe*kN~lZUPZyg-UMmZ0i_U^<~;@7N*gAp*-8)+)FY!p|D7D>k(TIcF@_L z_G^4!($^jN&Y+exbkvI+sv~}D0vF|sE7Rg_@89zv7tT%e$oborX705Ac!01^_+4v_HS~QXxp8FxW z0(fKd&ktip^-9Ow7qpfr+d7FpJ|EQHMAOh}>}Dx1d6uCsoANB>b)L!4;6ho~7r`eX zV|w?xUl{Li$A6WTSJfj@%mRjVcXyn0zMbMM|HwoANqW6O*3lBRC05|SQxp1xGY4m3 z5Ed*;G3q`m%X&@lU>>%n=Vht7)Es_T@GSlLDErJ4Kgcsj$w)aPCf)8D#YDkKDxjX9 zJ=7*-+aFR;DAxBm*qL}>sc6FQf2mQ1rqd}K7^)ANMy!;iD9kpVLUYW6nMA!yoNmwlsmj>B2?r78M9_#wbbO@di8Yah2b@_w^AU;iVoO6vE`qf5y z#d`@br{*=Ue}3Nm0>tcMzz-V0U@x%V<1iYu^4r4a=@qcIk_f2DjO-fG#)KM< z7RezZ%<{PV&1A@p!OS#Qec#?3?QhRNIe21GSy`7bqzL*p8r&prik+;T%M1jjy2=f9 zp+HK^q|)g`11tEKQCsZGc0IRE*2hV=yX@3jBse(<>lwK%wt<#lL+T=nb!TR$^b&3102i=8H{3`IIW4S0!cCu?;Qn0_R`=Y zQ+wfXY7EBo!o%H}==J(y^L6LC^~?I&0iv|U5Sc?M^?;d6q;vM&W(@ef8I9eC15!JP z%4&^zrvQYy;p4Y`7CobicrF#x$Y6h{>4j-cPQr~179Sd%gVzMd9AlRof zQY`Egpi?d)5%hJshvN;e+3Gu%$)u4ak;I;B)zK<*K3Ma>qLZylHQ4L!OHk~g?0vui zxPzAjSc5K_tf#dY--X;`cM$6MYV-bQ^3zvrpt#LyHP+`RPwbpNwn7XMiHQ3b|Irvp zrPW_xGaY}!sl4=n&AvCcxvH}kjdVInT590~{w1ROLO13shIC>no+Zbmt8IMHI78Nb z7{_(t+Sg*lsLQz@b$&7`xO#;TZVyYwWBgQPSAimHxF++SdUBe9WgtJ|>H9ZmQc2pa zVKl%ep6wldu;5O&+xXU_48a_-e!z@ zQwo)~+zo5TkxUx$3&L`liZoXuZ)~-va&E>n+nA>gRmXz^Yu~;A<6J*@iTis+?hBwP z#*BVs*dy{et}%e0-HB`MbSbFc8UdFHjV;1_ql-C_7?*&xbWKl6c_;`11%?IzD&7)G3~oJpR4Y7~uu zY@r(%mV>|VX+A>C)uK1=w2S6rM36k$8SM;`B!8!fK=6;6vi}toGj!f*o;ib3Cj6KW zG}8J@(Vyu2%d}vT{au&e)A+!jzoO@_0{N>WfwK8)1|ST<|MyrrN5p5&93P8SxOpIZ z`Yev16k+v@v3%1wBYF*+$|}21_bBXntQ7SFtGFj|9sTY7v2~;ND}?sGDxY3-KmTN4 z*rDr9I*)_aPxAKdv&zuo#ca6QikZT8Yipz<-1bFM1lttX`8`f}JzLk|HOJSOVDgu* zk?_7c<8t-F`0*C?VHmyoquV@A4mQV?{LmmAC7YwGc9M=aSB@(s{5^t|Vy(9VZJ=j< zye4_Zg#~taA6H=X_&DQCw>uc-=XfQT#mI%EajboAAchC`@uXw0?s|fzmw4;g^twO3 z8*}E|Yskfr$NVS1`3?*3MX)D^%4!TO!XE>6%klcf7ewjbL;lYX{wJCr1m()7ALO+H zBMm4Rc%Iy8Uj_f8AETZ*83dJtL`eFT@WTtp2_qnVcI_G@<<-P3B-Fe{7cU6wFQkBW zG(RBWx^O{w{$u9TUtiKB2mSXW^Miii?d&;pIufBy(DxpM{SmkRPU-(4Dk-cCx|gqu z?5y<2Cym_Ro=14q8K_Qn`1c7#E+Akw1hM<$AFyJBRd;1gi zKYNY|8}I^!&IqNw<4QR{#ly}~@f!6TF3@d#3c>GewoU`+zOviF@hdZkrM)6)$V0V@ zLNa2^2;jS=#73_sHUAzot_0ieWb_J|_6 znLZQ@stJjle#DgKm&a6Ovk;EG2RB(LSU6$JR=}oa+xG_ zZmL+j(wj|wZ7;WnuMRe~c_12bj}WprIoxK{78~4}ON!>USn4F@itgk)(6M_qsj_FW zFvyfqY5_9xZ?NtLbXE8CsYYAuZOTZXRt`-UMjD*46Me-?2GAQyZ}8?S{&YtrDE3~tCx-4jDAdhMNegCv}M!qxl(f%5_5tG8&-Xk>HYf%p4AH1~Khg$f3 zS=on8>s5pcC0LJUvnMcylK6|JdWWh%2oeWJ!H2YxUT#WxB+baM-+nmTklkc#pIF;Dh`pwc{q)Co@n;4ip7S6fYg2pAgy)QREkq3B@ zQw5c6Rxzq4h7J}|HWAsc;1fCRbZpx4K?7oCh?SjcWydxd+l1>@J(dd5$)o5g47*cM z5;xp%b|$;Ns1NslYGBn_VpE!BTcF0S=d{}lSsic6a+BcM=X5weOm0ttKT`iBVCQjh zI=XP&+}It20%(xjhnODkC%~v)T$N9djQTYY@LFbK+&?<(cB0`yK7D5I!j}qQ6d($s zLlYii7zky3*twqj`P_y_xN%wL)KqgTU0Jd->&ojyY_4&V1{xM7qX*JYQO0>p!Vk66 zR;QIr$bE^e7p2r^2(5MHWShW#)yA-S>s2Mg3cXyjjL54*43L9=eDcFbZb0B*t0fDEe@%wCc)Z<$ai&U z>kQ`N373(!q<3UdWGFH)P07u|DqFaF#_P_a5ZzcVUO9?Sg+1o;+=%`Bc$7Q;Dzxly zyR!O8xbX*!wOt()H`R@1|EPMDjEf-6XzjR4r4Tlpu53UatZZk?x*xG|GQSiFs{tcl zdfj%JhTDl&O>yE}FC$k1&aE}x-6ynCzRmWJ=HRgiACymmhd8rjIVf*2_8m1rM$JQ+ zoz+U>B+KSo8QNw?q2zBeRZ8Qh2*kkrLN&+U{<_u#fRv#aPsh7?A{a=Lz>L`oJZ+Fxut?&7h z)ImjQuTL-aq~(*5m4bH0tE;W0D{nA)_MAHyH{R6ZC{Z`Fz{=E1<`NC$wsYGq2@X*@ z5N8%I@7*6ZRWdUv)(XbJiRjBK7Yd$52}PsJlg!mgw2&s+e6F$8Q{~gA#yD^f;B5Sn zv;E5#&;MDgC8JU+p*KF*S#8SijWk62NO4iv-QmdA20j9Z&Xia!;n@k0HPPmZh2&Rl z-y{~Wx=AJ`6sD^C@nRhDycN3X{z}<~vCT$HVU@CZ3+}DCE{@IRk<#vRFIfJ3b+Q!9 zLtf3EM)6Sl<#X-Ogg8@bE``!)hjl%h<3o7$x3rHfhhV~!VJ=`&m z$!gMEYNA8q*3H|vmL0of!=f=oylOT!4;Wo3v8{c1RW4>0kM*wZ__w*y z^6W{?vOD(C@9jVMm_gF;_Bza+htS62Ht}IOLlZSum8swXTQlQYo8rMSMImMp&sB$S zA2c~Rj4EY!de{wmEDddql>H!CjB#qGIul$Z`+yZ~M9}jHsl3`TfPAwe=_wXUcp;Qx zuts@rD$OHrTg>(?#3?hMAz@DhMx|^fJX_o;lF$|AEMep~8Kh)_ITK;cx9NKL(SJ)T z4&9khoO|sJLXi58yrZkRbj2k9Ml#sueV z4N!?~ru|>Wd`scs>Cct;peoj*4+zXRulonmsUT%}ka*Wh z0cfLSGZCU8an+Z5+eWeA;6{{GGcM0z?pv%VE&5E9wYpi!XvxBxE^)Ry5mt3<>Yqp80bcR2 zL#q+CnvKB((SDb_jn&CqpMCg);1*0CLpM=NePw3QJ=^{QzZqbyv` zvsrL=*2h!)BqrxpoL8k-j{?m7jhC{Qg2p_PqS2zA*yCGB8n#$q*NJQw=i(34QI-jv zauz3tm_O!8tX6xR z=7n*|SQ!0;cS|*+4{wsz?ix+#7eAJh*a&yLK|KaCd_QVt%JXO*t&BQg-GP=hhA`g6eYgW z@iIKaU0-A?aK1S2*sVow;jKIo1tzrSY;eKr>!yV7=n8agNKiQ~jIL)UYY`U=n!S73 z_o}q1l&H}aa_@=vWhJBX5v+Rq9_+*Fk_EA_xT30c65%p0Ad(zzj_%ATwG2}74Ct(I zRewQQR&%0YdoTZzZ8AGf5j`dVM^*RB#VW$Wk%&wOX(u}FqXEjaB)-&Y3^w}3;`(Kp zcdX=+@-8;AAI@WNabG3euFU>+Zo^&1mxq(UX;=^26=F~P^4X@2!QuYJxi4$fdy5jSjKlP$e={CRPl?K z?9!wEb-9L!@%m{p+`u_Cg-?u619=nsf*6YYUd0!X3LG^$$3c8C&!!ntp><#%xDi@M z<=NFsi_%Jo_xpVbV8AF5HOmvr_?A!d_BiWsm`e=yOIB_?z z)hu5J^Ifzz&gF+zMH8*1)oEE3LEOjlqI8h7u&YJz3VMhB?E8Lek8PHBas2n!8nK;+ zje;kOlnUpnF@{WnxmgligQ(og);rBRD$`fk1~jJlpc#dul-|TF+~~v+=B>EuO)V_p zi=pd&B4s6w9|VbRVX)6;$14LfpzN>-gJWsXusJO=$Lu)=YKAuf6D226hUEIRlE0jy z=l;vO6!HX~G9z3Fg{8JDdM_rG0AG}Pha6oW%tj`+M_8O|swR76rGOS2ov_ElX0h_Q zQ6T+Ldaipy+>-4$^F)T9)rgVrYIAN|(vnGWtw~iQj~c~5>r56VXFv4|hHsfeRmU0G z4vGbwLQqw2jB&|^w%8uKqqRDGz5XPN%Hvv`H|Y0#*V`JBqONCFOqXm*6J+o9*e$q6 z<-)H@@EnTS%+Kj`*{|(8?$bD5;j-Hl$_m8T-s?IV`{+R=F3vKl>x!@f1C4jd{Loy^ zo+m~j0?;$dDg{#nW3xDt3CzzLo44jpzHR_1^mx7@&l`_VLo0*tw#?X zj;7>0m=)06F3v<}wR|eyKSCJ`V5Ai*sP#mvRhp|&59-t@&<)Qt7w2GBNQk${NQS+J z2t+DO-2K(>_6IvcEFh2OhgU?x2&8{=a+ZxgR)wzZpiit zxnALUckYaH@$&qv$7{xT{!)`~YSl+51QAhcW*xRtF`BNEoTHuEu0AXhD_r}V6Y3SV zGf&$ldMKF(44WS6nhmCqyBigTV%Zg~#vYB8q0x#o*1Nn@GA`kEw9>~|#(QbPy3=0m zFF4NIxY2f1e`_rW!8U_;4Re<;ga>SD151sg^BhG}UW?L}b4EC}?l#O->@O1rDgLvT z-By7DnW1roOig&A5m+~WP@joB@(Z9jv|9Zr$gDIru`Wu7x^S#tK51pZ2Q|nlq2637 zm780;X7VLxbPqqtykIvG-(#bVgQFaF-H~~uE=jGLG z^>S&vpE$u)`mIU{J6FZPX&TiNzMQvlo^k_YWx3LX_qkPQf~Y{1Ah9D>uc}G)CZ{SF zI9U0J$O-F^m*a@@xa!>)oMSP9puV!!+-!gpCvhxdM{#%mdURoaU`OpC9okHUJ)bh9 z#`)Ir+jHdH7uN5t?&7GqyS`_om}{L5KVUa3udQn^O(1C)S}DU3XUm;i>7Vdrr!QOW zOalyt!h|nfI9KXkVUT{y>fjI<+r^$}k#^g)QukL!=(~J%#N-$i@+4NS9Tz*ThvHPV z6fzm{qAp}{%&jDA2ja{=_KWBB*Sr6#^wAXnt`0U<@K5RU1l-V2Hc(iIk(G~} zpI1cl7c+Vr->nNA%=?7B2c=v}LwUnNi5Y>M80?d8EvDYv$O3~=A5|_-xn(kioahSL zlQd`^T!zRlB#5K|TVCa+ILBdn{J@lRj$sjenL}ym5keT?eJxoq!Qwtxjf+3v#~j89 z@_v0cL4sg$XG(E_EX|;%5BP&)tJC2o`9K_~MUsx5oAGL*jao#>HMP;1XekgH8n*C9 zkCRmmtGW%igX2(r|K)+HU~q}QQ?%3cz}Pxfrg&?xlGPmj>gd_Iaz*%^(ZT6OMv12j zbUe?UU7vPh3vLf<6u#`ggAxc6f6ucwdn~}`m-tP}VeF$WRTOEk^4;P-1|hTQxuiE4 z&ZSyrSqFtEak7uL-^-yaaFpmh=U&ldsEuVLc^rJegID!}IP2E1u6?L)Y)_Cj+Ds%!bYXaw zKq5>)P`s0~0KG$9dB8B{!*UczP(Sto*#7QPFvn;NldT>(RQv6r2ZtO~N6sNUi@lpO z#v4zW-@c~8Ptclh6^yc>bYR}u$kvT^^A8jc8=?N@Ng{lfyvOg9zTiTV3@OFm$@Yeu6w^f5TFLZ(bg1in>H7+M?%e9`8F>#f!q?924m;YXkcr`EZ|t55?Q5O-gy) z7J~?E>XDNx)e<;YH3+)|fkGaW&@Ss1rmFltWn3bg)sJ$n5H;sR%bfvAgZQO_d|7ej z7ze7L{>3Sr@B&pmTu3pAqMl~1ZtaJlXpUuqYZzPy$&+Tg8-yJmHdJv>S@e$1PZo`R z?C5<=b$EcI*EeovaAelG(}mx`x65I+%%^Ji;Kfz8#+BhN~0t>7Zk>C_$8< z-Lo*$lV?gj$R;huMqFv1b9?8edLhqp;gYtxKYF;FkQd!z#mYzFe$Dr~;mP+fcV$Az2iMc3cwR$%*f;8){xs`&M36xS{@6Ch zGO7pR1HFU~aF{Klf=Z0mi`WcyBFDwG?0uqrh-EN}FncWK`IT}i>cT|^f1inR__6;R zR#*Fk0ahV9iByw-lD^DYQ&09Wa-zwG~2}P!=mq`soI>?;lYYR<(5+ zIlbpz&u*tfWys&wD;8)4CLNYjL8n)X7x(_~hPnr#_0z0;Y(rr*RuVs#&+s@w4@V46 zO79~lvFDak*wxY7{lyAtfv+LX&Qn`2)sGW<-i>>-avk>5!0_5zQqyA5A3CZ!|M{07YH!h`{KgaLyusQ$gn!v9TD= z3(}yy?nZE$$3m9T9%GLpw!MJT5GR(A=}qT?QM)GHsSh%b4wh+hcH;e-1aKnonFnmD zdx>CGP8z9g)O_8|a8IQ5Eu9xyqpMN;)8{3hE_2nd!5clz??n)cKJ zh=Pf|qv^S8fHkr&#`C@)EtFOv^xo*Lj|VqBNTS}!S-h3Ev>9{5>FT-H*fea`wWZ!8 zD<@9ZLkYU4gz5;Y2PHaNstvv-Gm2k~$;TERADa637F+(H@@3x=i^mVsNW2SE<+uzM zU3Ik2lyRn0&TrBcll$sIW~J1iq5+j^fbr2(qEZ!5O3&HR6R=ei!O?(P#77&^m{oR_ zQF%+^K<}PIqxO5Ba_|B)xV@vgM0!MB(I2b+L3b!_M}`PO0b`&j9vClNr!mCfnEaeQ zLsUvm+9(fU`(ZdStC+0W*vyQ;t1jUi1*uld0Gk?>iaHv^!7w>D3$xpzk0O-c?e9-7 z(6NA^5pVA~5Ush)u{^pdUiZZZ#Evv`Xv?;yE30>smElAq{LyZ=ysIR-JkD?Hp|ZXw z&1ShanE#l{SZF!YalYdtXvE1vg>!t|tYlJ&KWej~ zHa6va!052HD?fKgSTMq1xSCcmXDfvXqMYoxbpk)7=ouHXvwSy9&Yj56SG?L_ZH7w8 zkF;mYt6s@k>B73En+6SX@$Ho()Kf_0IvW_&me=Ey-#2si_W5aE_>l z3mhZ4ptvFGo&09e;dW`&E{oXFw+V(!Y>hPAzg zEF9YC+ZDLn2jQz-fpWr`^_7uU^PcKMMg`(vwZ7*Vi+@?|!$Kb1cpj?@57Ll>Ni$Tk z*C<-eZUCntp2qTRa)p?Vv8e9b&vwMsYqt5|GLlgjH>1vNt@k`;9&u>;9;}5s{9^!no+xYGcLh}RVUg^tm&>B2M&Z{H~wCHw!>$00H}Udk2WQ5Xgm<)S-cH(V57-x!1U7N+AF*++euF_75_-6y{aV3$NMWhXtz{I+jL(Ak~^H!ZT$(dgJfUzPC9S=`OOjt-`2m z-!s*)wNqFi2l{{-Rn(ev1wJ3=ZHg>}-6*kKxCIm8QFB<+>euap#Sfh^I^Ku5fnKTH z#XK2&&?0F1-nBwmqVfAiEH~9mc5S!v`>NqEUlH2l%_OQf#BC2{%!V0+q$}1%Z3*QC z`9AuSUvdKk^Lg>+v^tjC!Q+S1D0^zg|Hca zom*bZJi9gb8ccUha8!nQBJ7>IQaRTe+160mxi8F9TswBISvtQgQDCZ{M(k{FD0!Ph zTRFr&imkel>gb_Gc!g5GelB{~c;Tq6TG8`#f}_ok?q`5|nK_LR;pk z$APJOU8g7HlnKBEJ=hb}gt~#w??3JW=DP-&d+`lV7BZWHkWQhhR zc@9uZhB4gV+#Wi9|C&uH9zJoVjfMpl;x0R$!PiT6;D?^XpnhpZz*p?Q-y%bPON|b^F99_{*tFK zCW10pl=>Vgj_8rM<3WSY*diJ&rWz(;5XfdwDa16$9L>w=YzIcA6ADYl!yH63&2@X| zUNi2+t?{->m@Q9)snap!b51Xv{FTy`GDoxWua#GzShiL^K1kmLA;0f{1Afp>xq<;~ zEjD=Dh2e*Frb4=gdEcS3+c80linlKakDHksY_#ZNznfOdu1R|dd?<;kqz+8n6}*?$ z%G8I$&ahY6iwvUVGb`oC(b-3r!sFrN#Oo$y9hanTYwprp^tZpPC$tuz3Oyh%hnemB z2T{e=T%3>T!Mc$r(>ar9am_$AbxjB~|J+U)EMC30D zv`Vxop+Rd{G~m&XU1FV|tW5W*VLolDN3kBZlGzy$p<*tzYRV||xaZ5AIgoGJtIDw% zvRgkEIum6gC+Khp+OOy{Cncab7c|c64fH;GkYHAEWm5PBVmeuL6$gq%vz-QA#@3#% z6;=zXG^6iMcN5XSsU~@^;X{slCYlZBvY47|tjAgNK zWi+AQe+4n=&;6e?s)^zK99Qk;L$%xTIp+$S#2fTUr}_-8Lp&FnVlfk(e1R5K{G{Dx zF%;6d6$CToovmBQ5|>ByTnVn^F#W7iH|x=vVC1LBO&lFa@N24+ZVMr zmc4rd{g(AcLyObo(HIyr$~FQ!>uFixfuW>>Ioi2CPT1p128#Ot;^LQ0uUyn8-zEm8 zjWet@l8MsgKkiS15yyU(gZC0q4KwjUtDdHuJx$_4&;sN&2s2`kpK4x03b0ZCp(H;G z#Ce9r4AQvdl^>+HJ9pl~?^0Atn85@gANp_0_RzaZM#TcG`^_>`*uJmNw7H4WmBtPf zP$i&@)97svSx0Aoz3V6v9OhWLH@GmC$JmBSSGg&YyRfEDK$Wr6^->6H@wxGUR@eyD z_x<2ur+t;Nx^QE0dDpiqPr?3^S6b$kq(~<@>7~@PbOPgIaGJu+IqO|!s-fc}>#@n3 zz7i{e7~|M-B5BE-8pX`&1_DiDh3J$c z?+P&owPpyb-#hnn4=Bc3a2*-rLTq1r8&RK22&SZi5(H;WcBb`lj_vAPdMcL~w=X&R ztq|rE?hSXZ90nUj@WBP&vBD0HesJpEdVSbM;a(-{NKK`1o6-n0+Ol{BuC?26m`QZ@ z5_5}utDc+!@JWoU-xy?-b4ONo9`0}*99j*%aoiqcY-V4ePH1YWKpD1IW1Wquh8nC9 zZ~>j%vdnQu63ZnkR@RAoXLdq&A^D1WN-d7jM%&}2wK}Cq2cnFl?(%)#xQuGMf-;pI zb3jmuPiK+oTtGy&e*<Jy47K9S*$jo|`ko{pZTVI4au+pk5x}Nqab~I$xI;)Pa@^5yv z&{umi7L>L~3(yn}*lwrPS37L9D3z!Ludv=`&|LMSUvxfx=f1@sV!vEmHEiC(r0^>K z9!Ibiah<>r{ObF<{mQXw|6QdQF1!4`L|lUes=HngQ@_c^COYB9pwYPN_ny2Nuxq_F z9csU96<*t_=Cmujd47fN#>RklI8!srN}0dgb44rtBo(W7a?`fV#bx!x`)<{kTt!;f zXD(cRm<{;(yhO#v)07*4lWPc^JZQNjeBumz^#DKOo>teldvKD5!^q^DSGtHAtbD^R zGf%p-iHddW+|8U^M z{DWAs7eJ842IW*ufeT4Pfct;ivO16}&Wse~E2Yhaq_%dOg@;oY=7y&7-O657I=165 z%ozx3_&UBMOys!Z(w?DIc|#L+B3|k=Pu|YqIZlU*vLT< zUnB_zQ!I|S6^Yyjz>)z`T=TYt9~mrN>vQEp#&?dU{3Ab`l1E?}7p+alf6;i)PQ6&t z)H=YFnUHehKIY2+6C+01KrxLm8>+;fK4tP0t@mu&m=c_OlC?`ViJ#a&9#gw8B2+k6 zGspa^wLXXt8shhBPbZ(K0?f8S^bJXVCYKv&e%&V1?&$bL%!3vi2n$h)gVQlF;MvWjUVtiRL>hpJ^1inFdnFj57rybVu^wzb-6dRj;(8*SC-sowRJdH$_- zx9}@oavWd-PTox3Kb2bGBIGsj-LYN`|I{jx2(3cBAt|G02_4I-2(!@XuUTPzD7{DB zou&4Dvs(uCV+7u_ATEPS@A-SWBr3ou_RC=;O0OknfB}uTdY+uHQ8I#!xVrO^Y=ua( zaajZ-<>dC>I6+U$7{C|a5%!rAU5CXB=yJwi_wuixCnFLe&(_nN8o7fYNvXS6#^~fn zE_Q&my7^VYe|_nzI7nmRFrY!Q0iG%tc}(r>nZHx|OC_M#|1#cpo73g*TKX=o{)&em z9_3#l{#`%(6%RjbmA~TQhhg$pJp6w{Jj5`f2y-83z6A9^kDi--xX53aseSbNm;CK`8dGyQ6-qtXInbM*={=ZUq+8+C!*+;6h$j+&bk z*eG?I4(DiQ*znz`Nu@)YpUoF4cDGAcVQm$c8C&T2Ln`|sLi5w^AkEe9qyMKliU8I9 zXeFH9V&^Trf$y^V7uy;Ao#G28mhi*G&-JdASE~iRujaN5-fQL_qW1B8(fzB5d%p-6 z*k8!|SCM{&?Ty)HKIj9DdV%A;?Tt5G+sw9hX`608NcUAFCNJaZ@aUtBmYazyy-aC| z`TQrA`i}&7TwqmezRds~i@%S@8;52wO(M)`&0j%)S8-?@AS;Y_D8TIW{c(HDxr81e z?V~lZkA2FyEXl9ce{m>zV|IXRP#djr4_WRs`6MP6MYvI(bdITyK)p4J%Sf2Vm{zrtUzCwWFC7BaNorx9M$uB?A`lyW(dV4IPl*Cb)uiN9cw*#vzPFeRlP_ z=VCfn3At#g^}-;jU}p((dnLrm<B1T7y(_$YPTph@4ae zK1W|yoH=s%f39GG4oqz#yN!@FTY~$Z&MglouJR>AZmr$22uT;?3tm_dPpNuu!r*kokymkM8tJFVR77n)mAOQTy^t^GrR$oXzY74p`6TdUtzEdBCm55= z2i^8o!Tps6r;kD80T+0K?A6qB_fE!%ApwkYF;uAf1YRV~k@&0vEQ)2NkM4Z}nI8{` zARM%K5+S|3KVL8d&&|;5PZ^4k5_sYt%P7fDR@pDWe81BL5X8{(hVu0Lx>=Ocz z)e(Qbuyq|2^7xj@A?sz(5w5;pmvQ^)Cx5$W_e9LNLvxN2IpbV3P;j%2+wJ_{adf`~ ze7{4de+M};Hc;Wv(4jet`YY;@55c$oEn%(aq;~4h30%+zK^JK%;jPmLVp1SD;q@5x z^v3TOE+q)>J5X)nWMVFL03yBgU1r7qK7&~YK+{N**3A$kJLw@}Bum|Mtp9xq58r|* zyNczhBja}K!2gGOD72&e_YaV?7l0Rrn3qrC?0ycGd0kBEzY&cIY?7rsNS;4M+aS21 zH&jRXFCXwj0c^zC{F)p21-!@L1AJtI^hmP*Mr7WYE`TD7!IUo3lL2{Wz^2TcgL&dl zPWE%4IT02p_M1PKXIu_!BL=Zs47IszTL{MG+Q*{18yY>MDQTMOT^B@`fCfK63 zh;8VU{qIQC6#_VQkFat@2Bf+XMP=N}ia*K@{}ZewQ0=If6<~Jy7xB{@gHmb zK8TA^^t}}O7o4sms$?tKI*^;Gyy{>|s2us1kO8ef#=i(qO7ZnE-X-LW89^Az!d;&aO4OAKx|QT8see-!O16@0SidM__w+GVkg*li2&4?wD`y3ca4f1 z?ZRa-P-eD9dAjU0K^Pi}oA3Vf2d9q^7re4ycCw#rPb<}8@Bn)CyE%Wi$xcasAqTOo z+p36MuP*L@ne}$$&-@E^CY1rUXKuzLtA$)o!yUJXGQ)C(y;-cCll`(L{Fd7r@vjqP=cAJ_H#Z) zl1LI!kj11Qg%mZvAqvcI0D!KklR=6Etr{$YhX+(=knjJY%gulV58b%Z$%Xv=2f$fg z55wz_=G-51|Cjo}+asru?LS_)2~bmpdP2enk4$D-|2ibPOPX>@9Lfrdj z^Mc>M;Nl6!Q@x0s-ARsS@98@Ig8`EfQiPHj^K_X%y9YSSCrsxz9{5SR|EUYYAFsJY z0OC(NyteQ?`K}}cfPc3o@zd{6BUzRAIv|!8Yg!$W+rFj%a)^^ZDg0w#70_W`kfO95j8Hc`V|&FpDAkUzc<0;Kx5>?slds^C*pfQ$d%s^H(2)agq9 zt9JjY-BV_G^;hlwRlBF$@K^2rRl7fx#$V6(f9LuB#Ne-HSM8n(+W)3@L5GMP zo8cnUvBP5wVdYz(eW5%2EYhL))m}x|_Vb{loAe59J2HJW#^r({8q{HRZASeO7Uf(y zZ&rK0Xz-;@`e*;$gCO~j9TQZlc12J!ct_2nL!-{!b~=Jdc`{e9J2W*S?VX>;hnewY7oc88g|wu zs=#q6RWoXiH=hW=H~`!0NKY=5Qx z&Dgz4G?-H#7MW*ZvFCYGl~N0eTdz8P6+b^&?~DrpB6I@I-9{FVNIrAv^n$Eeccnrm zszGLdT%d5ByNRG9@kNm*d;Rb^&VGl*0~|} zwbh|(>di&E`{7qQ=1Fgo1@wG7tMMH0+nQs&q7U2T_&C0GM7;}XR4r6{9Ewja-I^9Wh2i}Pd8Dz6=OKDq*0+_ zZ~5;@`9b!)&XIPAK%}xkLmCfJN*ba6!`@$qMYVkoz_{1OK(EqL%1B5_NT*(fQA#DG zm2RXPE-E4*;Q-R1(kUSwN{vHGHz+l9H^a=kkJlOF1JCdG{`o%7_m9gxo-=2kz1LcM z?bQqP?@?w=UBjf{tyO=Zm`WZDxHK_9Ak5-(M(i@@9Ig_F)EFa@{)M}WT~rIK@%}@t zCvb&H$cYlrp+s;POOzf~dk!D*X4H)0AbM5P4K`-*Jkr7S4;uZk=|DWL8WcKv{Vok* zvtRq0`HXi69Wt~~Bpd$p3Vuuc56=F~KnyzO*>#;EQ?ucQ_8P&Q^aoBz{;=kFJ|Q|D zNMd&V^gavxa|i4P21uFb1ADkvz;qw;_OI&?asfaakbgx?K;!=`j_~tbV9i6-sj?El zv+srH0JOW@eQoz8NCMXAgCK)9U$ff2z$JYD_!tOVhHczWc)IV0f8x>%@WNcHuKhg2 zKff61di)Lo7+E6`RAPW{c$m5iryueq9@F0-wXN+010N%ANH}y z|GL`GQ2Nt+vh=aNx%Y20)(gm4Z!`n+A^HC`U5RX`cCH+a0;!TToyWnw|~F5|e2dq2VZ z8&>0U0b%IVIS@wWCNZLD z!>xe&_+ycOzDHQcZLspoL_hK@z$_z9=AU|1V}*EwyYROd&g4N~n5v8InWgLqgXoK( z+U(`>c|5r;;(AM09MkJ|&wL#YQDo+Bi7>o7f9-tM)(8~`N4LgXXHu_++KUK8|)@sJX|1WR``M6<$1lkvM($ z#C2E%+GBePy*Vrur|)v^QtDqRQUo~vej&jQ4Er7D5-bR}(>R0ZU=d<&&aSJyDodbM z_08OvChj|I_jUvF_hphMO5pfb7PKP-2wY|L#GHlS!&=q1aADuI}TlXhTv|jZ)BZj@?j@b%(V2`@_Zp}6Gz&Gk=enm^jENPPky?s9Iwf0DqDL&Ctfotl!f@h56X zzJz2DEH6_w_Y$g<&OMeUj}(5}dHPhq-UCqifcF_9jNxrN1Zvmd{2GV3MDs48I3q~mLwT3r?5s2UeDin*aM34B*vAscD_qu-R~i8i zmWhicLSPU?2vypan`Qt(!~E}pVHbv(77!cb{2c#vKRJ)A4Z6#wK8 zlqs|TK(C4Xr6vD^n;6M}8~(-xZmfU&@`u363pv`J3z*j7gJ+E5AynCpoWchQ{wdi) zBF?)A6FPj@S&9c3VLR>+!+sF`?rTdN#k;ZpmB<8oRL7O%X2Kr*{X!xW2wWXPsl){K zZva9rxOpho2sruuBAyqpp4+;#FA2pznZXUJTHYP$y}@Mk5n7|;@~Id+Ievj&=zB@I z&lce218c0~8?6ZY@q$7i1m_nEoxjfn&%!0*slfA2i8exQ@OXQOJtFVoV~hX9#zQ%X zSFot={^S)QV;1cW$5niy;*@oH6U!`O`AzVL4ngy zxWXryP*1`q@k4#bhXex@mz8jSArFs-wTpn2&`l5F_tKvTPlEzl-B{W2zd!7r3~}?r ztNX+Qo(-=CP_m(Tut9(#4}g-v1LnP#aKQ5(RPni1m5UI}5&=jlWU6MHZEr*~L&9io z#BDmfG|KD-ZnVWQ5CWKgTk-+VI$(xrD?$%<O@sGIS|hg z|5yPx1fpvA8vF-LJ?{ZDvUEmpaqJBu`b7|t=%f8JE+PgxK`B3w?}q~_==qCZA_cBN zfe}9Y$AJt&FT{M{#}7yaiT&?zMEG_XzwiD7p639evMdpC69V-G!2lucX4?di{SGT; zDF7ntXr*Dide3|p+&C9QQN5q{zY+|%$wB~J2NWz2bJykX-^V48cgid9-2d0Ort{zf zk^4umbwI%pgQ_n0!sPL-fTQtl$Km^O5Qyc+1+x`2;W+wgd~`$zsALKkY)Z1+2g%kD zl98%wvUnt)gOJ?n!?n*opkIuFi0uCZNPfLg_>xXsjH6<8*>P1ou$H!Tfy|rMRe`PO zVbt>OLv4aY_UlIU6anxPIdv0H;huu5;J`3+ThkIgxHIRWAhNcaXKW}%)5-2kSYZ})?;0Wd_}qDBG(D;)Oxx#%kVBKNeE3x^7f^n^U}@plyg#40T3C+^ zMRVg}Mw#%lGnVg-Kh5cfsS06NK0iHT$MFYI|MwMzEWjJah4L!!<~D^V2gj*lf=-jE zR-iJSih}rV{-dRCQHN}W;Rmc1m7`gX+YyxMZW$HR^;V0L$C|AIdYu-UOXq1pOH*f| zq(d9SoN`pW^ySI~TK)AxLJ*K$$jakxOoY zeu=NutmRJ2RuZ`PuNTZL04In@;-`2L4k?r#rpfXUjJ_koxV?}1dBxr!Tm~S#EPnGSUb|`}#iovxV>idt znsQ9R3s7-YsJxjJDkbZ<)u}QH3T^rG4?D9m%3*|*RrM1en0G4i(5BZ>rDs#ItL1;b z))H&FJ>-quee4>&RWOh{>m2E~ ztn@=8U-~D7{JL1Kp<1JhzOT zK9dM1H4A0UBD@s(8&&QhS11HXB3$N6#Y^aVdB!#P&Oa%d2~dCM=6$-YBw_KRn**|* zHS>W&n1@y0lp)zOdA?z@U<#}G-kkne7bFhLFIH)+He`&P8y)+^RQu#)0el8=-F!WC z#<8l#r*&cTJOh!cqMhw@zSAjjeIvG#$xFt0pvD5<`tFY8l%0~7->rZbbA|nk=G;J` zp|C?u3e3!|w3XDKfImRz-nmtaxYOKoJtb*jcV-w%+k$hD*^Z@FedKY8)iA$`+p?#L&k{2p)x(*v@I4&6*sVTNEhhZnb)h&RUv$ z7nsf>y_o!dgRpqCbg{rusr z*}RsMLp*tYUSuX0a7FXeyaM;zC8Tou8nkw1)z9XK_X-DZkCm%v*uyi*yCQoHuDMy- zJ%Lqr!!~GYB@w&pgTnKrxKH6%AL(Sl6Vdmd69-hz<*9rs2MgrHx_`ac8Jj%DZ@(7$ zPItZ-%`u7GdySC1YSmJijgOk|NR=o)0%vsJwX?-k*ByBdm2%3+YwojJVK2NH=f2zS zg6e|Fi(;jDjoVti!irG=mHiu zRchm<=M^MuQ7Rb8Zy7uIc0Ai+a#u#zfX*az{#z4KoaWA7KgEpZAme1R zgRgrxl1bADXIy|P>|xM3OjjB^Q%<%7&K2z5|HQNBnP4;k_?f>8)=d=8f8>tTvrFHj zwv|?{ah#t*^04|el4`rHDPbN}H=S_xJ4Cm|Z%Brnv@y%>F;1S-jYtd&09~W$<*0cD zVjY*KFjDUvKLq#9&VWLxe!)l2usdp>>!J|SQ?Q-vxu}k-I>}iV6zr{ZJZe?^jis&F zSo{;iW2Zz|vP!Xa`5e7f=rq<_$=$_Ot5fZx^8Is)?=$7k`4;JQ9LZ7WFX%OZSXc7} zE^e$7QnMJDiw0$qc_5DPq-45+-^oZ>64A$-$|j9?_KzH$7#m;T!|L$LHCL{Qt(?}p z4J%S*s^W>X2Ay_U_tV_6k-RKjstbWfhFv@E4(6&ak5%2N#4O)2@CvI$`f>W->9^~1 zZ7&*GzSC|>2vYuybps*p^e;jjL_NNTj(JntN<<5M-R!Wb+&y!pCvqbG5SI4o(dGw< zF!OL0MF6^(-b?7ujmm;sQs*c@HA422%CDJ+hLZSp(%2>!zwF|CNmVp1dOV1jvm6R! z$K|_8YMv8t`dU$Xaysq8r1bSOVOsV#VV;OF>8i6RWe#(4jrZ}D3zayrF0G9-N9?xJ zBgZThH_G8VBTnqZ(;Z20eK$>yxzL=}=NHDF$BN=MwzzuFO1tZ0*ZC~Vcje3qtDCn9 zM~6yrl#WR^z7|>toOqnP^vhUH@C46{iUKN{hHh z7Tj5^poPe4#=J1OdnTjYV8CjJeag;!O{tfkw!hTSR{X8$b{3}Ua+nrbGO;*M`-3!7H}aS>DnBuDJ*uk7?|fm?;v0i2({+v^F+i6qfsNCjARv8DBy5ZB zE+%G>Sez|Dcs<-K7$uH5Yns{4Rq15#0pYLH^W{^{EO!G*N9nJp1;Ik{Ur^^%5y9)H zkbZES1XirKI&L*>Y?vU{8}zxy@SSRVlCfs)b~R&^WvCXD7wfG+U%8k9=9D`ezSVRM zoWmTn?u$cX%3iuPTjX_u=$N>!?1H{45uHr)(%Giz+346E%h~dUzlocl;U^!^V&Jm89mktxeuIhU_X<7IiUeY4Wbw2bO|H7TU+i5T54mr+%9TzJ zm)$7=oloBG+^x)VeC2}kpmlnf#Hs;01~CfcsXlrJX1h!LwZ?ZPFrUB!rOUx6ZZfoOvRavz8v!R5(E9w!OwlfxZ)=q@q@6p{Wp7y19gm1Fg5Z8#E1C zl22VY=h45DZ9F)eA-(*ik}e>&e7ljG;*~vC{gK>Jh+O%Au?<}rLKJ*T#8^6!HJC)$ zZt}gBhWk{CQG3$z7MsOFB!Mu0UU=3414)~}M6U|oqH_Csyv7rlVXT{$*PNokv7+4g z8!SKub{}gM$Zz=jNSysfOkpi3a;OESn8NnZ29|~*N)M*LWo7d-$o~j>gj%8hmi8*j*iHuv+*)~wunL1*wv_RHGSQ_!v zac0h8B&&462T1LVbpdV{RW!eqE#10MW7n;r`bKN-6IlhE$V<)ToP-m(SVHhHJV`!| zHp;9*2{cASUt6E)FkUIrS*<C~=Cjr)z#$!|iHp9EEyPso=8)mma z@%$IWGauw<0Mi7bx$sDj*lrsdQPsq+MLy#&tDwMd-j{DQ{OIWur8`$H=-LfO*=i|) z-kP!#ia{_sV@RH^^5d-a=N~+WD;v$isP|ecr8g=XW$a}U^UQ^e20JY4o_LM8?Qdpp zdXx>$O>7<8k3EtiWiNiozVA`#lM;*{PqPpwK*B3bnj^NCN#~6h+3lIzT+;V?m9N2% zGxicE!h;#7dV;TW7kbmU>%sb4$LO-lR$Uu~7s`y4tm1XI#`|ll&kRKkwU9cm&W5Vq z)xEVFS1@#iqufysY%yB?`SBjRjqT2a zGg-x#_=b$CHO?FEN%c@#8w2jIP1he!B{WZj+z8V9@-=vL+BA#vA;gAS)ASpl1@R9# z^4M>#XfLPhOa@8j3|H(V*dKrvSjP^3Qhh@EJlAm4n^jJC06audHlD@P@7x!eh`aRaJL zX-MTowrnNc^d%AC)SpU8qla5p^2wU8N?Q-Tcx5&J;w0jmejua_Ta=DJfAapG_2?Ft z?kEtuHca0QV=GqT3TygIGgTGPd!n)qDDQ2K=%Fxz4NfEe%`vz;b_+eZ=9;BK$` zj``@@lg&|y<7Da<=jmKG@AhTg@Ln6~iZVES@Afr-NnbjY6pZzm*()Lk7;CV|dQI@s z$3tflhhMWQ7LYcDg614g<;5;{!(8N)*?M(1GX3+T7rB=>h1a^ugcKYVY{Q+1`dTA-q_R3;iFLHDrYut}vE`=$ionM~=6n(GUQ=ygcAW8O zk2%4+N{^}z6ru9A7kvA1_6HkIjnB&UYi|({<|rw8eA6EjDgf4vaBb*C$IzkUJRa%C zU28AN@}{BM-KJLmhP7$g&A!PoN4(!^Sk43;|Y61>rLMH-Qx z${q}w9l&(^bhF)bXM4yIIv)bv-JGSmtA$ib%b)+M!R#&M99-#Il{Q_A^O7fRwDGxC` zrP9}7$aOKs$ZO_m#=-Rk&r#PJkH8T#c>T~Jz0C%9e!DJ3M#8Y_na|X2{Sz3fX}Y3N z2%J9Lsi316sIE+lq!XtH9TD_Df-l%FE?hO%-i&JYL{L;K(YO zC@ET~IJTVLZk;>W1N=-%eVvg%<4sw)^e}MYT?#{O-jA5G2cg;TNC+E-i z8segeax{Y69MmFMGRoR~Tlw=MZ{?E8Efq*Ia0>9-54e|NCVH*yhD(RB$7=7(i${qK z&xF%jDQ;AS8d>vuWTPx19=G~M1l^sNe!>-VMv&P-_3?b$2N%!oZ<@kV!=ujR`Ecxo zQY;;R-D&seBCZ$i7?10Gg4aAyM5#Qk9Iy+It=$za0}2}FM%FLE{J*Psic(V%k|Yo6k|lQ?$seu4A+giVVrX)x^Y z@_gw+ZAa>T-KAP`Kd-)m%#8HjwDe||^!`WYQ9!&_`fx{Cc58Apa5PU|Jg_*jnKV74 z%=&|JOKO@tAr<&nT4jd(RLgO?+62oHC1jzBIB^JXq5735Ah|je(1l|CTYQUYiHC9^ z2&Iuo$qj$J6Bhd4g=Zhs{8jXhDBjY2yzx4yxIiXhf#+m@zTgbiD?k%15;nUoRB}Nk zZ%&=yL;3BM_=^y!)~NCr;JL#g$Vn4_-mvG_rQ)|GU?)ZZBUDX_JO33?0pV1;kaG=3 zO|hTD;lDIbnx7#2A7A(%)>9Lx^hy`p4i8rPt5Dyq2viDQiue<(iI5wKYULpbe!Ja* z#3tsXUa7s|Dsc!Z;_xxpa1#H*bKnnb`(n`OxDS+afPJ60fO4Z8|AI5Hm#0d(X5-B% ze@>|r^0ggBFcN_H0m^L^4shO&G;BaJtezZd8t)I3hzC_T;YZGq5=Qv_B0d4&;kJp; zDSXnq!VqK%7Okw<$B9f;0miG2yP=JTw$WAa!fAuFQ+O9T{=pAqQ$gvk3^9&zct?NA z8Gwdug$=pAqzgb*3Xm&oTHTo7&07=(M~^#GJaTWS{`-*{;31;=V+jGV;Q(>m)z|R} z$A8S|_#S}+fK+6dx(o>$kEgnwq4=_y>DLA9UTq7?J0r={jY>M*j*h- zNk1*mX(Y^H^eEv4|G-CvFJEZ8$EtmE=~lO#n{=wV*qU$Leq!UN4n~;O_Y2Q^kU+Xd z$t{Lo3ndW{-^N0_hhFkD7c(d6E5qOcq zLHg(X{?jy!m?1*YPkI^rL-SBLdT{(KOmMqMlIzh(cWbVa9|hlgdA5?zRlO1LCzlPQ zg|BEit@+Qc`RoP+D5+@WoHA%)`hB)|@Jjp-|5;=E^5rRPST>Sm(0RS(%DPDIos78N zqO7i1d@X=FM%*Tl)+Yz%ip)BR`z76PJ~8vHlxf- zWp?#j>B{UjN~F!h?))t3-AQX0()r7z_meQx_mgG)ozy?1t-33PDy_vd;%c%heU6YB zotMq*x<~pBU-1Ax+5eWG!VhS``Q3&N9?qlZjV2*g)1Cd6roF8FNVBM&LcUq=+bS4s zuc?IV%A7;ax>~x_ux>!gV-Ew;`^j>CQ5K~P>qMyHySSCv8^0^}2Z5i}Rho9YHHa<1Ssc zGQYDlIVxZ400N%|nxbk+$zx*HBMvLfgLWj2f_ZQ2fpg(?ruR}7!OeFJ-N zK@(|Oze6s~7tjxtmd;~UwdArw{}J*R;tTKX%u3hDisz&E-k0%*A=dKn+$nr;?w1FL zshorKlvhH`EN!H_wR+#z1oT_d>|pJ>RYeg);)Y|B6g>7C&N15I$v0O7M#s*-GwyA2WQfNj@ z>_Qa*|Zbi-|#|Uh^;zY;~neL$|ZXAias_QDE&m)jWe&8H@<( z;`dVb zESj>c+o0_wmf(@E`S@12zOow=>l;}5YZz+=O!Bp^v#tLTp5f38ck4k}+)JDoPKP@~ z!2I)4&g#88dJWjU-O--g;0RCE_OQgWRo*L}*2JLMSmhFz(PXxIEs!!Z1qzUlIwSj8 z?z;!vD5Z%qFYT@Kc#-8juf3>|8`G?c4yf;1`@S`aoM2M!K`l0V|U=raWTvxGrJ6d!EVK(T#GfO6HmZ$$|S4G={RZeu}iP@k<>HJxj z+1wk>9?EW&Yt1fPV;wembMmv@z|!k<>%GBrN2=Az+(u1HrHqd(ALa(%0Kz`wPu~B7 zzpOzuyDe9>@MbI{*Ms9!u&t2>qE?5m??Dem9`6_mbF~)jOd@x%Ge^hTn;Qy5akFoF zNripsX02vFhHtL%FQI9$Bs;Ukch>FQgA%kC=5v${ZX)UfG$P?7gR zwqbB%R~WSa37Cz}-sm@Bs1^U0T+2~V{X=~{Yu!h8DZ_>>NrYds=;etxkBSjwZa1L9 z!fN6d{`?OruSaP}>wQpC_WjGvMX@MOQNv}_+-mqxpz`sk#}pInkyAz_(e1`&tJcZ%5mYi#apsktOXNvzK-5x_L1Sh2b+KC$nfZZ~j{*)Ggj>&H$Vb7oT=xC!fvuk3020Ij?YA+8>KfZ!bBWrQr8GBHBepAH?Ks?4SZb^5yg_U3O}__DiQ{?Zz%s$T z8IlKVqevuhhMr@hcF{{*z#*F)>MG}5m?w-|MK~@v8T2Iq6;fjqrog>kt1{BrJ^TV) zS4JX^UA#Z9e2Vr9tFlIg=03p#AkGpLnY!n(K185&+k@lhU=gvZs-yXNoG-hzJXnRj zrD&`ko4v<6UyE!B({^KcZn%9r;BG|d&50;f{iWKNZqU_iJIW%^W!?j(Rv0VJA!7Jo z$uzecRak+bL{CSA-q@ydQ0z0Xe*&Aj5pjK4*l{EkJu=de-Joolw`O4VwZfyx@-2P_ zKed4Gl(7btppakG!dD_QdI?QX5I4d)(u>ymEcfkiP2v}_d>WN??OVHJbR&zjySXiv zmFCHtJ42-^lQYeLAD?G*kJ0qkYxl>7murJQ=6ZHPf0Jvyp37_av{OeGW$|DU>wx1I zJ9=7FSRFJF>95r)T?iS9QwkdMvdT6n`0B(oVcN-UKV%gJ>4V#5_dc7ax93y~S;=fP zAKB2@M=03^Vqr&MX(3)cph|?yl38WzNsg1Rm~vMNo0Ow&CO7e=*>a#wgj7IdLPY2g zYnI55gRVe`NW-jn>sHR>rIX0+VRv%=q2-pwX_^+p1v^o_${w@e)4Ai2#Y4S$bxNEK zI3Q&UhMSM=vUh1)#z5U1M#sB{8Z)S@W$iCS7Dw^>wI8LU^J6cOF3?==&UgY`k~^*% zj;PMwb#?s;@&BSczd`?9x zyEJE&oBi{lPcDj&9U_0s!fs&Ox-2@96z5~s<;JEWl?ibvS zoo+aK%X)!TEUu$cB_scO>62;T#6&D@ObZK?M&6$Pq<1m~1^R+jnx`g`1-@aT4LhAR z)+eyfA-xL)amAJM;~!JSKn9+&^}2qmdP=m4%TU?{7+eSg%+GEwvuIt~a+-w=ku9CcvdOx}D1OEUS&S^OU z65ovJP!5;-Co$}Kx8dg^t@fcPV@T1c6QRJ{FyikX2Szryrfl07o%;*WGJygT0znVn zp%Q2F?O3Aol>x8=h)RE1=M@JMMt$W=@NF%Ge-JE8n>U;z<+FYyUe%B-kO! zJ^1#g=Qj=F5~56*zyFTFz$M%uFo%_G;K!qS18H^9wsHGx>eEnBCP~?_PlOksl8M!^ z;_vX%#P1OBT!z#VoZ!0Gk8{ow0TsK8NWF^Jxd?OlaiIX|m5FRu2*T&YGZ66@tA0ALSZ52+ z7q8CqDu5S`3RykJ>%M=mK7T0EwN>WB8GH%Cj_ScR|l%6uk-0d4Me_BH|_9vh(Lk6rcd%M6NTp@R=$`B;ba>PNW-* zFpJ+VcppM*6mfcv_ktN2L3*Hh-AjTE;&)6?fsjAAZj{w9p5doJagxq88z%c11fcd* zCiKSmSXp`nVeo|yqbcP3yl^!cAa)`o!<|4IR{*^$(%;3e=?^p-0WU^y^~jI=o=1UQrtM);y#(Qtzr7j%3bK#2`E2y!U)Y8$ zJ7??`@P_F>5#9;*Luy6nYr=;sK?9jFhl9@h#^yf)P!c?q;erQ>CnbOq4_nA@6Zwe> z2^KJ=kdX`u{CzOciP7mb-V+b(x89~p`xM$*ibbs zKNtBsb%#Ja$hakt(Eg_$!h1$iJMvHH{|9vms6chdw}fpNUWZSGZcH~)>}Lbn1o=5N zoOp#-0>nOmWbv1Me!oM8Fe?dv@Z6;4mvUc`5M%yl-(qg@^U#3JeNp%U{h=l^t+>Pc zRde%!3{{(lDe*}T@zG#_dSgB}0>}uH`F;`d4uD9oIn5HU6NaRK8$NO`AMI@#JK#AC z2E>ry=L1aZU_#9)<@=Sd>yQS|Z~a=F5Xyv+-MamXB*Xfb$+oGNubgz?ciPsFr|F(D9*aEyMx zumPVk8DTk zL0C$#lC;6mP&2o7Y}H9adv&~>^yjYl?^EHUKoGHs?ja<29kzzNAzRLk;f|>;8w1Im z)5W`=U*$YR27Kr<)U;d2^jKMSm!GMtI$7ussc5g&be2k-*B0Dc~kg2sa+4&5dKmnucF8Zhu=V!egrT#(kuF_*S5%>%o6{>F=X^$foq5SO@IPjaPV#J#0P^0{Rc~ zfi{H^j&9~QJd0yIBjrnUN!3Dj(v=SkQZKj9q@8FH;pgKh+N^)VsG?~t#|wg);#+|AI}0r>^aR6s!WKii# z)>|>8HZ7tK&Z5FzdaQBPF0^#_-M8&#B9m;INW%=b8?QvGVzKzSN)gccli{+6gR8Vqt_ASdAj*wi<9JfB@tR=@mO z%fv^6{0Mp~!tmVAK+ea5=DF(Wu(bv3!Y)NvG5^*MDrdIsP95l$OL2#-;z8DwOp7>X zQC9?4RkVrmBU&1C7_P&tw8T+L^-mTJ2}LZshKlhUw{0v9ZJud!K{=9}lsyE4LQSmjmY1_;QdX3c)@3c_xdiy4ogBiw98 zu-j3V$Q?R&>S0$N7Bck&a|1b*fXiXwYy3={jyr?6TF7?{8cJWev~pidK)2F$!XX+{ z5XozN@9J}mp9`HMEUR>BJ>O)w9Cr?6C@Ae|UXFhQjwIfY2yVP9UxM;D4EHf=Cm6%9 z$XO}lzah3QvZHUB*_M@1`9AHTw8dA{6^5;JZ^d;P_ZKjytX#fgtTlR_L8|Tvjqn5q zdU6Ypfjed?dD2=h&eoMR>%GRQ`C;Mm3Zqq}rLlJ1wwX7F0>SKDi$Lr%4-ws}P}&_3 z71-9BrpFqkcNw8dbK}^WkA0;mPkI5$aw*87d^9XBIY~AfG{OcpzO$(B=o3v~- z&huZ6N**V#E2lDyaj=5Cj;cXpo*p6nWc5*4#S6>(d87@)5TLW zVQMBz6k=n~R)tiUi4ic#7aE%|BKv6W6zweIPg{@+Ld9MEGK`3G?3mD%lL!s#TDzs5 z^DShncYANPbrVH|-Ur@2uT^`i6dj+-Wd4@i-w>Ek^|rc7Qfw?yq*?N&AlG)+^xCe!Vo=Z~y^J$G zMq@AYNkrTE(^PA&zF&RG-903(cs-z{-nKV6h0o7ca6fojQ2JgByOA~U89#&-hjpL> zrZfIQn5t)i;!qoxKRvljrp+t1JL|_A|EjdGd)^}e2o-##*XmUn+%*T^Igb14je94_ zMj6me+4p!xWZoyM7u6aURNWn&p0Skehdc>Zlb#-TK`s0fhg@x^7nj2b-yNVICTGHFw0f0offv%8j6x#lCuG z)+|0cW_vSeMB%1;|4njr0jy;oPHB?Cs#33J(M*A<+r2|y1$kaxG_JHU77dxf>&>`U zhBh}?58sEJ+X5918~JWlL9p^A^@~o1xg!nw3@wdWa=e&mzqa?XZLqD0P8IFT6H$dq zWzoG82FnR3V{%kI47kbPs0~N2w(AVNKGK2~ns)sv)@r=sZxa@gX{WNHE5WQlJpo zIbaB7$syvR#v6A8Px8K|hjV+^K0;ueVq(t1mo)@IhJ@(Dg&u1^*<2BE*@$z=u}+d@ zJ3~7xDpm>?i(i>p1=D2|>2#h7fx@bXrg$}hIwHikapBxRJ*BlAqX_oH?xF}f{si0qd-*<)eW~Gb-hzQU= z#62p$$$a#bn5df=3+CGS++Oe>JOIR%!s92#8rLP^%nX8Bj$0oDW(Mx|TF+-cVp{Vb zf314OWve#q{9BdEs=253I1oy()7+1aL{`$dUCJt&NLm$Q!FvtwWxW3&doL`5v2EqW z4em!F_PtKGl^JeV!21e=0AU{B`5~Ok81L<$Qr|n|b1J zbcciR?7uw^ch_>iy=@W0U6K>GRQ#mOzs33%HYcTF>m;&)@q=@@kOVUHIwuc*ynnKw zOjUoJRQF-!JDb1+bVK&GlnAD$eadc0B$nEFLo65wCTg z$@u$*Jl(5OnytT{77p)fpAs97Ue(Gy_}QuY7!tHwj{@9%*lx4QOLrmhzM^LM#oT2XH)PL=jC;M-fdb z+Ecs!ll>&~rJTTgii}aBPQ#<*mItWI#V$>gE4klZk66mhYlbd;KC=q5LsQccw#P*NmVWxHG1}O{tMFnA zbDL$Ka*#66F#e!z<;`HAKa2Qoba^bue>#6Oj*QNA5xzat?!6v(IqU%&$j}DoEiBzJ z+U`^063D+aj2gGJv&Pn{)a1342W4tod5!Bl#tkjiMJ+V)N?XsJF06GsBNpkRsaTM9 zYu;dqOrSXQn(qx#RF}Y^?e)*1uVX>NgSb_vT_ISkB%dxj6EK8m1zY)U1(>1Be}R4Oe}Um4TPIGbwCFD;;?) zNUt!Bm>CWs^MXgVVn)NOY6NrbPwRxl4d8<@KAxxd67Q7Z{@aL65;1ZXevTR*kRayJ zC>Nff@f3FB7%lnSYjs8`MC{BJY3_%Mixjv|$SD`^Gos&Z6F{_qWXBKuL%#Q$i|D)A z5Se-)py}_JQo67%JBUL%K3z*c1=0?3+OOr^YcROU`e96njo9UsxFFBOW%9rgh1%)# zOzK@8R#y@_keKL9EqoPn_iumRiuhRf#9R%^e9Z@$uOB%?ZNgreY1`D^G&IcidkTYs zk1$T&;)&)qUhrZ00EPfDF$Y0)be^0j7n!>4s}sY~C+8HqRINVh`Ze*r;5ar%E-xy~ zcpg)9#Z52RU#RZyn3IL-^@D8080z;6v^jHK>SGp$E%IV3t{S#eFX>UQK{%T??PlQ< zzMszGp9)_H)GRQyC1{Djo~7ft4#TgUd^L6qZgP&1W!^>(3p*O?daH~dTfIt}9NoeH zTGJqPwIr$58+=U^b#eW!}7y#mFz*tTOG zZ_jfts~k+V2!iD_3|BEv%>p(Yu^yqlIln{q7SuD#Rg#TtB#ko1AdQB~oV+Y8B8Zm2 zr>C|8lg6VgE1Bktb}Ck#E8XnmfJ2k zKQvoK25iZWV7+PmO|e$$Y8zpi1%aXtgIs%-&G>Q_I9jNx*eD6DD`gH-SqF{HaOsXD z&JU(X`;2^)(oYW{DNJpCEm+&4DEMO7DevM@I7j7VquKk1_Tp}XER|jL`64E)6FU1ZFDG;y|Acub_xCWYck1RjR@BxX#XyRidcVl(XI-mt z%WuI6OC@vNzPc{duiZFs*$Yg#v(WLQP1>!+cgaQb0xKd~a0KDFl&I2%Ju_e#2!b4| zNcj;_sC5n$A7?Rx_*KP0VDT`+*7Km&l)|FV%SQ3Z`gRKWR)T^fBP;R^ji0jL1Y_jn zgmFC<>KtHkSYim;>j`jS5 zV(|@W>Rg;wj#9Gh-s#z=C4oEWVk3eJ@3pM|2a+fdERMF-rvVf(>_!mz(l(!YuGq=Z zUD3|NQ$<`{B4}svLJ}c~q&v@=osTgKCc(Y)$h}k-S1?d^T^wv#uwBhp1=ed$w7^J{ ziRZ~1*vj?RwUi_gkdp1ZcH!O^fSeB`mY%_`Hp_FUyVPkZC|_7028}A3oxad+1^X1f z@;r`&eIkM>GfK!#jk?>z$aSmpS-wT!puO)PcX|#CIScGtKV*b;o5{aikgwIv5FnQp z{#d`tQhh$(giF3JX)`K#eI_IC)~ew4;=sG!m4=7|s709|7_MmbN&7nxRM?xwA+_d1 z=dsCBuG#zgTCMt5-Wb!YBG(R8(kGl@T27i6^patFg#nq|?F-Dm97K3?GP5MG?VO7y zxj}Fdrhm6JIjjX)A_M%hc|6jD5Le-L{g=&WroovyyA&3q(|VIEZF!BRTT@s~59?W( zRZze}I|_xo?aZ%*Juva9)d3Y5vaB<<08$V3ngxsh(|u#T#S+1;LJEUZouU!DG~_rF z^8^htb6N#-C;!dQ0?B=CDGML(kBkP2-{~*35cQz$U5lUX;vSZdr4AHZw;izUADr4v zc#Ic9GFge0I$%(RWbwL3QfusvQf)2Ro!QvIp~+y7sdK@yG@TyvidW2VGakg_YkM%{ zorgJ%f8YmUa<|CJIJYLmGb$}$lKxPELUG)h#DF`p+&D@ri2(J1TgQJ~sFru@$W_%9N ziStWpyPGmmVxs1GOV4TEsfQf`N9G$x+GA%XNm)6DIUM@;2&t5K3Y}tS+x3F*r`YmY zD9~Jw%;3WnxOt1BX4Wml%W{E)yD>YIM?MDNO{2zJ?m=KeN4kX*R5;6ZWd zdGy^cQcB=-)Vv}`+rt#8_k{R6ZNr0LQ|O9gbpEE6MeHofGn*^K9=>kt1DDN(f#G!Q zjru}aAw4sq3UALl!H+#}NKM^S~RX*!tt*h^P4#CXcJ%Q8`+?Gn*-4cAYLPuL$B074zQPhBNFeMK1YvEkp*faZ*%R>pEo z5RRBo-H=^zkrvx>-*}jZ6~+oIr!Qc`QNq5J7l|TZkAl+z@V4o+NWBG8OHoBkQ+j6Z zJq%_g#L*Jt%`<0e(2p*QBO*O5s{79;GX79I#53vbwP{*#IO=%y*M3US69F<<$`?6& zBX1eZ7NICjL0<+k-qHmTl3d?SNaz)oJZrF<5958UezC4D4W{a${$J5$u zbRhKvIbUiPM^wstXC}DHA(-)IVeD>u$+hkW*8&y>MS-kK>%Qt!go%u@4r|@|==QdK z;sQARuIJX*i!W60i*0@l)=+Ymd%{*aUu15UGxgk9GiwnqN0x#}ZtR`eqB-R$JAJP) z9Ql03!keNUtT(GBDU3*OwsL`%R5=pV7l5U|r|y4eI^c|~9KjyL-Crz~o?Rc>ZCOfw zIouck$!AfP(X7 z#8|kDLPA9#EM#SiT)TE7;&HWx%{3pdJ`M^$wlam1P+3iX%eI7tQNQd8Upn_q7Pb6v zCbH(2k3R8bqCd|B<2*K((|x$CM!LCnaY=^C8mjbUntIY~0?rh<6tc9n)WE>R-j;M@X{$ckh^S&gznFnUAUaZd?e0RR*2VLY@AK+S!e$<*| z_idfO*roXwUjqIJx~&Qh?hr#{4_^G#XM~)0*~phZVI?XWOKK^pM)gjpAi3>^pK?F zVaXtip!uvLuOHOoaM7v;(HGPmIKx0*1BP~^Ia-j%X2Q;Zvl1QA68rSP!Q=G*LVwPN zO2i+7H|}(Ko-5ZF%13RUD&0uE!7za+w9-W>RFoB?nD9US{`=qlxLj(Zid{zSd!6-uY4ed`g5$=4y{p+8v zu?Hm2(H{Bl_t1klBGA{D@*4U6uY6?q>j{iD$LcI$0sk@`JPa4PJ4*3$m_M&vnZS}9 zx-^J>Ozo`nB6#NUt&+y}Jrh*lU&rw8r;qMTjiXZo*C|Hrm3H(0~927eb8cFh)a1aLhpI80`U9aIH zI-~SI7b<@kvcD}Y!~Ah2P7bDmrukHeRhlP2d+|P4p?e<}?@2uTA^N}%`Tfch#BRbJ zNq*jQWdtZQ?9e#=#}9w`XCSoXnmWm!3ynPvaQ2VC?pXh$OryY0L_eC*hr>90F3nUj5k zCAyZywp+d;T8U-(W_`>hX=n6?Ln89VTS84Q3&_5gp%sPI@lASgHpdEEb~J=vbbI+J zh%~*c9^GzuVT~z(md&^=Az+Eh^WG;{EUU^DR`W==v!mvNC3bhJI5c1X<8uzKyjC6? z^cvM(q3HQ;4Z&wl6>-%hNPBR*YpztU6)|o-T6Y?rv7MxztcLD9n(wkF*e&kEkaEVtP^jt6J>EJD-2eXH|d zP+-u`=zm}PVB=;`g1)zw^-ceycjuOT+cK6)&zu^@44w|5_aUClpBq3fnzgQqrm7`g z&o4L(=YX3%h1sVQy*mc)<28Q!;dDX_mJHU4t2AobmgQdX)*hRU`200wO4O?HEd1TA z#YcZkTH>oDpev`d(;DaT)F&Me)`9td*n8`+DA(?7*g(MoK@e$_K?J0`(}4j2=?0}k zK)PWYNY?-kjdX)_38>@{BHbX}pu`|Uz1Qr$f#>zv`#X;Bd*A=Q-(TJ{ICEe3x~^5{ zI@g-|nNCDEEaDQfy-UQ9lXrTaiO=mI?fGzZL-k8`W6E%5t2oWkJnHW(7u;e)vrFo@ zVj|C*o|`H}6VMA1k~hfgmHnm_hi;;<6 zJgN$N+slDFvp>QOn+cy(^IU<0F?$*JPfW;PR`ZWtU`SjhBRTG!T2V>hL?Bo1h%kgO zOQ*!~*&K=Q*?umB28y76SY_*#6-%r*4jC;J2thRywEAb)$7_lWcZG*Pn&-v~dde4@ zwm;-bba{N0xNYM_`~={7)|%F@j(a}|As4xeVDXV1$cssnPQK6_c^6DB za&*tH+(yQfEJV&R!=f0{jlM=JzI&)oe|)s>y?j8FEV2Oh?(A)IRpUSGk-$A^pP>-kQu^;`0^t&LVtgj|u2 zl3|EsTzg_-y*lCk=~qI}sCt231eI67K|!~2Iic!1RMVd82Va`|A(VL`^%jDmZaJxh zNI0zf%cV5=wPiTO)?nZKV{yT;)o7)$FFCJ^hNm}0#4Iubc6Wzd{eiZYol)1kT`+CC zTI+H4WH4wamL}?VUVj(Ww*8h}j+78m7N3|)#&fR^hb6q1XE?Moy`JORo1wtG9aT&2 z)-ub%pf&HvEwp>D2$#!}CM@5}w3F6$T2d)f&C$L$H)(fYpmjonLQ<>dS;V60vVOlv zdH^w-mqK0Q?&~}q3ePaf{Q~u46AFG0l@JOMTFqjkunT)zu>Llp;0>OA$LX304u`Bh z&*F?&(p)%1t!uucur7QOacCS5YN#HLe3U0v;x8{p4aN_lb z#Fs&h9JvuUSK`kRRNn8sZthz9c?)+%*@^#im{NXfE&AGIK*`Nrat> zpO2Qe8z*Ug%hu(f&()i7x4)v!GD|1F&)&AOY#ww!g!CT062!Ukxz4_%FJ8vigQNz& zjvEwM8a37>RpuV3weAU=;%FbL(nT{Yb{Pb7vN!>@Zemn&2woYneO=7J4d&$>CV z-^qVx!AmnZ?1#*u+koR({^EC}FB3mmjh^@1rR?KXWi>AQ(o4Cb`HSJi_d_WSSc(T) zyoHsI^lFeJA=c^66;9Akt$BtUh}&|L5&p_`tj?NAu3PhwN43?`Po5CuKCEMO(}H); zWMrK~+E!#*988cgi-pm1m};lWm)eZ?;b;xx>7f=FVo*&wsae(uq?_w;g+tC{scPHn zB4f#qZ5Ag>xI2Yk&<|ef9NA?}#H$cUxLO~}J&+CYimOB8vTIlAUvj&AgUZq)DcTsV zEaJXrx1gl_aE(h`@D}t&@do#R<2ReeAsF+dh+F0n?(#sgapW@!M1(8I%Nj3o;+`4% zKTlaD3pix!*BI#@ojTkXn4mWyz&omm5TaS?64Q|=Fln}py-~-kFyQfJ>hkK;Yr!h} zJKml7`gKd;SK--;=W!|d@2MBO3G~#8=UU-LRw6lqw-&oE;XBt!7S=Y0 zu#{L*E}Dd#3aP`LYGEh~pPf;Z!sTtdJ2xKH8`+Fx1n#_2s{h2IRQVnv50@TafeoiX zX(QU4MOv(X_0B*Vb1}tglHKe+0N19|l3me^t zRE8+uje)0hKaSEx} zBlQ5;u*Qf5(NLV?=S>zRG0Se=WXboew9A7Gak5fKkwL#=8SX28`7p^(}jWg>~XJUe5%X1WQMEzmFiNy zh*gfdct(64njlXB=QR%KV}EAp&E4wY6si1g?K@;Vc~AVkZpUAg>>A9a(PTGeUtXp{ zQTtB}KvfyuMa?4N?QQjg$WA})YK>S&Df4i?DfqJ;9_>$`HHaCBB~9W~^Vu2!z@c7Q++=m@K=-`0U%3 zI5LgN|)L9QKsz?uO>qTWk!bzHZ>WjmBLSBdFEQ9Jtl2-%F?18?*RGO+G??M@ ziVbwpV*Tm%@7R zVro)VA{AZ7FxqSw{d7xb5%ZUIB8a3!yL?DzMt5QJD&E{Bm z^dOo`Td3VTkJEzRTYXVfB2({T(1t99*00AtcpV#Jo}ZRDOw+YDK11WgH|OE!J!^8Z zKOH&ZL@$O1cewi%@S*ot4TjRqq##99HWjg~Zcidpg@!|aI*cPG#j1nt>-ROi^eP+8 z3Jxc+4r{rOwlvkkI{BnicAb}mp2+9Y4{bX;m?eEBfbj{5d<^3-BW9xL@tY4o|UUc2FN<>nt=ig6mXFG^Uz1{UZH8*3aV zqQ;q5uyHBgd0mtuo#S`dJQ64r=<7Zo;iQWuF%Er4PoL0F$TM%?U6gse{81wJ{(OO% zNGD>px`*zfJM1g@F!}PR6h$}9_3f46?vZ-8HE7IxJ?JtmH%mqS#r}Z3V)alcyY;N- ze!+gZwLDXb{)ECEz6WiH1Jq@g5jtRb__&(CG}Mq+ukM8%Vt z;`IxRxr5eEx&UL<%LWU^{d~gu@nDJqFFGPL|3w7u`K9dV7K2#~m$rNt$Jtea;l&cc z?@}6WUhH2aQDmyS4(IeJt6RjEU2XR(N+P2fWUgQX2bIuK$6 z`d-*e?$1Xua(`|_)@%&365V^=TT~%7bwK`nqE1$rLI^oDxR{_mm#x(w?6@ELvZ1+c zy>>r%qdO)k)b37xn;whs%J0$PpN4FL5EwE!xvyuq(aronNbYqfVZl<;-JS|`Qo|f> zeDhutdGAb&IQLe#nG#~Yr%2=a(Qa>1e(uM;GKF1~Qt6BH2sFB|Q6YhUh5c1il6QVM zJ;(f=o&6n^)KtQb4ZFOWS4R>ptrMQuLAv}O4fMB5vJpELgoHwuLKB?~nj2f+2Qm-R z#BX0clq@6_^V^+pdlX!-=@(0SuS)Uucqp*(n!WRjD+Af8a&`-C_~Unfj+o+2ae}DZ z{KprLBL1YqIZ!=mybd4g?D*z)bj@2x%SZ1>sW<3xEIKMMp9JD>(dc(Y*#x=FMHY;x zoUa-ifAZHUSRD)o?zPu)U8R@lhGJK!-k6-X!@^ez*LqE(??^H+xaDA>Yzg;x@JiXi zqR@`Jj+lIca4NLjFCLl$Eo&QiATl4gZ@Mx7uh8=z=4~Fw)4vQ8cXuv*#cZP85&mMe zE2XL{qD#UzR66N=5~=UA zIg-zW!p4&T^79>J;kTg8uMoJVUsvq}Pt$49yi42dceX@cL5fVa%$u@aLD1>i?>b1* zJ%)QWS9{AK4<58y)vSVQJe6~VhpSyxKA1b7mC+RKY@-#V*nrNqG-3yltb{NH3kVi$ zuTXaEnbbWzu!>Q%=^09f^AFkP+*!erXPgi8&YXV0Q~%^V&VDQ7yngy9-nU)_m52IJ zJ&(GA22MT{|2~Qe{_Sq+>Z4|%xh{l)p2s?UMeh`^O2|wzt}(aD@Ddody_+tRZdb(r zI@8nOGHlz7TUj{p`e-x_)ldViCC~GBUH8Afxbh}i44#z{9L=#IrTJ?`X@)XG?#^9g z1f9LRNYM>=<1Qa7i*;hlxb$5Uiml!}OY4?xb7A-PBOTh_x@b!ks;ta09OQ~!7T@M; z9w3cb6W-4rck9i5vd0KH*V*dYyOHkW;OUO=grZ1L#$iUQe&;P>vm9`tD0+nse0y?+ zM%xCDazl3w6)`f33(I|mf3Vdegw;dHHI1`KkuYpMgPLg1IU5?xrY^2lUF#`wpKTSg z`G|h?@yP}~vG~^Htp#d-5*2$s>VQ}430g2D!K2in6<0Pfg03}Qg8x0!`&0vh@Q*9V z;WLY$g3l@eK}S%-IG^CXK#)MXKl`yY(CD`|rSqYvk+WTIWnSbgHc;qy9_o4NejKZ8 zOkJHcp8AOiipBp`fPsVdM%{}G0G=FV3h@?kWh$}$cL3UcwKS&Xa zAX~w{yK6%~chKL9%@ZaU^;CbmeNw$|k!mN&eMIkh{)*!Dx&t(}r6i%HuU0opSWT%V zvu24=IkEdkYbs3$GmDVW+AU`>coYBwXO~JS#6>ZfThXnIM+TZ= zKASVI^g$T*eA%>b{Lhde>EhOi2`_|Oj6k7quieJG^|;PI{+^li@j9GNU4CnGAd`>X zui_OXBEvH|aq9-0&KZ}4U__#$TgBnYV)G8po`WHb^SZcG_uyNPS&Jvnw(`x;p1fLi@PrlOT z5%#hb4J(8}xMli&)6O4002tGMV1MHbZj-=(wjwwaVBA!~d%cG(taI(t>E3O`_H`Uj z-=a3AoN^G~KGPUazHBfEdG(ZGB=er11?R1jg2NKSt|Ls2g2=X8bu_5)LXEi&S9@kY zJTgpYI}C9hd&)&BJ7ESj1dwj|s)&n*TGo>5wgZ!!%;nlh22Gu6N3&;%_w|xc**t#- z@szn+t|N34SK*P&y5iM2qQP4LFA(H%?2$>;P_Vky?{@L5jQnbufbWq?`oWv;C73oipHrjv^glZI;4k~ryWBPt zgWrg&XkWBdzHaZ_dxJ#4;HTko2Vb$r9I0KQ4DkouOL3z9-KU zuj;m!+~z_%(?zsX*}DB}bna&Na$);UPs)1!pM5@DV`)lr|O8#=UY6VKS-b+HuahbNlW;}eB9`0`)!Sl|~FRmv|m3T&_h=i%( zV@9UuG@w))y1c&|W*q=I*Sybc=Z8|3Ek1)w+~olmMtA@2#mU-QWw!_*G2I}7IG^IrQUi`@o#DyeIW z*%|>v^bmY}p1v#yT^jIkJ=>mxibku!L6%{S!d73UsOp&dbJfa|z8a4U_Xn{+A$*SL zAX1o7Sa-O24hS^5aY1g`*E<=(^Z*`U`$m>eSdu3>1t$f9|ZLd?|`a^EMVQ;p$3`f?#G4;i=8*bB4$<99}(Z%g%aR+nNO@{x{vBk@{N;K88 zCBcgZROX+??G5Vb-52jwd;J)&e~L=5|Aw5nZ$I8^2_5@L->K!y0WF}v3710e`d`P1 zSxDeHCgxh}vT4o$RuF9ztnK(N4!^8jJ06F0@4+TMT<)#ti7TQN2T+_p0- zntIl?e z-wLQXawwQtasupN)Y9t?ODLCId{vUd@0x3XNpt=;tNJ4V7}P@6iNT0p{L(2lCXuA7 zMyaH_k}PiguhjK(Oy7`(B=FK7`i5}p^(}kyKS+_v{lGHmQ7^hp$m7<{vN>92VcMe^uepj6%E4;NmRfFf~ zRa2q`RcoqeJHk7MoL2^z`s?!9ZI=qQ3c1PR*4e5)m?U?J_A4e~6D;c08y*%nbb!#6|_K(3@`eak%aM7#~cB zld94=zF=@CvJ?^?EUka)wb$g<^WDcZXfmR6)Tu4g@SfC8H}NJgIuUFUMx*7rYWtn^ zA{pL4%MN}nSmX+Fr)mKeO;NB8p$zBL4Ewoyo8hLxo5Z|pqiP-$kZweI=w*C352Dlp zRX5v8xmS~aT-)2Ljwzj%Z;UW03-32ttR)CI_nY0n7#xPc6}+z4>lQP4#W$2ZV&c%3 z%yDb>>)qFA$CA3Zyj#5a7;AtRxMT^clC70qaLD>=R(V8VZw5>KYlntaWn3cl$y>T3 zTpV8@Xo2d&*!yy!*6G9`>|*9fQmoa>w|L~b(K(p6aZQL~*kq^X!_z4P7?{vHx>T`k zYpc7x+h2F+o7ZsrpjyRgeFT6u76a0r2rROd>RQx`1Xpn8lwbXRI&Cn;{qhG35~paX zpAsnDM{;_V%!pyZ zAy3@ATRBmNEebNBUJ`4M2C8wPwW!UKfpWc{+p)wA%(94yi$VC}+pZ2Gm$U!yp=4{U zbK#Iqk+@m1@Rdj2#=D?mV~nW*D$m;HD2k+nvyK{2`G0vBxNJZB%?r*Ydmk|#h&ZCx z@7VjAu0y=NG%<1Y(b5_=KBRk)Lcq2f6lxz8k_O_|racbtnc9^_bVlFd;|ed}bVVbI z?fTtPy&|AD(oJ?+f*rS7{ss)XHhLA*5Fp9yAEQO|Z&>Bx*5<7(s`cbtd7c~}2h}Z4VsZ3kw#U~?ImFaZBg<9>)14PzSQ66d zDu_(F6?wNHJkp{aHimj0gl>ci?c?SsDnVOZNcT6AYT3mIYA&z*=A-^8{|4Ebnd^Z; z<{o0rZ5B`Hg0v3tejkN0-E{;ZGNNUfkeEg9U3iilyg^EVw{MW8T||XN%TiFWDS50e zIVty2{Sot|*Kr`5+#Am%uPa0Ey@Y&O{i;L+4{{#IA4N6zU7Hv79Jia3AxjmWMJDt< zynWbXlR^5qfl`OfdD+lnFo(hBqtfJ>=Ssbkeu-B}d!V2t!sYLnj-P5lxO%QbR>poKTF64n~chkInOeXT> z!RUKv5=lNZxqC2?Yp$L+3sbLo+6wAr(+_KUu%>PzMC0(jy?}IDd%x?Ziw9wLf z=mk5y_{@6S37g3ykSQSFaRt`4Y7EwoGWIWuj;MvEJi~JZA-!78QKFzHSe!kp46#>g z9hH|-*Xu2NQ}UZ#Q~){RgrbUkZ*(XxPg)%w?ugfB2BpX`d}LGd;oT+}Cu9J~ZPd8? zF4>NoqZLgP_$7`nhDkl`LnNJ{S>nRzn?eN53iS{14+Y$odW?NH@}@8LxP_z)$*m%XVUDpch51PGhE1ZTF}+n|q-noTe>Xykn)QLZM>| zj48PcOLQxa<#b!&nt8OVM>GbD<&?hqW;U=Wfh|EK<1kSWn=OF@{Hv!NDpR!a=@~Li zI5@V4-X=&nsj&|i$s?M%jU$~4HeY}rtC$o+yQ9A+)7b5< z*DIwxdCBYPyLeom(jeO+AJ1JkQ)1LcJ#4T?ar|QUX1>&^egC@_3#|MVg$rk*>;j_v z6%>2z?vHO*Krl;LP4@otZ7 zR((5-Y;THw(gbN^3C`%*_wLGpd5xZ_bB#EH+n7(HnP)GbVFqVc7bJmG`ojTI@$%$Q z2wunh|F?Q6!8n3( zvK#z6)%kO0U-R45feu(olJm{Z)6nz+3_~j2xxf4BzZUiP_uljak9?s>y^8Kc+Mg5v zBkPxW%eRLo@0{+;sqEIl6HO_ezP$M(25ec9qW#@}|A-*qtxV&SVV`agj>|o;XScXQ z{&j;yKRJWNd6;%aojmAEEEcex%j&cw{VtLOX>MY5$RQ73^7WNV4Y1J15_EcH=R?wf&NMdSmME9AL|YQ|HD1 z{Ua{mP&U$V)2_Kga1t z{jdj1@tP3h#5;Lp90=yH_r;A}`?sSlN(~WWcx0*PXvMi$$`tN-Xd|CGmfFvc&ULm?yotHoyXKv$ z{PzWn5I(*(qdXOYG)4&7JXOU1Dg;rC5d3P}giasX3nK&tfwv~7@BEh#UMPSdGxVQ= z%zs~8IC$jHA>K2#`=?vrWF%(UUO!pwe}A?P(Eg;k5?;7cH)~}V%@t5dVcenv;#;;{jCto ztXE#bhb$CNz2=F{I63j(y?`&Nn*0(^b!uNe5#aPuBd4;J9~~1X`G~?iws~)#{_!_u z`lJo4#z$JD>&XJnV;K_*ki$8zoN z9IVTGZOt=vd(JKq__4G(-g$3@r~_zJ2&$#U!3!nA`|^?OJLGENg7yuZwQiax-cPb0(6rI9CKCWBxq*%*APFm}Z*fp4+{G5PMmJj!|Lo4Cxo2NqT>j&mKu_C47x=T<%d>OO46 zWJ7z@tqIV{+W#zd_oNNvb~A6yvqVx6A1pt~1m*782B-~xWYNzbSD%9(BWP%x!T6^d z8}6`bM++^`R&rvV|I==N55+jSB6YLVbm->$3okLpU3rApxt1lf(iIS2F@#cCL}4+G zK)KcZxi&!qusY)(ezwIcNwovdzNUsRQmns%6xMIP>hng?ukboEqMegneG~1N4q_Xu zUqNueZ=k+iUkDv>iXa=L337+Qd5azcg5oRXbhpKgV6xve4s*_NbbxPqISPI&>_n2^ zoiE_IgAx(A9ld!^_OStJ8H9PK?Q!(?h`KS*OfgpJ+B>H;bHD3F^k&AR2xnT=4=aH` zlxPEBfS`?-@l^c|epmLda`U=cZ*PVi4Td-k=Ai*5!}d~7TV?pgP6lfc=bS&fyk70|gI80bs3mn??88xb)uubiS3^nJWLs)ej!9;-dkY($hE#=jlsdA!&kMGYBf% zEah6l(2AASu+kT9tAO%BPd*0!nzyCvL)y%`bp~s(oK9J3PPZUM3|UbE&g-pjJ4a^$ zwSvHj?XlbrTn*FVLe%!|ol3USNN%+*p7Nl8!>?#^emj*fg>Pgh#`tLU>&7q)3Q%5+ zX<&0lQAGOr;b{I?QQ9-z6}Ft*+F6$Vnc@?)CRk+$YYtk<^JL4%w2Q>B8-}RnAK5Fk zgTA$jKV?BsTlhB>tR%L50Wjd&-{lMFv>9H7uL``7Mo|eqH0??t-V+J$h~a=8dAJ7c zKaPA5-oEGX8I)CK(ZOWRb)f%E#^1mvc;{B8eEU~j_t{2dFPuY)8w`RHwFc8!{pj!^(y~bY(~8j3}mP;3E?leKk9$w)7o4Pzg2p#oJ50usXr2S=8ES zvHQmP;dC^Ud2yPOg4{kFRJ(E$^{(L6Gm!O{IG|TutGw1?jannSU2xlSx=prmsMgDi!yIT+nJwuuZ7a8`8Kca7eX%pHjxkgf-IZz{ zS1{0b*BVdB=Z3z!VxspvGRvtZ5|4~av3%oK3&TclZNlg$F%>a*TU472n0%?acRxg9#@!^sF8#b@NTX7$LK zx3I2n&+Je^S5tYJXsPJ$=L=gv%$2YO*8InGK5i@QO;(6Nhbu7c4+B)5PqCC&=_>PH ziRG(MBXt6DS!H#JNJ)gO12m01aLK+Y+s$=Y=dEl6#^`&_umumc3|}zNpJAf;5Ug zuEJ*B-R!+PnR4XG&<%fnkLh*($bXEkzcT`jtb@<h8fM}l(D;_pO?s#CfU^fgbVIFSo!jQO%8Z`N-MK5N9?S1lnI;`Rc86*DRjk3>qJM-P$Mic#kBlG^h~eATrRRyc zyOE>a{CREVL_%}26%()X2|N?o8YJ?q_5*~ZHydY&jMm;VsV%W8&@_vH_JMg|!9SyLqF`ES~-A6)YF zq*qQ4>w1gFik8}VfeYtVQ6n?K0rPV7;56>sdw)^+557NPs*fFcU^Ah!8A_VaBpe%H zEp=R{8m{|V^V10D}p`c#MkmsFj-6DGNv zVjtZ&QDS=~d~t>n^rn@dgvg<{Y`rw=3G6a?#aICr&pj~e|LNz*B1DE>8V#e>T*qn%wLnQu2GKEM7)`8 zH&p3<7BzA4t-xcvS#rM@>!SvbuEKWZ^ZRfpVL|rf!frZlzFr1;J-ZUY2MDr2*ymsa*rm982%MY-^1GK^6=?L zG^5E9keU8S9@J5Hc`uzO0}L)%9gST$TB+JBA8leSbJAn+Z$WVqW+RntzTG|vL8(7o zqMWBjZZfD&j|r@Qf!N*%k0#EqG6}9)-s&29Y&An3_CSYK61}s!*fW3$I(_PyO4_t} zy!CTb9-2NEP*CKsGt)=L(w3-#rV_<_9keibBW|ys3!^mXH-*Oh*n7CezNxZi(X+TV zmHyyvzL0SWsV~o(78ScdUonbP+S-}=Zb_kC9LtSxo`4FWUmHhJtqx*e^bJ%g=42uGQq-43d;8atgt6gRuY z@a@fi^ps#YE*)|!C8mo{_}sSB%7kpUpMv(Uwqw)rMOG6u4q=K(`AAv^$=V zeWMP@aJd!25m%^MB0V5fW=~31wmZcK(on0&n|D>75pXzpeHq3s;dICf07vlai*T__ z{&Fa625LjvZv(n=<+L5kG9=8=pS6`{{23H4!$K$p?h1)en&`=FtB6EMfgB6yo#nEf zyL1De3Y$~oImyvX^flKUlG0<$!|rERMUWx@IgG zYCuUo$ck)WEeodX6vt+ru2%8P`&eq$)5`mNaNU#L<(H{qqBu5vj%xM;u0Au8`E43h z8<2EI%x>z#1IKK~WvER(=r7EEW;Lolb_wRCUoLLqwUe(X=ld)}?w~LFLbI@H05ebK zPEk^W@I7QX#c?ZL>U}Ll?^OK6_=wghD$3R(Ac||{$X$4i#&Sr42wh#GQrY0j=eiu^ zuqM>XP>v0Y-$wxU693mNy974j;)G<>TD9Msp|jiIjY{Eer+isGyLDcUv@vlOn= zMKVyLeW8^L`#4M}dvZnP&Fx}2r1A$*;VWzX$UA%Uhb9zxIf_+FSG|wQpcx&}hW(x& zWn9gNIX7dtuQfCj$^@c>I_3WE_qVVr`Nv|>A0imsG;`uT)5-A`{o@)0Oii%&lf;(V zP5~aD`5D~zuJqI*npb9J?%todQwhgc@E&ozbZ5nM^E%cC2b#2#E$d81pr*n;dP(ky_4h4_&>nD(fKa)1rY ztP8)AdUS0>L!d9f{(Gjw5`4@UL$>?*+WIn&!b`V1pWf{=TDRmp^ z*gpE;lrfO372UhERujB4?v85L_b!akb>D3t=2*~0t-KoOoKPQpV7j)Y7H0IbM`RB4 z&1o)p{@gR2Ekh(>z6W(vT8>L8AaMlBk|dXS*K*K`dtit2erMdr1=i`7KZFw?-qON` z^@rprd|14c88V5b;`-ufKX3D%?^da3ek>Qfg&~S3XIO$L5epp7@*{i4xt+^hd_13r zoDy+OdZO>b`5j-xzk<|C_#&rL>GHN#vaOWTA6r|+G4cQsw4}czoc0ELfQ*omPO)*P zK^{i%L8Jo;#YSI^R!feDN;ckKzk{B{`7_BF7bqt@)ZMX;a6;t_a0=o2STyXpW-I0w zb9x?I`C7Cn#KYG;kK^98lJ7Mp6Of%8$Wy!?18@BD$85k7PV6cF0fP0?7i44QwyDpp z`WZ&7?QlIR$s8N25!C*kIh3AXq@v6h%BhA~Z1VJLopfU>*4Cd`hhPyP@L|xP zL!I*HxSwhEMxQXK&I>kSn@c?PL}JOTRZ%&$FA&r4(>{i8M$I zDerxy`KckG@y{3?9wrBxvqQG)m!yE%WU1Fs(972E%Ap!1T0k5b zx4(nS12---MtPwK_)TL|o-Hg)^Fo)PYVR1~J|Af8)lCLTwjE$@styVp2Wu`0UJAht ze4MnM4FOS5mI0os3jxG-UE8zgrDu&a7~^4bD@b91{skC2!gA>5z=~?3l?H8|?`T{q_G-9okEo0*p2y&UWrNg~q%^vGrOs0H(IS$CTJ4i7 z+-<**C4l(&1*S&(&(M^MC?UA!9IBw+674rjZmvPU6CnNh!~>LNlQ)k$zSz|1Q;QrH4~;pH^gsK%6Da?1sW&~!BzQ7B#pjwd=;kOcKtF?iB#=5U+3{b9XvjpM&MKxg}et3P?+f;>Ro zz8;}Z%)%4G;mv8Y8KUw)9ZHb9KU{oBMIvJURw8WbC&OI2aWB-wvB;M}@CtnKmg9uuN5 zD-7f*EB8E3Tzb+n=CTrum>tfvjEn2`I@;f;Z!W!$3TFV>-#UABfUv(1)-*W;1H3Vf z>^fAwtD%=$T>yFRqPgdYHfips;{D?L;6BuHjPakYEnf8BfMl& z|Fd}UQ-CK4w?)eH$)BOM4@Mk0?#$_AZ~QT3_u~g!Zlx(=z@j?w-T*h_Kf8zsOmMz@ zW8^g8!KxIc28>A#qTq_^ZsAKI6^}S?z&d3oFZtRF{37BEG9+ z=jmd6QF2lU0$}1A#)|ozZVRfeIcko+!H5R3%Og zCVS7Puh){Ur8MxZ^+$4|@J)ejqeyfum{7LhjGIgmR8Z+%(5dm#eS7`*F)o!5{#?;K z?gD*ahuz_Dmbts-8=~<A?gpMYRr7+@$fS(ijo4#{Q0%FH`2fWCzJ4`@sqQVh&8}YTN!uS~gc~+wWlBD^s2qqx*_%w+1#MRWXyZJlg1g z%b)TY6kA|5^_5G>4SqiifL7%tBrZ1^p(P}%iJ7RRsP}~rG=+36(no(y5V47;vwYV9 zhxl8*r_Z^y!o5iU&U&)WP_P~S3sgv;s^brmxL3Tvn3YMjgm+Z6bfM}6Pya2&`A4xR zg+ZRrK!-;d=JAOHm%g&kw8*~QoD<@UH2nC0)oI#6v*KBxV>8I?Kztt3+pHm=`IGVG zA^@A8XHj;h+2(#7P*HG`O#DfM4GWk>dJbPNIP1&*k2;)aJcyVY{R7)hw&dJ}pZK7% z>%a89)7<*NoS$s#h+uQUm@61ID({(>Z3KoQ-PIlCQfP(Xb=L07hRqg&VJ&Id`kK}t zRni72)AN6O%zxBb1F-x3po|@SJl$(`S1#7?D72{_>o$|#eL&@|AnpItFO+^L>kRO| z`b#2gcD_@)r`q18Dmi1OTlUDK%2;%=;OGi8_a7nS58%8NJ^K)F8lk`NzH|X}_T;z{ zJa|S(pxh-QvcoZ&zAMV-iH`67bm6XRns*ZO(XfV=@fC-~+YAd|^#Ow9z~0*&`T@0P zVLUZ0V)S*4DyW?n7Ow$pF+zv0WPf?{KWI&75a+q`p`c#u^0?*YnPhE=aJGSDlP}V` zRedy%_(|flq3jFoptYH4h9`Tt+DH7}^m7u$g0DRg2lO?4hl|0dzyrZsQSmEqNYKPA z^9k|G&in0W_-r@8doH^JM(Tw%+{z73~;whD;t+K4Ffp|E|FlE&(taCF&?jUdWOXw(MH0f=jKZR7gTtRPLb6 zy`B0&GC>xAqVAx%4tCzOtLMG@z&NGE3Mq_a)*RE1TOwW?`QTc!I#C0r5Gt^A_2;h< zTczNiDdJyO|ALYpZwx8UWVRudGbu=8d2J1lk@TrXU2x6X_slf30YpwiU6(K5X}z~v zsIy~Rh0Ai>_vTgsJOVrVU6apEAKLBqQA}5Gao_cQ-rd#eF01()YjOP(`zlp*P^D@G zgd5ZLB*sCR7qL%c1l^>AtiB!Vfdq6S3&AFVzVrGnqgsAcS7sOYvI4n~2FOw!Yr(-pL#YvDCuvs->?_J2|(iO=H*We$xCd&9-V#ZEEaTwRUay) zUi9Mg&85DAy^Egal)C#vh#y!3R$~n64GqC$lgZFJ4d9H5)&r?w?YOnG0i-ulMrMf8 zHD{cu?$qtokUapg-#7Eos?e$7zxvsI%&e}!b>KsnVXJDlpm^|7z)7(rlNLB+l1@4H z?vCAwGn&AwZyn|c|4iTOJ`GSFXz=%j-y^|KuK}@Yle(|!bhc-hD2Afr-4e%=8gc&` z9FgSRZ1$qDQ`>I)7i_rL(bw^!ahu>h-4X5*P|oz>7~y=r`}=y;>9cbXQ0KcfkmBPXYLYi*}z&4)i)Wnw5G z1^S_3V|*pDce#S)(>Ft9QY8I#mPfBSR2m;PGudr}!mGz>{UiG zN%CN(fRPS|aT_7-V3E!0u=A6Fob2mgoWl9H z-fhEX64#EU#f|)|0MM0lbcIaTy+?f7B#5+Grk8x&COGuH$+Zt-Tf@=H}8=R0zHNpcqHWzU3H(DEXd`q*1w!|n03B}umzXO5V8=or{& z`<#EX%jsZ|?|V>|1>Ak#4;3;uC?%s_&pdJE`zM7*$V}F2xCf9IOSa?py|9>TjV-V& z6S~v%KY5qDrJ&D$|Fd<}s|VYL$24PAGxU}zpS6|K#`{p=J zhfL{lT*IgGLDzFRt(X&M`>>I-79>X+_dYG0Kg&%6M$kYwQSr_2{urw8tv@5n+aNf6 zdMvc|Rl>IVrj>2I7r#eukx*R&zJ?W(KhP&LCnl>WHWx?_BBAy_a}iX`6tszD6vy+Q z^3Tdjv!U`nQ#BND_CE`0J^?MeP!OL8_w*me{JT}VlEi89c)nt2;@O2QdLxa0-+Cve zT#sf;o#2O{vx$$W;%uU2gjIl(LHlA24rHe_ew(;$sE?C-m(e^q%~qH2xOo@+skkU7 zalG4x#-_u-5`E-`yBRQ}c4z*GV*SP45c^?mTvOj+Hv%52R-JiYCi@039TVbBw;sCs zc47=Tvv^JVqVenva=|CEka3)(NMnDqVCx|^^xPfksMav^`H-*jjS=t9VVjAlMNddx zz0v0sC056}zNWFfRWA3H&1PPa;Zg<34-vK~+m7GYo@hCcBa;Z>T!JPrwE;~aQHo&h zUeZUoD-{9g<{#qG>gp4-H%K*mrZ3(-;p-Py*8Ocb?Yh>~+6!00<-G>!^NZR!{ocQ@ zuT>5+a+x2E<XpCL>g$j4 zpSJshJNN$Px$ok?^zKW8HyY36%{@vTDkD;~k|Q|t>glYUjRC<+B%`M&T^uq)lgWPI zYgohXty&gpR>LBOa(>iWrvAA00V#Zy4Hs-{1H~~ChMKNS2lDi|x^?!Sf?50dB{Z6F z#*|e0p>K)*HfT{0u9j)hj=~yNuQXv_fW#{A%?INV%kp0Ldj)-mInDPO=$dmn3mszn z<^_`~R%#$#he{Yy>+)1}y=Oa9Z%P;_40F_xv-b`n(kH}NPWQEc=65u;SU}~4zZQem ze8aQsGbGDH6Q~YwM?p#b@n*^Du?B^pjuNh%fF6sDXyQ@Cg=J9u%gLAi8G88gsX$6u zmoJQ4PwkV393$s&vo3S3$31)AQW8(9N0Lgy`_hi930_aW@oV&fDbiR1ZVbP+fg~A| zPH}<`0ei)#_Yib1&?H*!>gxDs+9sl?%~fFzr3N9+%qIz2kwwT^E9PCWYH-6*9ZN=acHNS z9gDemd>}OfGnr6vdExn$&^je4$g*v*vA!1+6&{!c*IHr^=a|V_G=gK7S854){K{|) zRZS|qk$xz({CE8N?8_EHKgQ_;OHBfH8;%LLJeJ`QW)9*uOTz~V8g6&Noj5CQ@POm- zGir5PbL+2nGk=@0am)WqLq($(Rs!xa#ppN4ebPfm+7vj~CND7_H$ zd-ej3CT;3Dk&80?NIO~5d9g%|w@871(k2vPwe!fS=oUR=c5%1clN@weQ&L4vgy zLvp+i18)W2jB_GTX;dez@sM7{RIQqY?QUFmeK&{j;;spYsK8Qq-P)`7L)|Pn-;Mfk zC@U+=r_zlLs~%jZzm`#Y+QTWc_oab%D*kD@lOG-sEvujOY!b%$O?LC|)-g()P77>S zxvT7_+YlA~`$f|z9Xdl`310|a;E5N`42u;-tbQE06WyytSjpg@aKVZBt+ zf;K)Z-heZXq=vQ&5(g+srwKp4J+^tYmy4YAYU6A(s#NJvoMu)Nj<1KIi?<1JeUvddW`m7sPASBn_kcA{Q3OPyBumy-=g+A|qv@GV; zsa2EY!i|Sm_MFWM-gmgxxB6QT4B24FC~l*$r`8GzMV^L*jxOqsqv|cy(D{fk9~4f( zunCOy_figJaUoYN1-uMyFGQt@-C<0gO@!05+#+Y_7jMvXXBW5eNoT2_H9jV_u7{oJ z>~_+r`+%UH1Ux)#cGkGgS9##|A@A?%XBXfDM8TN1Q}}o}cM}&ukur|^(Q-Bf$E|Y` zd_HP$s(kN6|2xXG_$H-LOSPe5GUAi&n?C~j3{Wc-*aLhG@9?;+bTQ|~aMqD`-TS;w z4&)@9jmc{n2-~bD2yd+=D(BK16vH=~d+QBC`x-bRJ&sq2+#-^QNR9)%M)3S)&}y%5 zBrXJOKV|l6%ZICQwQotO`L$-4Fw*%(RbyLkdr_nbD>ExnO2JVA9*#gCRSzEsz*LZ~ z0y^pc{(0rsF5q99{P|ZO!~q z$D3_-ig_syC>{!i4KO(7YPEsJIux9HhE>D@q`jJpJNym*QvEqrRF*Ws%zr4$=Y|yI zrnIjTY|(_NzfuSkiC5uX;k`SA@BFla8^kA zmQlH{awtV!S{+y0SBZ%!3?j1C$IH+yQCh-b>ehE56M!vP5?e^lb^~`d1T0rvM2)!$ zVG|S4X~0Oz%GC*YOol7Cv)WP!6V~k8tIw%K?f3N=Jp7A6*TOj#p?r#yK6%VRib74M z6Y%)UANS9_q*C{nRcL7^>+?Gi$Lm~U5nx29C)_=244U}d7(G8|(amty54(NcA?poe z-3nfwy^)7t-D$9o%N8^lRD~<|KZbLzuz!Qb$R{{DRk_yM*R)&&y=iz?wG*GzIyAaq zl3ZE4eMtctvP9*RuX^Y3BdcTA3hP&fuI%E@Y|w=78mNoa8x^ZUb|)!My1^O6?LHR+ zx&qrE3RiIh&?Y8Rf_j})Q@9Akut!(ie7*`^{|aS4`JV#fh4hr((zLI&r{kL}Gvxqq z=P%c6?4Bh6@8Eq2Uxm*Y)~ilD$Ws6o{G5!JGc^EJFP&z$X0+Ud3Ja~pd;3)WfPx=_ zZH|*w>!R8+n&=7spY^5*MAl?Uy96=sIPa6R{{qKxznBz86a2>N#k1nK97O)NFq zd$zSGMZkY?U4pJQv-`yf(ASu5hilr&rK_4|(C35gdlM$w&$HS3vGV4;KHWNE2UDxQ ze1<>6AI={S(eIlTUmby*67TWv1l*g$g*!Q4ySKlN3cuX-uA2)gUxAtny$@jVv*;nuLA>wn^|fo8uM)&*!cV5%MTXoqC04qt)Nvd>a5mqiecPkY(!yr^ z83Nw>!I~aw4K-+R4uPy8BwM+#JZypYJ68U-RGIl9_((8(na53gODCWaA z`uXbwMfaGXse28#G42`R!d3k1wj>Jf~Edu$NTkm`D<8rk$G4ZJ6vSBYI)%1vzrwv=3%rrXO4%? zI$njj7PZll;O$#+Bux);0%Nj!uY16EkPi62f@_$Qk`)meGPX)mS6G<^ZuopASb9Zi zPY{2b0vl{qFD)8+mtlFrLA;btFdowA(554 zkwkx7lKD1lnquDp;>Eh4kQE9ibW_STxKl!Zg+I z<&_OFx{n_E-Bd`Nd~LW7o_69?HzJqOq*9rnTol0^_UBC!s&DS;zef=G=q!xw$&xpx zvRJHR>LGm)Rb84iY~u=vV7UH#8Z~_tG1~qSFlrOxJ-rLh?g2r$9$5v#eN0ye$r>D7 z>=$#o<^vlldk#BMd%DBlTRF5lmPY09?16A?@~uLhF!ynN@!kCXHW|muPSC|BQan}4 zVB+V3^(t+Uwbpn+qwf}hI-sQeC>%gLvw1_qT3hezwIoQhYoClmmvwvUGS9iMq zG+^7yISKEw<;|yqy;v>uKTMmld2IDGC|cNJpb?|^&Yfwk&IS2jWp`R(k4$a*b|2sR z`?C7y4}xmDoy1caef-{6rlhlX?D!2r?nO401;1nogi3rOv_r3;UcOs(JGK42KdFn3 zhUn!-)Q0OLlyI+1Pu-m&^+At>H_REQ-bmgs_0@-Si!#KgXogotvRDQ;$`h2~OdIHl zdZ@g|91wmPjv44KrB0WbP6B(^P6B}^5{wvkzr2Zv1q*;wF? zt(%%Q@@v4|QhKWZg4mcK`QaUvJ{FUZNI~l>0ffzPrfM@UOoUW5c`LI=mk8>n=%qh$ z856{4&ZnQ%eZ$GzZg?&|VOxi({>rz&S(~lfnhp@()+OEcMx7Hu5Blef1f|Z6EsXvv zebNqu-FbPWiXS-EcP0d5?_^URROYGLyA7c>P9i82)eZ zM)J4r04sD>zW8g4>yRh9b;nbu-HtX@X(vaJ>mf%>6$2vRE9#-QSX*h~YA_z)atI$B zLYg0w(kG`%6}Fb~7&5zz>YN6`^C|ubGOKH)qEJ^OQSvB2BaG6W34EGBZNDf|el8#h zFFB%J+c$;0d3wM2wC@jwiH?B-ZEi)5EVbh{rN`ub-&+r8sA6j7lYw{zI#H&s*^57w z88FSdU8|M+ks2M)k>q;}!h=I{I|gjf>t+5^J7`oZ@~23sYV}mGyruU;hjPQ4;7$Ub z;^S5%qaWU^FpvrDLkf8D?+NEjnWS7+X;GT^G`xOWA)GNuNE2bC>1qdg^EubwEIIv{ zIO;ye7S%^D?C0)PUv35|*+Pv<{oVh1tuT^yEjVOzuD)aITwfiq#&v41X1AaJU_%zYC zt4}?Q`YQj6{bf(dpdF@})6&eXsLjbnF8~jS)F2Vyw?d@-B8H|UcY1aGxBm9_%QhfL zzo(HR`1)rzky4z{E9`c<@9nF+4*JAw*)R-`r@s9Z%U?5v9F0*}Z=a4>6RSFo4P6-e zAq^6YNE^SG5gBHWC>~JU#oKk}n@Vet)c zHW036>EFSd&3TR()}u?O1i|pynUIVj+bqv#iCGkD%H|GVnD_YI{Dk%*1_Z*sr?d7> zdn0(t%zqSxJsQ3eZU(BSIMOZ$M|MgP8{Q00W1l?7=Cix+*AK1f&U5B zGNS>m;eWN>w);d}!~6b_$NO$f0@r*BMWY_0uTrT8<-!!2Q-rfg{MXT^v78{CnRoeI zkjLI3lU?|-&lqt-dh9FHsk@6Yyc}qlAmqazT4oXR1jw}}_9|_!0Vg^S^ngiR@P)Iu z0&#i~gt<m_4lC53yCr>$wJd1k zM$7w*VK(yDvZm>YbcQqUZ8d9>v6rL;5AF8LO|_v$-mL9b^mSI^yyT_<$j%&OU*Td5 zjUcFE8dnV(BPW3@M+4`Y9*g7Oz(Wh!<~m!I`3Yw^72GuS@LM9BFp@Ef7_AG(=uL-T z4t0gvJfcj1e_jZ+N)G_XYHe@4gL3481}bCcp5c_Pz=GSK^ZjU*lNLaq@T|iIO-sK) zLqfYC`mcGbc$!_gU$1BC#X3+LDsuHAX9^%0tjbOM76(qtVIvo=bQ6N1Bem4mIKK{C z=dNsl7-Nr;utNkDUZE{5h11eqlg{^!00piMZ>`jtct^vd8%C=5uZz}OUze^XD8HDXZLe5W^|1?{SRZ#QQ8mR? z8PYbk-dZ7fZFD|J@Fsh$9hZm*9(g;0$6t_I)(*Yb(vpXFZM5Cfx7GIC4JnqmSfPy` zzfPLrZ_1)B62Ck59MCJ@eM1WB)3>-Vfv3}lrd@|GXs6U_;eJufg!nH?;dLt}`}8)m z%s-l->%uR*p!yW<(oXI$Mf{GY1+z|IIhdk&oPhwSb)z zCqBvnMhA1;oT8nCCIorMhjr~y(yS^3A`?J9=KZ9h_$KrRYsE%{e6ak9K+I2|?%JgD zS-(s}2Ti&3X&gb=jOdy*Rh;Bjp6k>$zkdRz@?(T*=5asif&AF-bp@-mqd-uc zWguWrLK^jFV63GeC&M7ROc&Qmj!6RmB${68uK)ZJ+ZBqZZai(oY(Ddk@<|9CvKGH~ zz%&r?GuMZ8w_+lM7K<643gC{8$`yzWv*K-_$Q@qGDeC25`iNGtL&$)!8sL~ z=JDn;kySLqUOU)<_QK^r0>z)*h=J4*y)jBam>oha{+&mK!cUksuB(#bopUR{L4mrF zll=ZLAsWs44lTc2BBHChVhrh>0*zftLx=}f5mR6puk8n5&D-F|Fpd0132M!+7Z}s4 zKJXC&a~25i<)8xy7;S|*Gb?ZN!!hzX>8W!4RdYm1ZndD2tKAgJc?A-rp-JiXfzS^j zcF>}P$;&=BNJ{{{CjtpK-W@gCvzS?3k2R>TMK&!~D12yA;XOlvyb%yizhDc^Vft=D z?FVwsSw=9D8fo@;yn7;*`dfZ2A?Wi9&?O|}l_Te(zKA}~kWg!Qmkj`ZgWL>mFow+| z^R6Ice9e|a3^zNm>7u|}_O;&6D|q}sD@H$ZhCPTC<^M1cel0ym$PqkBFavu3QG#RS< z#3wVv@SHW8Bc!#7vL}6uUU>zLCyZmXbwhhIp@ue+!qXN0(rvmNz*{eWW#FUcwL{!C$Ir|HqUJ?&^XJ-`^Z2qXtE+dRqs09hO1=_)y> zFIPwGu25<4Y~BuHg93%z)N77382fn9b?xLI@-d?k@9Kv}it2ms4ifuwUMumR#H5A? zBG5My^XijU;ENQMbPs%47wxH05|I#MJHDaItUvhB{P|b-=5fMvK?;X-IT|q_hBV+0VidSvC&-c0{G}F zuD{HiFLF;s*EHY3gL8Z|)(2El62ZjomIqotS zbV05lh~o|s8sBK-y!W33{NDx7CrzlkjKSybIYX#*`>8ecu8Xf7h`3eg5rpAaEB)`> zxU48S`F_toV#tWKt>~g)Zycw5pa5;IfMdB7q57C1n#0;g@Gq0$K& zM*J&r)_ZQWTq0;}dHDYz<2u%n4dD{Po4F+lI(UZf_olP$kMAvtUj1{n) z)&w5^uex)$!0EClv%BWu&^cJ1~_}>>*kOCc$6*&tDVf- z=n3HTo}MrplwB~1T-xCn+g(8O?cbB_xyn2BZR*ri|zZi9oG|F?rGUcu} z-8)$}^@2dDb!zF-mVBLN&j7@F+OkD*Jz>C!lE(>WGOzpe(wGjYTcoV*LtSNTK(uCQ z#L!K5%5k||K-dlhM4LHzR1JBORbE(X*v7TP8lqXXM_OOh-Q zP~}+0$CXerQd?aK(-5@K+s&3aby6JxJRKmrQ4n*jb$BHkUZTEQFXJB~&Qv^tNAr$^J6Dm1fGz4gyC zzj6rQi7{YH^U}G6N^=vL&H-!A0GuzH^g1mddQSvACuiDMl8ZI1Ec=|@Zz`#gJBzV~ zTu}Vm6b5z^93ICFt-T;f5e6+W)VYA|)5p>Eb;`h*MWgyUI`Eu_p*7hmKwcvPNz#GC z1oLFJN%lst0X7Gm4SX7*OiERvL|oQFyyc zV=S2`qn#|u>Lyw|0_o{)0}Z-bBQo!rMtt-PiRj2A2DhG{tk7%(nwOGh<1+6?d3MSv zf!?Pbhqs;&KT@4eYr5xQeBon&;MR7Jh*=}4Yfm!oN|TOVZy_2Wk%{QYQ+g_DdPnFc9N}yO`M-%dczD97b{swquY@(nC@kkaiQT;9 zv+2C){Qu~5iiwb9l5Fmnz5dVpSpMsFAmiwD+jChKT3=X_j)`7*5{KDE^r;RU>0j6- zd8+_rTk)R9D`fzMM#<0OR>aaWzO;=0(lX{Jpyfs$`;C>_Zg{E&HP`~nX7E|lvdb|~ zx&`(fzs_S;(+yj-M>>t?5U1MlK1(`qC*icb5&tK@ak#|(C%hW`V9eHQ-C8rAb$iXE zssR~GA%wfW0UcE1`W_HV0#(~{ej;2|k%@(LiTWk;LiypoJDEnUZbH@DztD-n7NEJT z<@tl_ru#)h$%dfdw1hm^lbRLuP*-V*W+L3d=h7$$K#0d%26$7>@@<_7(Gf=MyfAmN9UmOoG%f?HEgg9hP_wFAOJ4_U z=^UcUkJo>?gQ(zBZ`pW*R3N88t)hJBlYm`LkcCbLE%zQ%8F}JY?T8El-ns12S~EFA z(mV36X91L7@x887Y!Y9;$|+?$LS6*}o#cdl=rY%InexWD=g9E>lsBAp&GId<*y{}` z4KshrE1E9#|I!8X26-Fgiws@nowP4qlw)<(xBiq-tX-1*MQr3~XtUK98UOc&`G5M* z6y5^Wr+3^=um1~{e7g4JVV-oT;zB`)#2&^g$!)5_kLx8|-kcTKC-njD2-vwb-?=Ip zW19OI_0+Spm#wKHEOTnCD|M*#XwLTz)K`F}D#;J-9r^`+TNQTA!PPq#4KnUS^4x_a zW(EbT8F51aHCak9d-!_b=z?v8+FqO);CshpmUnQ(kM^9=tyGA^7S73u9887!dm)|U z%K!DP>Fa@|LzuaM;u;fCK+xLsB{~?DE(p6Vm*eDr!Z72h=R7BcVzmgVJeko0oYU{|z-bdD1G^#bypdTt~3zgzrD#tje z$c#S)cpcI$aPi2~k^A!>U8vATP191-!~*L1SJw659cSrDRH&n$m$FWot)waN!dn>5 zKA<%$m#BO8w&);05FuO=S8g0$PLNB?zTU$xuU>e<)W-zf?+_i3dE>$jv1BI?rpbSQ z_6nTyt+{$-Sbj9PXm0?TQl{f#fb~;UcrQB8fp0?i4+=8u8m(KBhJcYWM-?h^$u&|^ zqG2-uoP~t_I%rWg{bsh~UfUEc%rnsz8{i-w<0U%iL-MU4zbqAA_Z|NCm63JIql~J6 z%LJUCpa}GA<8q0%$60?n4Oad13-BmI0oQ!M9yrKlIEV6)^5%;Zseda^SnIF6SEM47 zol6=^1EAF>NX?kA4qJ6jzT!HC!_$RG1{()7sp|BylP6`UPE#h(qf-+v2--=Dve!#( zYiAObVGcvQlJ6DubOtq}yU>?Ht-9R1mz=`9`=gn{l+jW3vA#JBz z4y~xd1PC;Ry1zR#Cj=(k@Or0q4a^n+NV=edqK<-an#0f6MPm++-*h~tB8oH{pbZdz zekU^qwx*O+K3id~Wt-8}DT+2}iZp~seeysJrUjlMUdfW9h5n+Zm~agDOO(1OPD`3H zJE-OVH*%*O5sRn+>qvSj3(WnUB6@(gn7zo@YS_w*NhlB~>!=$p+yZ+tMqQ{zrf;Ju zKGo*bl#YgP(gmXEvxm>2nQrvlNC+>s!-ZG`r;h4Zn^i|Lz{5_@8e6_ntJz zU?1gO5;}j=S^%H5b&gg<(!f3Tn+&4k$+~{?A1j^&Ofv28b`8-jF&p~0z0gVKJ^$ub zT`EEYX`~CL0(-q$aPJbqHpx13hkxYV^wjKRIj}v(?k&hf*C&@cHHI~tB-c+-(J&ZmY*+_MvcqCB2?zSWab984^XlnkqQji+Ei4GL)s`BX;4QUhGO(w`D!eMP5`+=J*2nMWWspIOxG)a(E~& zL4W{)5dL6;Dq9s;2~yz9Y|bxSbrjkGZXygop=H(Jv7WbAb!R!T73Q@XxjYr&boihv zqE$%@Ts&)V=BReN0raXW8KGdD=qAx}UrL44UPHdD?lO<4$%e3toEx+cGcDu8)tCnl zi^Pt7!RjqxvyB?}R-gD`3!Pg3biOGf>j{+Omu4RQn@eCm`BrH3tMfM%K~ERWB9hm; z4{CT^_zz`Tj3(KloOis)wOJd+~sk`*N`+kNR zrU*5Kzx_7VZ?l-(m*FLOzYkP5SC%acHVu(neq~YR@u%qFG%VRgiS_UR#&zcLz^BAa z>oa64fsD`fSMJI=@-)!Nak%H`qmB%(ga!effuNZ2XX+*$Ea>dY>lDz*`Qf$CcZccC z$NSbV#R0O#6@u(T*qWb04eRD~;N|FzC!BLpg0<=UAX%C*TS47EY9bd=9&%izy<=U+ z5fL&j2ecAyWOAB=s+SJ28*mxNpep#i!2R3pC5X-e9?$(8Pa(aogiN*W6rlJ3A8m z-|VwJIna){nmy~BJP*bd@^-0!#JsxO5e{8EKedtgrn{%r{k?m!`g47V9q7cOeE}fb zhG9 zcSNP!*0ayA!0YC8;|rYrHHZ6K&*=H!n24ukRt!qEPEj61T>Sdr`g`Ip81_L-A^9P* zCSrKEmWw_i$p6!Et*bH(^UEV8T+_+LMCC4=$?omUH;*B|FF&-&R>;T|A0^VQEykX# zN8mcYMU7o@Y-^U9J!leInZ!WVR?WCSvAK1yISa>tr#f^}M>LQHr09kI*AE&xCM!< zh?!@JDON4(8kQ$q8kIEsJCh{Ues@WpY3?YRtqPR9<(MIDU$>oEa0UGDZxaf82r@Oi z!fP?EyGk{SxQfuwrFHgCj9P^&C8x2xso3+(0+d~49U2_FFqf#@6J*@uT+*xm)Nz1} zWFX2V&ms&`d8iv7vca6~O^eq~D)}yb-VH{CwI&Pa(WBt4MVax%jl=?$)v%(9BY1Qz zdzXvfMRouXif<-zK66!;HxiVAuN5%2g($Y=aVI*bo1*Tdw=98%y|N^jopO=H`d#?K z0D~4YzQols!!k{V9|JMwxF?!`qd>zKk=8LM1Q${uqQHD^BClzU3I1Dbf6B}Uz%MUu zW5%*fEuh(~eYyt5hA^7>!b%0c`?E01!sKt2{$+pW^M`^KCcRf8VkT_Y95tm620}yp_kkN&zQttc4_9412)cuugmHV-i#B} zfcIgoYUQY?5ir4Z(d=Y0(EnQKC-XW&K&MgIPQ@K2gOP*#4W-YTv%=Lv%@BFuX0tjg z-4^nh6v&;1-;YP`aN?WjWBFDH)kQN(w{%q;gOlM51-`2V@;n}K2U|-S;dj6gk(?m?ZZ^x#rZos80u%&#V35|gi zM%=hj;o`<--yk#BM)SRYE-1&PI9Fl%_YU3H=Nx&zAfctBglIZ2>AKi+?jn027I;1f zaX|Za_|6=iH%B458WHm$`kZJlMBZs}O_HffG(rJk`lZu_m#=w?t!qid^+oyj2GM;A z!>OPgSwnw)rAQ+{{`08+bxeUSTOI2m8S z!r!$}Cr*oTE-kd!6y`Vf45=QIA;)R$*h-?;OR06qwMQ;mS1tcgbfC{610grKpK8R? zXIT#w;?8A!AEjoly0nv(XW? z0^8QU=dJVvaKUJXceP@GimAR$$fXm@^fte1`f!MkBFYP-@F5~pAQ6tQjjIcIB9)=J z<1j$i<;y59?F}>!lUy^%GCTFT8oUB7wQHpk>9qdIBFQ&Juy%3~tn9d<6+>0Cwu z?HSc5BR{H`LkD&+t(-?n0+ z--30nG*4|W%K>9SmQ`rUcES+1JS#hlem-tL&Q9`+Kk2lZy$D$;hvlF~Rdn%Ei zO0g=30Id9X=GgP{@YK)u{>1o_)vgc{K(!;j$UFU~JW=j`Pk77fnAMjoe(p>4XB)sg zfC1cbGU@MfdE4*AA`Tdh{?5*bi2Sv>{ZfLt4#2iE!j-}#5J>g?nw@@^Ewn-hb>MSq zv3g2KWx+(luwVAZ#9fkvMGu3QsVFvDIHypAyN_R2(AH@^=oH-gAZY0Sz3HTA ze)LV=sVR~*9iR;!E)qmt9s0d_Uag?y>mk`tfUTf&I}3CFm3i*z_{RAg_Y0KS`GcOC z9vZA#r*07aXvVDb?*6@Edm`V*?ue00l&w1H_aGKFKL2o?L%&t+uR@X}k$E;JgmvK9 ztqg}UVO@Q4Hfs`+%2018aDacf$n8~r>Ukjy6n-gWowcv=kkRs34?6{}X|zdL1FaGz z+J7$BtnIfS9NhONluu$7MSr31naw9C;$wOGm>*~Jt1|Ho?m^tr){6ppG7DqErbsg zy8`0x``s2^gz|x6gfX3af9=|G@Rse*xY~4bu>eR^W)=|LaM5t~HT}`gDoacSo0453 zIBW$QY2OU}IS#gxDdX*N*(!9=33|l$Zh0m-INL&FJLa*X+i@p`DJZxPz1A3_V@c(}Zu_au@NMG9FP%#clJRZJvU~mOc>! zsy@8k;Bs6i?kSEp5R)U~N}b+Fh865fLG4Ac^QHAftZL&W`3*DG!_I#WR)nR)^0XF{ z>@GB2Z&*MWu6&_7A0Yq@@hbRRLoRfF>d$z4eFy38R?PaDAP-qUN3ybNZhH&TrO{+= zMwg`4*v*0HRM;cxV}|+S!ayR`TX)nwQzd>zbX^P-)M< z=yb8&E-_Ck*|sNk{ceg)N=3TsnN1XXE)h1nPkuQ$ zH0h>`WxQ!t!;APz1tUe}ip)L4fy1c7uBw-O&SS#HoQ6D~k$0DB$!$gN#ne#Ae;|pjFm@s1?p{{Z zA%OFDW^9E@_F^mUdui%W*v{jOfTj$`O){0=Z<3rwdIfBFBnzmI1RGUTb0%(az1s3< zo63M-LS3-HMoXvz?BM2fvmgXv34?642?zcOUnI;#fQEwdnxU4!3j!NfaXi5_LV>yx z6UhKuYKl@Ji*XHteF9H{D1QJtiX#lT-k(J3=l_7)6$w_yT`fnm*+hO463yltQH)^1 zE}CSwkTIq!sBuzr{BswoIb)xH zGnj<1MI_m@%*F-GsNje_g_*n*FOT!tye&IsCo?pa7;;i|M|!QtS(McjdHpQxwj{_k zS9;n_wOmf_X=wI@*)mX|qB-v=iIU#VOVdCvYznVC!)%fXNcb%sp_NQrs(*^S_q6xn zTBVptVo<4EmvM%p9)!KWaw{!Vpj)0TT%HuN1@1yJ6Ny8QejBO9ChtLdo83Ay73`>i z%}_KI^`rx*)A;9n4=f$3b}Wkdv$qJ5on(nOKHh-svIMv5gBrT+^ttLdIP*zZ3pZ-| zw?R{&P)3q%*FuT<7RdneEuf8kgIwdY39^g`prCqv1??al1}{s}q{Af4LA8HhMg&kP z5f1OSH>1hSo^k16G}o85eGsNM2}mZpY488F=eWFTXRG?*gRd3>9`WLya6NTx+qO>4 zpPf?*Kn5a?X4C{C9tItamtNaNgsss>?$g_>TTFIVxfYGHC7jrU?2K)L`xBd0mM}jm zmCm05YL@-y^Ofqb#)(`7+WMl6lp*f9dodY1m0SVaXsb!|Jjy6h3NUjX@m9Xk=n+)= zG5ljMNEmpFJ=Aubv&Gb&{ODENw&2#{>2q@7w}*v&1exlL5ZI5k*q}ClL1CkCuHSUt z28bfcZRtp!x*OkGn*D@dn8|gpIRQ8zbs(>=E+3?NvVdY2+Y*^5E`w5Ma{q97nc z6pfZ{OO_Lbl@z3Aez|O9`b#ny`vU8EHWrDZ*8`o<8QKsyL-gMb31!W{j#$eKmamWU z{<3iPi+{y6I|Z{l4c4VSc*(8Lr~wLf#DnRX7W8L4>0P(x)?{@vpZG?yec8oEcw2<) zcJlk{dyr&vmc8`l77cpR@f_L9P4~8I>SKdEI*Pt^N*;JRvfQfBH&n4`Ml!?MIxnz< zyYlps7tv|QZ8n8V<;pS+P0!DhOC=9+d{Yv%kI6fFwycMjWtzmvlYReiVEo=Uk|xv_ z4lq+`!o__>u(FrnnDUCeOfssLZt}uKt`km2IjuRJGkt4S;907UglCJEFFNeNX4pyq z=MX38;zt*!D`!FTUr2132Alq*W%(r_4)PS-GPW{f4bHW*#p0AzQ0NB1u*CH7MKfl6 zpeTvOVKiN``WObD`@`)SP%#6<_}gfE0}H}T>b}(0ggC!Zv1{Dr_xV&uCllED1@{N< zz-Rvz!_Rnh1CBrEhS~Ef%hvVBzdQF6-`r|$)5p|NH4Vgb&IVN+QfvYJq?Dn=AJ|R< z3SltIvK|+}_MuzEhykS+Efs^BS8c%UYpmvOXBVl^WKACFp6f31T7zuKNc}Ou$lW!H z&EzrM;8d-Y(qkpp%a*lM9O_)$pzgr7^+U(6Bd~BeWAepSx#`f$`V*DeHzI)5iR5eB z$$R21HcY>A?0(+%MQ23HgR?t57q(~h`t(P^m%Jnpnj=TUcHy$_R?}ZI$)a*0QGEDu zG_Gd)e0`dB*%uaC6kZlpI<+znPn0#>HhJu3b78E z&6A5bkW?y3Px|~4ydG-ER3Q4XHSHKFGKK!H5Oz;1qd`^@U{|mcXBCM0;8PN;wxtVcDo!h6^jEh6MP0N+Z zV6Trl!MtJ0!wmFH+~wJHEF%(zVafkq=Q?PTJOIxKLNqLGP%RH-IMux`T~vzv;K6h& zob(-(#xh@IM{!eMme>|kO>MIC>f|CXa85Z2zsy?(4&*6Ck)cSP67la0h6=EbxTdUx zqi8sZh2;r)ZKAd%)1=L|4J2(TGiV4N7c3NXpdoacblD%Cqg~VQWXf{n*!<&P+EeX1 zXwFr-rPl1k$5+2sjJ5J_lYK2k=Tf$Gd;%>cjn735K%|os^Pu4@?Llpxt7OaF{dc@Pil3P6tG5Mzv7A|Tk zBVcEp@+)GXQ=E#V&-b<(G-OH(Fb?8T();;iX0vC?Js2g@XVnxZ-SC!45Fz_Lm8y(4 zNxpWRd-2fnqwYG%L&=V2C|B4hlY!I0MI$pPH`fp3gXE89w_jTY&@`5XR0>*(T-BA0 z=vc&<0MM9EaMGLoeK0ggai`(!4Oj5vA)s^(9=6>aQL~Xrxw-b|*`jjq$<>F%+Mf6+eoD{noJz6_atVYQwP zKcbaN3{A``I9}eIvAuvB_KBpPGMDl$g}&Dv`~>%Q$;ra6Z|_pm0Y?o<`azZa#l>p@ z9_Lq%)IBrBR0b|QWxpq~jh$636){zbwdpiNTrtHE7hA8aR=i6G?9`p)n_GihJ{A~| zNcG9DRlkCQ3d^WknHyQAx>pbl))#;>!7|8s6OMlmcHQ{~kb**u|60Tf@Mbv!Zeb`u z1ZM{%lI2RJ8~xaETnBODU&4W0Asdu1Ck87xJw@gKWN`Alg@>}R6Jn4DyYE;P&2<17 zt6-W#RYwLybH~W_Z}fdeZHp|ok{4GoZz}lne*Dk3Z9l=!kT(zOq96LRqV{Iwod?qG zH~Oq~k_-8#df5up;>A)8+B>Ju29-uy= zoJC{R-1s>T`2l!L0Gt)yz$@(68{M_SUPX6oJa?kqUbixPfB5W^aA+2!(70-qV|>TQ9U}1>YbbxW zk$9@C@$koK4PU>Kdwb^Qj72)7Q>~G8wzhp_?h6P?p zOLW9rgX{CVP@hlWoG9)7@sF~0bFE?e_Q|DpjOA%quP;~h$4^=V{zZ$4=YRRoIn%2d z0Hb-~lk)*f`zxSDPhM24+x;rJ_+!Z2?kBf4*I$|j<{ENP|6aZG^vRzpXP?}!#5Sj2 zvulYO)gC>hYX*oCLRTH}ANM}$j|%9@uSm@5NcnM=n=&=f&(ozKMj?gRbW3blni+Ct zi$Z&D637-sF z-xj>E`S`dPX)Tl~fV?!WZRZM%Mh^I|ui!cuGc+L_ZUu1Vs9=*E1x_+h>nr(#P8oI= zpro%WmwZmIT)&MPd9BVVabga!5w?kSdU4_E-%qzF{MsBJ6tgdE$36Yk=GNLL^L}5X zxXL`IdKW|2n7Y&y=TzNQ^5XOub1t_&s3ea^&czZ%@{TRV>VRbuCdOOTnBWn0GRk z#GFGuc<=Y$JpXXrg4BKaW({>Ws?SHdL04}$;=f&Hmt|2LYCkoRl6kocrvIj`=BF}_yvEB?fKZQi z=;BemMMuQ~!2)(N+&oU!=@^O`_*5IPH-d>8T|#;k+wIzo1so&~^-axX%_rtC%Q@3k z-(vV}L_PX<+3KlxIB*-G*m{h8GAiqJ!#g@fdaS!xNF?0+NiK#r!y}ipH9DGjJn5)R|>}uB3|{U zzOAQ5b@`S%0x?}g^8vOnau`anof`w|8}V7YyggRFORA;fV7?E0J$A9<| z5WArE0JA;gv$Nl`8doM|vVl!GmKj#4;Puh6JVOOA9^v^3!fRlQKnaBy>mms z?TYzJ4aY`yy`o9DMhP-*!|jUqGk28cEYsRw7+kW?F4!3bE7=wC%M~Fu9G-bf@{>Gg z8+mkTrr}OIBh$HJCh*=R$Y%Hn*rnS*l#>+|PJh@JHq_B!iSyBJ3D4+^rk)bp_yO+_XU{gzc#^_@%x^5`=V0wP z@!-BDq4NU7H<=Mk5=R~%lR}($3l#{R+kq-9y5Ds^O1nV;P>t~98WcduuCVl{?gV2z z!Q|gUl+!%kT1*6a_QKyiN;ft%R71?Lz$|MJ=$HLDlQnC-ipqSBm9RV;s3i~!?r#QN zbq^k@e9;{SS#U6TTwf@`l(WK??nR;A^-?w8XXp^FC^i2JDlCRBbmgso1~Thkv2UH2Z7G?EZ7KOKe7Yu~V*}#IhQxeY>e$Ny z2rZ!RVAh_FL<%5e9S4@5+pvbxbZ2$4e^%8khbXF8ImYO_-!kjp>Qx_iErlu^mV0F> zqGCrUvb~TvL6LlQc`ge4*w-#t>H7_gWOWyEZT%Dt?xE*&gQiXX|vvyhY^J#hR z>TK7RHIG-20QQd6Vl@-Kd8P@t#?BkNcEpAy_5?Vf2Bhzp@46ww#@48P@q0?sz>a}N z&{;vfE(Q7kbZmWpBXHMNB5rmv@}-)Tp;#S6St)r5lng{de|$gcEkD6k+AlssUdszO z_PkDjIHFj(IXvngbepiOb-DGcu|x8Pb%>g*jKmWLr-0*GL)3QN{s){CATd#Yi)+t4 z4;)s%vohu;{^nZueNye7$%~BfvCL=22aBq%2qnMDE>ivxH+nF7@ zH)ciK*>;&K4`Hz5F0C#dgV?a5p#(a+W=+Ykl)V((s>`4OGAK0$0F^lv8h|jI@)DcLp zUFphg;NgugJGnHY@mSG8)a#&b1sH&41N2b$4X+L+Xn%D)@Pv04j5KFE@9uk7w|)c) z@bsK-<~-jsh6}k`W9w8B#v5rx<-F)?*#6VJxBk=H`?f~lj9TAl z2Oeqla%`s;?U}JZtzWvt`TYq%Ss)#7*?ycA*@N0UJFs`Qxi}N0wF+4p{`^YAqC)dG z7v7)T>SjTDmtGri8Bp1O3fuW9OmR|Z4+rJU^jkz^GpX?&>-yWrH{NFWIk0A}0&Fw? z@p|>rou-QmvojYWVT0O>R#ysf-+uLD)T1ev4auiBB#(8qx>%vQ|4G-;Q-DYHG4?%C z8MGvCON*|te_uhhJ>m~>Pi_8Nao1cBS2+GjY}+YO@HWp4a7LY@^5LGc70NG!##fDg z4UnVz?X#9UKWP6P@+FJAxo%Z4bwiUf400vN;mgwdy$LdVgYou9Q<^^JHbei&c_PfF7eE zLcg4}>0bdy9`A@(Wo>_J6|TwKq4JlxpgYmz{bKEf{qjfDUi@YEOJT>J+ZAW=7SVv` z;D>r`H=tT>+Asu^3!4$Istz84G=UxLo)H*9PTzz4z?CUM3adPy^K1mbTRIJ*NA8*{#*jc!i1P zBkjr`$R~xD$EXb9Y-&|_Fm*B9e>qp`G_rK{X@6SK&eeBP*ACoVMqA(NyBH2q%uh+! z!$*AjKZ74XzI^-5wGmN=;;s>7_d`1&URNC3S6v!)41ASoWtqQ(!~?hml9i0vH(j(^ z6E^)FprL?tH4?N0>GLlLQQdN9q5u6tS}u zP9R??qDp{_>r&ovMWbP(Y{#Qm)!Y9{PqI?d}Vz!ivw%kP?oE z$)C7C%CU^2OxH~wrUKt7eg4IE5D)}tn><=b|FtmkB^ImW#gCvG69KkUz9n`K_Ux+V#VljO4`MAa7jNddF!tjOZ^M}uH_Vn#-}-I1GGenkyzLh8;reT( zw2qlw75BJ%oqGr$zh5ltpx4>K+reUx)a2s1cFn^~P%R!Gtv|JnANvX#gN zX1BdZx4XH>FvNDA+v9sb*9aJfF!U!RhM74;Xod9-(S2Vu<;ROv5y6^IH8I=d865d) zMr=kOQ(eA4PBXcQx}&U+s2{N?F^%z?l%;yd_&+xJT6`@XFM3K7xWoG>%8l%t-Sb7P zB38#YxAM2I&%}G2j$>=X14#iZ$}_r#d-*On2JLJW^+>dD=Kx2SJVyh#}o;YXm$G*8JA3|~!HrwD+ z9V6Pr1>TN%By94|)(S5-8k4jfuaMYgtHs~BX@0=an$MZ+)Ks@PQhljh&6b#a`qM`w zotqN@tP}^EYB~TeL`MGLq zf%Uqv-0|aKgTR6Dp*_YoydjfDm-A*j)`x5`yDcT=Zb(AkhYAd68p78;I;u0^+(UFU z4Ph|0g;itKx(Y6&;(EA&s*?o_`_s$IeRHnHpBl=xX+E`IiAG^G0)tD% zF;S9PZ31p$oBvGb0i=1_RDs&iRw5m=8?Y`~Hud?`c~Q%BV{gmZ{+j}4%1AM0pXJp2 z0?*o>v0P`;@l?ighHJh2uo4;3^$;4@Fd+FGNy9pQQl_e~6-3_hHWL(5n+O72isZdq za1Xj7`i%fy!ik8TZIUF8r3PPZFck=xM1Nzo+LHp>TP7TuyEI2yfr22TseGi#xj zkT516T6wafmNZ9zGEI}z)ng0mZVgNmXD@_fwmQxHx&GgAJW9oDU%a0S;Lm6EUzF8h zmz$m$Exza9E##Rc$@f#v3nr2?cQi(r4*bCGs+$-UI6s;GglE709-Wz10TnBLsY0^5 zRIyXkBamRG@~m+MorJBU<+1eNnx?vRI0F*tl^w~@axNBYzUPOz3i_|kuPqKhSp2sj z;(H6$uTVdslQiDImwhxwx22-d9qVH_D{Tl;BhmDGFl|4(6C(h$PWzU;^qyIo|IxX_kGqGhCs<}oIU(C^>1z(gW_b?8%4tmUSGJ;Jba9l4NmiTN{^K|emygJ| zQFc{yJvQ?Z1j~t49Zje9Si3N2K+9#RNzLS!;GMrq>H0l|JbG0xydh%EAzW4Z+s?3VsFr%N*qHa(eFSzs_<9oH@?gYriI}D{DYpKQK>G2g06VV;D_z+vOrQ@uH~>Krtrb}CVMo3O}`IDn|- z=h&7L`)-qBkyO=n=c@UsMCU+>h&mE0m~u@V<*Sj|BpHp`5@h~c_w$@q{uIY_6iR&= z6z=@5v=KWxT51;S+I5~y8J&UQr0xV{z16g+BG?}~3LinrN5Xq092RDcg2 zG@IhUNy2Ve*U0VnJN~#^CV6-X5px+@>(e{2%YDE%9^Vg&`;4hVlMS3}x#C+~dM)h! z;yP#`*DCnnIK@L%%yraB*@0Y}zDP~;R0z2bw8iniU;W@lyFy>vwr7zw_IbAblXOUy zEzypQ9p6vgfG5e8LW%+$n$@(sT+XokVW{EKqdfo53&EaQ2{Kbq^=Fo2W z9Yt)RmwPF$L*quiM4l<{pPe4;F>7H>MX1Q8~9vF0szL($7g|$_*zVL^qe9(hAgvjz9=3_1pNXWoIqQ>UQ@Jc22 zJHbZJlKp&=d4LNs=%W6JEc&?Dn(5*_G1e_7R^a!D#8)(;YzW<}+k!D0ACrN(+b2Gl zQpkUQaCu6RzQ-O~u!Z3oZ-7mO zx+S&|e04wbBI+#S+)`?18FveMKzys7mxD}}_Q8?0{msSin@E*+V)w&BbN}4==OFxl zB?*)2s~fyB!#@}qwcTuFfy?+|OoZoQ#B8XKaMsszoh|e8aCsN(v6=RUr5Uxq7EE>v zWuS6+X2!5U@{{`3ZU)xutkr#SS%A#!_nJG|eCiH8AX{{b?7dpHpJzzi_pmie6yX4O zZnoVSj4DgX(3J?e6laL-m`&kBMt?mp6@UUAn4N8-NaOSc$3`3{O>5ogY<8NqQ8fL~ z-NhA=Xw2^s3U?VJvGaWexp_2ibH|iI@j2;HjXs7G(1CO!i&Wa=PY?rP#7fad5j_im z#izn9Lr#}BgepH(2C->gzJFf+&%yuiC3@(OXy7`X2Q$ID;36XLr&xr{ZnNt2Q8 z&{1F(Oh6tmYJ2sSa4J@i8Pn(y)JMTJm65rM#2ntL=HwD3O?I#zaF{e80H?gZ zrtPcE(f&@DFW^#=d5BAnZY)(@1NH)kv0>Hdj_=RXEC!>8z39a6DgGlJm*WeIR}BNf z5=0r<*>NT05WDa)%{y~-l~+t709GUZt6N#IX0>JT9x~4jB=XYKPcbKXscAc!EJZBRwydINo*HfgYSt zJyJY17?e7QNS*JWyM%A6IH|A9h(8v4C>9CKWZLOlo9o$5EwfCfn)kU#n(wBKy0auB z0hGi!Be3Z->t91b%jVMz0Gi79mJ(0fKemMlXJvPaxvD3DVcx4Fly!V&a|o0RZc+zS z`nDWEtd^JKT;ls>mJ|=~tJyg^W4>EQx7_Wsrr83)yli8qcM2&1PJV`ZApP=^zA+ic zBxF=1SxicqXY>fmBElmnO}PAzT3{fnu`2;%o2g0JJx&&F%k} z?STN_g5QSU`6*gzl$vtA@~}4xevRlqb$Bu|nIp$*K1CP75Z*MzkV~KX@4HQJ`GAu< zAEr9Jp>t3-WN^W57jJOi%q;CGeckv!?K{(A_RIs4wg?1JHS`mNA$tQpksZ{li?-Ca!Q)`^Vd{ykgO3X1v-PCcT8hwBD4?rtuPRufr8z!V27OuI-vYD z8Px9*FJO`d!aei@$#LhlLIbroTa5(G06OWbO}>!mcI`{xoX%}^=KnT(G&{+dbai}8 z;sFQQ9|J`C=y6dw@;5YywEdd$RjhARjHr}g zP5sz=fEL1L#YF+m>V zx{;iqv;a`#xTj}~MB`__8q5ZLs+7=FbR(4Q87?`(I*7=t1m$YeSGG=2?sO;{G9ou5 zOFyHPf|*%rNtVE7TzM66IgO#WpL^f5Z=Sz^Cm&AS3zP7-+c;w^1%VAD*DNG6RTp^ zju2%Nhf}OMDczct&2I~mKVy-eA|MYikmlYrsI-)(4$S1L4q!%xA%0BFwJbWOdCq$( zH6t4TcCIHlmu~s{P~Bq`PVN#{o-DiJO$(wF@CvGV{`Rsn z6U@<#*S@Y;7$ng72w2LFaXXDkqcFN-K`bR|m59*7T+jn#0|5lRiE>f%Vkn#XS!Jsg zTZO&P2fo&PdmOi9--x;ye;tl3+$ zN2+ocwH!cN%+7wsJu8N&vv`Dn1)7$|Cac|&#G|Fjn{I+v&eoheq z(HN{tHkgMrw3Ju-jzn**X$TdIRy4d~4*{a)hRk2mUb>MNEEygK+krwU`qJ%KekIt` zLLZ=~a@uw>U^0xSW!E#WnQ|CJ8U$+ZK3!@R`Dt|zTVAp)efp6kqaw~ASrl()`K+C)@GXj0 zzXpC9<_1g%^G9uKXy(*;wGn4VDo2?rG@zV-{La3bs6gf2D$p6H>$Gl~6&%nNKXGr^ zm9Fa^0)?w5wK~AO40PO)(XX7a1ujewxAx+&X*IZ$pf?P4o5=Fwi&CfgoM}IO2%Q*uHNDUhkZX3C zEOZE*Q<+e#eMHZYaLawbkwn56%3n>yyq_sP;=%}C&JbdLCihp*fb0k}AExR{W7?Y9 za_Oe{XlEo`3W?zL6apXD@I}s_R77}L%t|@YkoX44QXX|=)_6}Hx~hJLSWXq364e0( zp>F{DIHVC+3@P@Vgxw7qu2_jcp9>|j?+J8U`QQM6w_8Ju!)DKbTDBf4a=TA+6}5&X zGK64C2i^GtH{Ht$=zBp4mQ>&K0N^PXFf3VJ#IZ0iEl*2fQ22gUxRa8@sFL%@g0JNH zfU?t=bywxIswR^jF+wGO|DU++Qh@xRchUq*`US&Rzim7lZ(MM#o*1Z5{f6P@bV`6) zreRB)IcTVc`4;to!c-fMopF z2GiukRM8IgMcCv-V+53;lKQzI<#!&tjPG)>RePu0#USu0cxWbZoh?4)5wC4C(a_c2 z<8x|eY0B&NG-hDL)KGQ=mG2*g`m+mvj>7+17EBKQ`W-v@?-Pjaz#FuW+&{eg5hOSjC zR(m)%H1Phx!^}{tH4MeOj0AWC=wHsmjJv*`lOt4yQ?M z&XiV^*5k~PuJ#x(W)=1*I5u=&prdh*bWR?0VWDjxVMkTeaHP`b&kO!Jr2kI|O*7!- z(x3j@%cXpc_&Lzw4(-?ULoT&;J6VQsvAe;u1l@?JAU`z8e*{!XRY*MKP9L`An~5+ z)#9iFk_-BCmmO50Nft4oIpiDhMq6*76g|DV(bvLdwcX)vjK8JSy@ZoxI&z0ZoH1ie zA@4%3PcMOXeuWbGo0uU)ujX7E-ELp6=H)W8!HnQseB%l^SFNnl&_aF^{T1HV)CRUJ zI*eMOnAm0b25dMdwhweMm2Yg7dy|2S0Vf;dBLBy85iB&(iWYtA6vkfSf=3q$4-w5WO#VcWTOOpg4a;e40^ru_u5Po1v)dgiHg` zdq#y;8AQgs>$=(6!~(b|E;sXTen@PBFhfGcK{i%+)MdVA?kxBBQ-PY$+*+4S3E>?9 zD~66P@7{ryW!rJC>0r!f>eh5)#OSwaxu#~RE6QiTGN=&dJQZS`kOG{Xw)M=ZG=Vd; zZ0-0&rLrr1di#9O6Y|;4Un3;7{qdf2zfH^oUZ%GCsd2oqJaL#gLgorO@=A(WVhy>o zJ;bX{Xa1fU>x8oshTL{X$AFHdn_VU`J?}GpB^I$ZfNJ#HTMB<@_`m7(O5j+~by05Yi>* zm+Umm{nmm7{BrKT9R>Nl9al)2z{)qdHe0FI81VP5jyutZRdu)roCerQ#jwthXa=%f z_fEvW_diidADbf0vc?f&ZT)M~eVQk|0d*Z2g@?nQ-fDGD&gqHZ+{K)hW(-9nZ~{{G ztQU=4wxLjNP+8>&MxK0+w^;`v8x<9BL(UM1^8-TBr+0tQlD2A^6XxqlXd|vUf$!eG ze|of2fPh`K-)5WV4VNuX*~gJt63rC4@+Pe_k$M-AYd2R|V5KPE(rw4Mfk9qhrz8Px zf+tWL`)3T+;T$rsO>g7!bW@~!oWH;$hl`2%H`~M(2m}$9=v7Z(yS)!%(RsP$H0o~H zg)%GisN?3ESJzG?Zi`z+M^Ej*X^r=yfe~|YI}Y#Y>EIj`QtlS9vMuMuQ;rldtODfm z$Wdl9VQ5z9L_0NH`3Lg;2s&#|`^okybSQ`s>(oQPHL?4jopxhA4Gv+XS^qj6{Pq4d zq$q6f)P%9`R5Z#--Y?g;6;e@HZ}rB%xf@Gzhdc%zN=8Gc+mMe<#Vdu2wi?>wqs!sC z(W$PE!5$KnrOsE31D+m+bq^pJi={@9mtd|f7vv5AJyLEPHi(4*+oE3(#~P|9rakJX zKp-9JCM4SQb@ZGesaQ>gBv2Ac^KVd>VDIAE`!LBpnh4KuA^p$a|NGzWcpBGyF?D>Gdqks*%j>Vt zYN}xA+t{&%l<^UXwhp`&PnftbV(utLV==m16W4Gmyyk<3lol=OFT>GW&I;-}UrGav@7al;V1fyrsi+wLS z*5H`}P_zBTPKCw5wO*M(GS@dB$Xw+1C8`B47soj7)je{v51jU(;ytCi9R9Acc`#cs zdDPIQL&#A(6uZLD;9t`bgmaFBm&N|2)cW9tldc*bUAPf3A3w9+7B@YahP8;Uq_j;< z?0T&HtKZhViAp9ELFzEoLGK~?=Axlkb<=Hx6q(-Sc5(#oAsuf=0^~n(eIc1ut~0K_pDocde0Mf#u8(#Px+v80$juY7(B#jolbEN15I5dHp zi-Z@~65zTvLsq^?rm!+8JKE%Y1b`apWP_>itr(CLO+7LN=s|HSeE1H(cKm5(33npx z&{4DDod|Ah(W~@_z5K%I9)&!EMX1NHW`|I9-`WZ$@+Qu z+3rBIVKI z5Ee?vBXNh0McG!B^7NK0p2Fu&)x3A%JjvwUkA#ee8edbc_pA_r^xK=TTQb zSmgWEPH;Hku(GZkW1Cl7y^@B)V~b)qUb-pCM*Aw7_LN2asmEM(-`EJHxuSnC>u7H6 zuw?%klK%oIvy{#}tJ}y)D*zkFfb!=tP5=jZc)lp#LS@1{6%@Nt9aVLpoh3t^RFD*eEKa;!(us{Y^j2gHCH#Vux?v$@5QPr@HfQuM8}-* z5TxZ%PK4y1=oZVtqHSCsW|{puoyS<`yOuamYv%`eY^$>|*oZd>W9bz3mC{GNd;eic zL9AsDUaMg5%`^R3UBDxH#=ZjapYy56?h#FpHBlsLRmyi=G zRx!C-pZ&*bm2X{~ZNn}+Q{7rv**Qg^9X8mxS_!)iq_OAL9ovRmwhy(eZl}--JtYTiFXbwzNZUrX=lU$Z=W;2$P(i6s zfgU%yvjtR}<8gWIP78YMo60%KogvmL5I6C0&}yCY=$OblKXvsOFB&G^Vyywm$36x% zagBrM9Q(Pd=)-VmBZ!C~A5Fet zKRjDEk)@u}`mpRg3VJJ^ko(7d=b^dqWCJlY=ZLvE8+91l+!&!~@0Lff^>r3@?kI&l zb?se23XlzGhvj8cGqdKc1r8k-Qam}AL%{<;|0E}6JH*$)Ndk@*GP)5)iT<|CYb=ss z9k%z!fe}hYq;sI#@BNPPk9byl^MAi?X1ko%&HAvaFHkM9caJm($9M7&aKQ zbJ?lDY399hLWY=h?lOWXGP`BV{w1^#!>0wKDyo+WXRbhC;et__Md8ugfz``kXtJnB zehZ{|pp!)UIBN8}bs#JU76XQFD;R-RAe2+McB`;}0wFTZju5{xxY0;esqZ{KK#Kk= z4a+s$qe31|bmr!>hX0I0za|5=c9t90{&2n|M6Y}Fi_0aWf(=2i25$3=leMzltHzho zp8;RRdBTE^+fgF$us}-T;@l(|*MTIq^F^+$q2dW|ST!T`e4@DLa!{9l0GS zkTvM6-Xg7A4O@B>Y4&SDmV&%lhBv^3@9)ygssgy%?z?jgO}tW%OqhLbhe% z&kd*kbutX5nRG;dfi%2++mDarJbXKSMrVb+!Aqav+T(lmn%hg0t_{6RcfPdy zYPOG(l&Lz4fBg+rdjOIAplixT+Lx!7m-nOPJ!XY!ZG-T&jrd7053ER02f{U&qSf6^ zLE+2CcHRR@W>Dxm-2Kan*1-DuK97sSgV7c_`uRT9@a5b2=jpX;o6qdyf@ybP(a^xc z7uN*z^Y=%CS_XU}s_pdrPj=%Lzds4M?L(!^f9(K^lZH#W(fqB;djg2D#4cI?cD88L z7`f7{YEsV&-9lwCzqs}`+c>b%cF=ctLmqt6;_V@sA_ z(M`-}r0l!1Wb9dBslAGT9KRUQ?tn0r=6lG~6!U0}wC^%U^Ng%P;^CP)N;053Eq@nE z+HUKY6t1TxcPB1K?O)M{9B@=L=ak24W6s?ty%1f7QWL_Pf`Q6&yPbY8#U07oh62Y2 zv`1cJ8PIUvY~T0+)6blhNZ3^ECB~-oY3S*!%Cyg*3t1i@2ogc{UHRiQ)R%N3B*HL<%y}n zhgBZoK(NBopgNn+1=1^giiR>Ej9#SkR8i_7uD33+Oz40j0y$>t%QrIsq{r196n!E2 zmpqj1(Vzz@P`M#5zw*dd){2N*p9N^!?`{U!X$EKDje(F4&8&c?8R%o-Zn68%(^ z1ZItsKr44s?*Zy<$2TNN-_JDAzhqp|WqUig+1abiNB6eHEsU~9aY}GjEZL+q%M$hpvnN5Rkk9j=!@nO*$8=D6n_i`=Yl8@@8>t&|Cv0{Fu z+}LAFZdg?bgoE}BO=AEjwjjA%kr$R&a1`ZaS&~xZnfiHWZwFA(5a`J)XtJ5!&za#; zWIy*j^h*7W|L2ZBhva`-7+%I&7-xX@vhqELO8`F)JMv_1`xDsdn_N$K+iiITE$QVe z>8_^I{wXC-xpC?IY$%3NuEgJM#pzXMMVnyHSmoDd&S#QEgdb@n(^N^9I<_q$M2#HQ zw`HcUjo8s&a*OIsjLDEPR`kd(I;6T?)zZ<*Pc;$vaN?5~$zT4m;F|xpyZ#kml(|)V zj?gVnSYschocxa7Pe^hsS7Jzn4Q5>}$!JE!AYQ8VHySL|a3Mk?$18xC9$V;uS(VsTMYNTpR+v@w{#;wtXV@RHIw7H4EC37ujUaHLk10uLdjQ(A78emI} zyaBpU^sWnkz%*Pgz<~0P%lHuM08F~^-Gheo$B(btVk+p)wYG#iZtNv{{D!C{MG4St zv`B{NN}UUQ{t#ghF(^}3PrYF0v*>uwevY{Hj5#%29eajX8)wPzyzY$j)%BQ>-Vu+8 zy&`y(emv1RljkZio{dI$ongTfU(%IX7;{1u38n%WaTd>`la4;}fi?G%juIO@wNr^S zkJF^&kP05&*ct>-H+9@=CwAiD9jkWu72!J8R}RIH9gRT7k0ow8-AK)yeG<)>*Jjeq zodP~N|3=3iZu*-GqWj9MlsidBrP*l!a;MXz=4~BNej3p~Q|Zi7^aAuMgNxhEHJb;SJ?(KW)&saQ^z0JKLW;-)0 z=V%F1pxQ*9`(5{tWa%-R)~b%-ndGqu3<|18U?7iioMY;(HjY^&6JiAyz5YF;I`pyV2o z+?dKpwp(5wUl;h|J-!a)dyT0)9GhGBV&hkT9P!R`wu0n5%h)q{Ck=NuAmMH1P~xW% zXWv!T{~>X{>o1sQw7Sxh-?RW7=t^k>iE3inoKa_au%PF}3i|P)zW{&!-vxeK{!QeC F{{x$|rU3u| diff --git a/examples/codex/images/github-pr-settings.png b/examples/codex/images/github-pr-settings.png deleted file mode 100644 index 723e267179b5bd3d1268d12bf562d210a5c8316b..0000000000000000000000000000000000000000 GIT binary patch literal 0 HcmV?d00001 literal 203329 zcmeFZcQjnx`#wwrK?IQyM2Q*^MDHabdI=In??fF8#%Mz%L?i@3^hlJ5(L19>i!!>= zTXdsE8*O-xr+mKO_5N0#=l9S1$Gg^N)|#0!=bU}^KKtx_@B6y$>z?o@TFT_4^rU!r zc;u=oin@4s#7I0m0-GyDz#ew&Gb|q7RSgFPg(s>C3M@}Nz;+JKws?3d;c*7UAiWNn zG-KU6m#-+wtF35KMqig#yTTi^$NE}L<UBx5k65yi^6M``4{ok|_ii@|_MO!sYe^_={ZJ2c_j}LIjmloKRs&5Y{VAMb@}BJ1 zL%eOy;HUA#csiu8-CUS1hZ%Ws@cdtr(sYENAlUk2%P#*qmw#xjc#O?Y-6V z4RlN@A%8z%X*=gUef4mLLrhuVL6`_)EwSYNE!*w~VC@pFFz#Z>7G`z_s%T0lf}o#P zmg#lZg@kV2`E}24NQq_ndkQ{hi}iv;F)~nA=?BG?kWtE?o~l}kJ`I|dHR37`1iNAJ z;k!QWl$eJm95KXg2NNmpH6!UhV(n_)cC9X_@7e~)-raBDE}56O7oZ`f*Mb>ozfC4M zfBjDOqucNKwO$8-2!39Hc{_s|G}Rt*idi)(Wxv*Gv%)XdU?({+ShdbtE&22@@FcZ* zCwR1&I?>8ZQLa8IfsXcb=gkAf#;slEwijLz>?w!%jM4rkhw}Fy>BxV&=bdstrltLx zkBN5DGV>SuhzV85=i?1`xQ$O8@;3O1p8kKU!I=L z`aKk%eSz0_i4-*0X!D-1CEs-4bX#kf`tqSXkzmjj@jW2|w(BRp?A!EYq>bs6;X(dx zDRA?`D+>x|7)t$)aAJw!Vf;)imit*rliq8{!&^Usv2SEKIzADT%#zr&oNvFT$ir)U zPmot~KC?@(4hx&lKlytl)HCrZ?#5Q33+CQriNlZy%8LdTT)(h8n>vg4v|q< zUOsytaL2hh)W|BLh+tG=o!uMX>H8W!nN*~R1@EX12a}1r_ObqCqq9K3nT@z#v&8PE2t9y@!6S$J%b0*(Tb_}cvx+~c+MG!RHVx`e$b-Uf<;qBo~D1U_0shgfw z#0D5p+I2x)&$jSPgkJ`)uyMZTu!<6Ab8aCfQT!f# zi=9OEu^=VMXL&o;-RSE!9J-Ot55GkR{IFbL_rP~h*kqxI)>;tRA!&W$^^{oTQGQ?a zU53R^md|fX;-lnqJ`t{5^5ixO)&E@EC%MP&OHvnH@rA#S>8gd}Eo!0G+YdzBT3L&2 z?lZ<(mKIwfZsfH%7Aqa56k2Y8Xb-5keqK#^CDnNI_KWuUOa7#?4CXHq=AZd62GTc# z%C*$~0$-+YdT*h<&AxqQ{tDg8+?TA6xGkmTDCe$Dsx!WXDzZkamT}fJ*L_s%X6<3& zyW`5Tc_)pff{K~iPlY{3wj*sp^9}niRahqLfaCz*fcwBRUE);&!*|Y)yg7-ZV>_;U z+^PP6jP-x(A9wopH1V@)ai)(hEBB*L&rZQkQ8l(j!K{#-#BYzwvwijEzZwp~h6EjX z_pk0V?Y~P(N=i%$Gb&30xfB>Bj+>S5I#LYFesjv3$b}5+4c%N*ABKKSf70;fEL7=> zVgKp#jxyTsN`lvD!vO>K|xTkxJe{AcU z!1;JkEjF)XBqiT40|wfSw~L}taQdL1sb7G;o}X$&TbAY&Tb5r&Q2uJ1tgO;89her4 zx0bH*DKviU9c>eDUg^8D2EuH6mUyapdSlAgJTZ=#nib9!t-<_0nQs$4Eg$tYn1sTr z-&Uu?Vvig?_CM%(&|lQr>=-&5=r%=69w2=V&*mEFK8JdT3WN$>t)#EGb#RN0fsjG<9hxpxa81yT z{)En+u259gZqniVgI@wv$p}#i*BxT@gjUPNqE1kb;Z$~L_ML3fLEh~A>~s_Y^%F&k zI`}vxmXlP98blc+izP#iKZ4$te{|5X{*}zp8sn1aP>tS6S@h#^%2UjHrN6J=o4=2M zp$X9EIaZ0Puj`4!`cPAaHgQZccy zv)xmk^mqcD6sJ+>RKyk(q6;o>EexBTiWD=A_$|}1EIx;R-M^mi9ImV`B`#Ob?DF#y7djWFn)SAWqNv`Di7pjaCW_7Y zML=sQy10hzOSNkYbNOc~2e8rU)#-iU+8LopUapAJh~0si4)Z1Wk8_!kSg8mPCj()EyaBbe@Yvg4^&U7hZxuCrP&S@uiRn3Z}a#EZyJ?$W8RWiSWQ?DsSIN_ z6Agnt)4|+7CAG*|i*1 zI61i3!P=X2%;JHl!^j&WA3lFbrbcjSB)eIDZfZ#hSLW*H`o+bl!jZ|Dy{Uow9<^Yw z<2T^rfElbIG-)zwBFjj3D=*Sp;5Fj*jo(bI7sSCrp>5yZ_Grv>4|MYh7?|({#@&o5 z=f(th+Z;r$Pp7n`$W$*(4ox;+cTufa*t41M804%}-N<@8aKN3Wht^n)f7MaZMRjx; zrjT~+v>&W9dN)v4x*OY3_~ZP6h~-+agV7-Spi%aOrnxlE@ZpmtexYr;7j!5gP5YkY zBR3j*bo|(l^|CP(q5vnXb8Sn$;-x4gt+asHXL!EnA^KA#!R8+6?9Z^Q*G*GB`J{?A!e*waKK0{nd(> zF3W+I=el)M7JeNEg;MNN<9Sx!v)ANMwep)Kpo~hm9M^!N0-6_3!EmJwhfSP@c(ZpT*ke)Qj zb_wVy5_^0>8Rdat2~{Nok{hJaw|NnU&0L#hf>^wSKgHjT-$m7uSdr}AqM^F zM(_f-!|u+kc_6&pGYYfm!{Il4;@iQ16O)>*LAq>)ufewYuv5#}R^~mqrvWx_X)O1z z!O~v6hWGU!0?KjYhrH_(b<>m6X58j&TPHUggLa?**T1$?{^Hp0LpXCjadM;9Q851W)C*r>0*-Is9YS}`3$_B_#T#3Ys-31L-aTOZ3f^UW zdOSj43mE{`FX7>ZIpAIX>l`iM`{MN;cwdzH^P3508f9;*GDW z%f1a>TiNpYxVT+h2T$5Z64-RH^|WO1adCEqNczaI{&9vR zuzj(cpOxj0Q#_qySV5XkSQNk>wk)E25BMIi%966Mut*_o-(Yg7Zv^I^=CV6eH{L$CRfN`w*}lF|HTo00lo+P|G73$RQh7C2N$5vfI4KwMa8B6IRF1R@;^2HuacnuDJdW!{@{NX{a=Uv-$kE6Y&{gfE;Txclcfu#MkzS2JkZ5f4uuPgU{ZQy={GDPrVM%Xs6Q!NZ4J_FIhi4W7Pv^YuoS zO5uwjmMdz7G;e}l2eD{g&QyuztfP%KBh*PEeR1RYfyL^|fums`#<3|vWd}A^h z{=Z$5Ql5o_@QP2)g2R`-f4wH&rOU`_qW_zg>IMZt)zU;s-~DqVFOCAubo$p@|8Fz@ z+nfJ3l>R$5|HXv-cc%Z>ApNtA|L+v~?-cr759t1PP5-Zl^iMA5pN^dJY&$dj!&bw2 zQ~8fTHE#)A*&w@m{QXguAo&-U;JE#S?nEhbG$NoCC%xF|l`rh`sXhfdT}|S8XTK!i z;AmOU@;m#H+lZ^g$$Dv~XKsvcf%EE+6s6kR-!A|1q-2!ak?uSC1xCDbxIM_zNWGMx z=N+|bidkVCr@z{nOME&6F;-KNuKfKELsaEXw;R5$bESI5eBjl)cMk{B9&La(cy9Hp z3xjBGkCz8L^)PXVJ&;gXI>$Dgcf3t3wLNx7?cfRz>Mx)lu1ln>L=u1e=VZ#u`ApU1>?(RL6nr*CP?@?}Ferg^hEpT3oh z4>PKKugvVq7^ijVANun)9sZ^1I|cBwqa2d}m(BNCtWp&&ROCzSg3bZ+nz?3!Oq2oc zr(_<=(ciUPzP7nQ~dSzJ23 z%7aV!Hs@yq0%FwCV%QPzU5xn@Ww};v42R9oYfGI~c9v~zDxa-7Y?euDZGPCqY?JRm z7K95^@>I)t?3l{1%qMrBQnapOv#MasRp$A1s^3`*N^vy+FjzN=IiLsf@twN_y>>2F z<0_0@B0dj|FB`siOq&XW}14pz9dyVSrQRT0qLsq;xbFj2rO#>?;?7tMWLjf2f8dzqIu zXR;tMdDegEUiSO)jZ#FznH_pr>LWP_iSbImr$)j~R;Dx>2*cU}=V_z}(2C{GeXt?Y zt=`MT4=Y1uXQy^AL{jZ|R9Qb!>@0Tqli}}?obR~?`tb??=yMro4_0qSNCaln# zJ?ZD1l?`+0@T13)P=Aernl-hkC{C0NDJSJ?%EQgtt*q@-4n%ss{qaG}8usbCA8_ez zX*sJ8(t~h|p$wD2v8)pR)h>^n4lZS0wts?=f@T2LyE+&`OraJ$5dk{!MW5AH))niN z*mPXCCS%2*8}98hJ-;E;Ta*!!=|3hYu)Ffz*p=tDQG8`eiOoB`_$n!JQaVAQS=_LJ zd6&zp%`&DeqFl3-2TBE1rydv&P4-zvwdPBgZ3eS>2R`pNQ(?aH%sFm zF*^Zc1Y<|BiT1WFKWMuVUM9=+ep$$0P5`$%KW@0>=ZHs=%hSoA(L*N zofptzP=VD=*j60{&|?3&SF}HzFe9LZ#Fobq2`XLasHCyi?uBVnS>E|cR1UdNuLz=lpb+Uu*D?YhW21Lz`HW>Y;YP zKBC5V5+#RYA4{qKSZaq#ntbEXo!{;W4Aa()dv~Un_ILBstJlY3vA~7RAJ;-ux@vxP z#&vtt$>G)y(|y-+R}YFd3<3f@Dhk^Z4VOpsBQ1>VW|C27<>3c2f#yODAC>dtc!Hiu$KG@_IahkL?UdVp-%_C}k z^hj%H$H%tsrm)-IS{{RTbG*33rUqs0XC!07PUoX!`7WNe+&b;h*7lJi6pvv(QUaDL`TC*q4Z zJsL^ZvF&@gv)CzFj+se_l06$Wt@pz!m-x5dbDB27qTdVkN;ey42V4_#nN;Z!pZDbF zWxF&@FK>$U$wh#w)ylXRdSy?dP7mj!Hf4JxmYu5?Vj>p5@f3dOjJu~2&HBpvpsRr2 z`fd7yex7vr=V_eR!e_GP=|-3%&7b=oyv5DOYMEbipef8r|^TZZ;wk~ z)6#dBfaY7(<2w#2XCga2P{Ri5<=>tmJ@YdPy`^Sg2Gi5UgImYmCx;Q|-{10Dd;*hFPMF5UG6g!U7YYCc@1U12H5YzgEh^`1(|BFDgmM{#r*e zSLI~jiHmMCH#Gjf;frKS@)At>_EIvqqhbTx)y}--zWddsz~nr>md7AQ?Iw?+Kg|n< zqrvL-t8}81u-wIwS=QWj1Vf=VW+qi0qm)CjUG8Kaz0V-3IEo=c*?&N`D_7K81ch5MNDt#mQV_#Q+W6P=3=E1RbrVMPp1V)F-&CHkGb z>vjjUKZn$-4DIH+dKiRTW#nD;cYS@8X~%!aqkO#BMg3`w+?>ixvb8DnIB_6BIY|_} zNLY=tCDW>Oo_@;YIK{e2hF3EkU2-ix9QJzx`V{cwaP)Tujd-?$i8K-F%Ga1#v()3t z_;^5>pzpzUa(zi}f`WAZ*6NyZIfrZq>0L{D+ND$a>Qoy%T<{J%TkfZ=iy!_juL)LPb)*M%Gb&x zeH%|gD4?qIoTymRl_QGv9;r7!J_Yx_M&I72pcn4)8_Y+uX3GqRdXxFVQe(kr6%;j~8)S=R1 z^7r)b(k7WYNmf4JcU#%sii7GCTa;g3t4Uto`>q|N-wbzOm zn>>mx)5=CHJpSFvbhu1Hxve#u(*DhIs!Ua3^$=A1HUD|(9N+~+s*M=M=cIZkmFq8m zCxjqG^wOCH)Y4YrUt#mJM3v4aTSIn{>wd?L!34VdX?}yZE%vt*oDv%bt z2>QiUa-a6KUXv~fd$e%%YMZ}nbn}bSig}W3TSf9MUgMef^jI{BtK=$7BvsBnI|Y6Q zqrz19A!57Jg!oxC?Snrv22eTa|ba^h`s zdbE{zF+VS(=*62#XSXeyIZNy|YAi}bzUG#5mgkNa)qndUdY$Vb0i-LT{`Rw4kO3K) zPy1LVwX~XF?9vyrTRe3qqsf6IIHC}9Wue)85s5DSueXH`nR`1Mkf44kn{_faHpk?i z)CT$SvW?l^B~7Q&PNOek;?E&%!%j?V3NlsQ-CLve&(0S2Iv8(HdmL|}6S$q5+sF$Q z2^UT?!WzOqk^#vs!W31IXohejKTnyVi-F^_G!rIpaW!=u8N+UjxV@WJUqUFX6 z^Wg7Y)XRW>O#JxKQdrcjjRQ7k}i?_xF*8AmC5UU@o(Lsf>{9DVkw88kr|@bfW`4cWE^bwyL}df-m(;*6xig#R3i~~8cU;|3<>7miF)&= zLLJFsQEwRoz8*jjvPj&k{{0AAo_hfG5{n414VaWKhEJtPK&o&R@RKQMiR)CJz{oI0 z|HKhHvl~P1J?ww!C(y^bCGyhl+)NVSNx^8nXsrrkG zr37wo&~~K2)cquQ`>8dAva``}aJ>6##l)d-Jlp{o9zN^*B4Xn|dG^Ra74* zQuQgAl{}k<*Q}Es2|G`9@hG8wEvkU9FnDk`x9JxHxSm%SlqF(T9=iRQRtepA0=Vg! zeo#K3yl@; zGN!(x7al!aiXUNhERFYeED5Y~fd-gu0_s=d%d8XzGLw3TpWh5>lKSr4YlU zb;|W6cCX3{pS!%Mf0XqNodDTeyj(Z((Q;@-V4xNZ7h({&c7n)*S?8b|@|DGD4qy&foJd)&hH;ek@O=8=~>z@4$?*Y2+(1}ELDz6avz#D0}@XlT%^)4al zI#B*985-fPDMsR&|%!>&Egz`i;ie9L+kXcf%kB zMC0~C_rXvgG#7M+&2+KL^?7p`8&`DIy9A(4bg9P55Ylr_*-}?h_R)7rV7#UeEe=5 zwoq7JB*Xb<>EY5!pdUlY8QB-R;@#~>^5M_cWcq&G-)U=_%@?gjp!kW2jWuhprAmgO zJN;cP;JGHbkKBZ)c8iaPhvmZ#`qwEgw~_kvrVBN@#7kHa$0z#WZ>jTDkLC^=@|K}} zalT$z6S1kh#*tc`h@i@P=}$Q7t*Ki68cMCsq-W}Ej=ST6yxvJ>;5id;#oXq-IGfp0 z9P+W{NDOaPC9F$7*D$Z3(iIWD0d|cU)RIDVeHL*B^Ltj&6jwGTjvgH_ppB}}q%8-F z0S8LKZ+M48M`SK!!tv*2@8;utIC9W>plWUo4~oE>${rbMm5Bs3b(QBOJVq{1VMfaw zo)4D6w|!1`i!V(#$rr+9Qq!cs^<@qt0ss$63w3;%Y&YmaXI1A+?fgi33}}6yrBf@5 zeb4Qkofp*^%O?}yhFGy_f1Ynuv=SrX4Q;R)`?NSJHuKk=IV5=od~vr8ZWm~7^=9In z3(a=l$aSRl#BbPeo7^o4KmY5wdQGi%vF)!hZPMIXvq*heM1PnhcI&YBs*^FDs>1J_ zUiro=l-r`$N#5pYr1GWKdPa2S20tdhmge&_e6xf|D|RT)pzvKM7BG^F-9o)?gUy!| zoAC5O@5U|iJ*HN8D{oeaV)MFW{kL2?l0@>f+HQaEEZ}iqMvWVk(tW=){g5JdMl|x~ zHkt)jNJEwN=pGWEDJ>`BDxkytY~^ zbL+=h|7xiObh=q{?W7XhnY8FJmX+YEs*uh+_fd8kg;BhKKlL*qILjzzDYZA#u&@Lu z0F(y4s*(Yg=EgmrWQ{cO>26M$$eSsCKYv9I6g|XMP51}QbLN`1r}2XZW!53r`SF}d zGXm6RKAwAH)Nh?77zx7>)}>z576)_8?W8`KYNwe^oeNB4S_@Ur6}-td(A{mlF|M*} z_YUPe1Ulg@)XlehTElsDWv1OBzTr*f{q?|Xy0VUg@p9HmYV^I}rIDnaR4H%DI@)9N zqJG=(YvapMwj7wp#HaLh**f1JrW|j=h|n$=!a=|l3NwWOP~ROK={3b`Ca&$VSR?8q zPPMBLN}hmEK&+gZm^e!&m1ya5fV%bDDT<$rhYwCxQyEug4!94E;{%)4_^&>yKK_fgU!kp&;8|vmXobEZGTG5Iw>n8O9Pa_{YqU&dj zLY1^M=XdE~W~2~kO(AG;Mcm8)LxjWJMr{PR!1qTx66Iz#u{f+@P>`Df9n3w|Hb zH7}xYj;fsLL{%!H9Yq<`z>R*bozDiU9MnrCFd*Z&KI__}Y=$W`>jm*&Xb!=U*0&JM z__H06-Pp+*I zL>7da$BiieA;8%+#Qg;a!^a&bL@CtsXi%&JTpB6&r)scoRR-w> z3%5!Zc`_Y#(K=n_p+Akm&Tp?6wU9;{)YX>gv6X^{yq!skFENh4tq{#|2?b~Zq*0Dt z0Ar5GY@OFGk;Dn2^fdX6RH`BV765I%No0 z8&RxlsLq#mm92Ul3L$ntJlgCy=SbaZ1hNY+mSDBe5$xWya&<`2KN{U}LufM_< zCC1DnUU{^?@3C3E+%oa9R(joc$gUsoj@nSv7)VT?)=wK!vlGj9JIt1p^B`a@^*G-p zyr`TY7_X@V;cpe4)oM!ApKSW&aMlJ)0H_&LQ9gFg3_tB(=gTOrU(rQ?bSf8vi|jg3 zp__#>56UP!V!6c8ySkr20@b@ctTFsxro!P!3)xg{;aNsZl9VoXY0nIrmqMdPFUcD> z8fwvKynEZh$yh?=U#(&!GyHR&Mye>=J!rn4lgfq5eMccZ0Y4;EkvIL%(2*Fc45Hd{ zW5nPDOO9bEe{@tL=(*ka+Qk80HO^n)vT_JAb{J=`X>c>1-T-gCzxT{(lt#6Pf1wD^ zx`^2W({`q%;nY|QkinULBV1N4l8%OUgvjP&w~bdfftR7t0iGGyq&L}tiY2a6{uD;$_q9mTD-{{9t3hVh1 z69Ec&@u)11EK)S$uQ^dSVp@W~#pdE7gZ~d9yzNTBvtZ|cRWx1;0Hs`nX`RWp8NYPr zsbjkqeHW1uO(J9FDGA8Z`)6IW>Hw#;fU2xWL`+v{Z!vj*-~E`)UK%{D$XM1h&FTp_ zux&c6g~Hi6x<(14H^yaId6UKV^(B( zxl55MS&_i`j=oBD@t%G(cKW2T0c3VdWO5Z#^$`h-V%=#HCZaI4n_XSA4-c%#m&lVT z$a)sLVJi^I3bO|?&QeLeK#H6sQK2~5xN{8;SI7ns8lZcZCUz1#ya364zH;gB}1D2 zb(MHqR+Lm!)46t2J=3S78OM=~dM-3Oe!I%Wrv#%X=)N=b^|9r|z}pH=Tde<1?u~{# zjUjL0jJc5g#qb`+i**7pG|JrdqtFr?LJ%p=aM{dlNE>mOdhY*<~Pgz4?#Vj8iDh3BaDtatJi>%pZU0z;$ zSLrgr2U+5MKL%McG#`K6%_u|bY$fw?zO+{^;fRc;vpcbLaj;jGf-h3ayi`dfXX$K^ z>gYPT5A!=eh;Lyn2u|z_ho3%B{yyFi@UJMD{Q)?SBrnea(k7G=q~><(FZvyb5SSNx z#0k5O3pyH(R-EFfP7DU{QiwjeC;-BP4gl(0ME#ZO9_agGVVw6&sWn{8OShjBBg=1B z&Uoi(__{kk(sZ;37P}1|%mP3$U)5_72dT?iSz0VLg=*RWJr_KrP@0|Ve*V2k-Vf_) zI(fDT{sAeKtG#T@1Ug@H(lSSeIr=5Pt@gkf)9|M-kEb8Nf~^{ z>+4Wp0tF!J1P+J}a*U9R8SN%o_q_rnAt4KOtNKp$OjGF}01^ujN{9%ok6!Vac)TxN ze{ie>>mmtqQl|ry5*EF;RO*)2@@&RpUIh=Wj5VAct5>+djVqi36sCQ4c?~N!hhGw` z8;jN5vriAGsO(oHm{y*jPMOB=$>4(tK18GKlHdNGiYNjo+=|>5)zaaY@5(y?I_}5O za=Jb0IuD+`afWNE(ZCl!g9+9S*uEFqliYc8w4$}<{>QH(u_A7c3`@O! zuiSamq|W`CgQ`tE^G{a}5^k@Po#mv%<1L<;KtBc7uVK;Q-(*~pens6?q`w%iHMt=E zh(qYjymXAz)?HTWws!|5VF6k%a9}^4Ol#JQ4&(|w4jX*hkTx;wgzjA69<WKo8>;wTR3QRfT8sjY(>kB8TcE0Mn~<8OgSP4V5)K zUco!?`+(L1AVTp@@w&CK=4jC{u3Wu&XQ)8W+-#AhzQh_C1tODK4Uk*j8kkW*kg6Dn*zj%5&;gpgI<8o2>EpA zXfOc%%}{}eoQXbq1WoIodcypkywK(CkC8lkDC{(RCE@iykZx^I($7Q6@$;Fu5R3Su? zo0XvLWZHhutBwMp!ilFpGhYqAEYC;RJcBsw{!plRY-WV8NKIJX3eQg#`!QypdJ|Y| zxU|w(wY@hiP4(+d=;%ak-g_mgrbswNN~8Ok8j*R5?m4Qv%bNO7|4f4ESgy#}M}%M~ zHtXR6Y1}s@LWpLt@Yh<;{!14Oo~G`QIMCBI^F=|ieCnC2>!KTyKTLA8ttV#TR4i4< z_Q5*a^(xwsj?XyI)(o75bqY;exM5V5pT|HivEhO4BDyE z4|f*oo02J{D&QWC#>Vga&Q?Q00=>&gPJn$5m6ggpCbG^)E zjA`iPyvr8q=X^1Onv>XJ0}sE9&*;TxTLt$CjwlPEz&<3}Ze$1r|5o4@ z^f)M}l8_^s-B=Se;VYD__2ClXdVo{%X>n4cD9jTPoRqMnjHPnj@dvm;595q&`bcf* zX&rh|@u$|~2@zIQ%z%G8NnLff9T6xGkp z)16dOA%1O{$9H>z+4y1<v>(mv_MlKI5FSP?|{| z^qbz1Z>gG&9ec|TF9ZGC~5|#CV)}{dH5U>+xn8Jny^b4rsa3i^$>u^b8 zS-=p}T|rp$B}Tg=nc>a&Y)$%(VFT0E1AN>CG8=LbT1Y%2(2W0DjNUe_^{qa|4C%4O z+X5~Ilqt`6K~c#Q@~O50xB; zO^3k^RGbs1v8La7FrZSw#sDDjR297{iJjt+NnfKW*GMjJ=r@!@RFoM$OpY3{TM)8( z4MimdJWt#z&>eG`n_UJ3BnD=N{LP_JZw=8!vV#o5uABfjB9a_=%PFYNJU#6KVp64> zl7E9@rrExAKubM}v^_Qd9QRWC@1LT|NKOW_Hs7p$F_UY^XYv&;W8N|_H{<|hQPHlC z)IQv!9bQJ(HXbP(Pdy+0oS5Rj)4|!n2(N|~N~1GM5pXXB#hOIepRC!7#5jQTsnH!R zw-Egpja@1V;PAb8O6U^%&~gE*--|tiG=aIG$q`)k7Zz#hHmq=Vpp)IJ{s`;-4lr|F zZxbPa3Thqvb0Vop_3a1}GU{7A5Tf}g8L7-Ro6k=y>^ROU49c}7!xTyj5lkCJ+8q0i zocF9RX z&aVNDBL&X}yKCkj(8h8b6u@h7?&*u3Gu!2Z9<9X8n=wGJTP)&BTNCZWpXD*1_-+@+&mIq5bRVXOY1SvnIUB)+`P{u^$Na zPB1Fi0&P$!)4QNbnuUw_a*at=-}oT?eL%Vx3a}}>pC=O;x6!Fs&*L{t^LuouRLw<>JxkIdrQf?(V;AF3#gfN7-vKdW zdYhp;W-hx$sgATE%u=OIB&@8#(x`;{h<@&F!mH?$Pt0U$kO@sslVM5$S zlV{%_->|ww5<%M&Yf@dX?{NkQhIQ+IP6;H}UNY`SCI=@J%Ks#~$krw0E2gCK2z1PQ9Jg5L)`VyA^`Ro zvha)fT@V%!S@YE$OS8>$Xu0G+lsE^THo()1 zvzuj=>2J6W%?1nCAv18qa)Xr#HmD>Kbirb)s$Qm3C6mBzi}7d)OwVEDfXb-?)Gi9> zV=$6UEf>(a{i6;q3-ZAltTOlfQ}V&ES*@1L(Ozk9NJLB$xE~Q-2ByUc>Ka`v*xglP zIoKL?PN0xyX1*8m)!%y-F9C1&TC6Hd!1!Ckfz@ApXdrI*2R=4Vc*UWgtFKT2p0PBY zJoYrQ@Dw{;ck;;NC6`lC(Xq_4W+8F)NV`U3GSTPTPhhUwMSX%q z$yRdo_+Y)N>RnouUAK3zZd zV==+>9wUm#N%;A# zJPMO3B_RBqSxG-)6!@-4YhTT`L z%E@tsUNg87FHAC8S^}a}?^(2)3eB=wbVMV&;_u%e|I|vKB<$osp7Jc$M2CpnhZFs6 zvjp6L)1tZWZ^$W8{zh)HT*~eYaQq<2Ptf(@A)2M_Ccm~`p^Rnj19<|acBy+mK^Fom zz;j$5uggJ-yO#h^J4|GTF`7F(QSk-DTmmPK1!QYWJ;szZz`v)ctL$c{*qwN zU#SefHVNCnpF4uswfILrN{$&xFmi6bkSrsqSHwc5?F;Ir&kPy=M7t#&^djiqU((CS z9`mBU-2J*cj<5$ADszuMr&rek9-bk!f9^x$mwolwBM4JkY%7rqeg{`u9 zXPT@Nlk;zRzB&RA0^&`4p2vi67aYA1gQRTCAM{6IUA@j`-bqmTIs>}U`&hQ2uSSD9 z>ZR9859s#-@*p<(#uH{wy7id`bJ_D#xf1A%!OrlAx$1%Q7>D1%y2l*hau=c2d!=M^ zxw)6rXAfDGx1PwIoI)y>>d^J)K^6h?$vegM0-uv_> z@OFozpLvB1-aQ*)o1B5(vT*xUShineVfE6z>e&3% z_`t-PXi6|U|NFlHR37|x%>7R)j0!jK5V*Z+?%w~0B$U4z9t3o!3;Qj$-%swp8=mWY##ntiU0pLbpJbC|2tg&zq7RZ!RCQSkF_LM**ju5BD^HR+Y1|;LWmdT z&W|4P)$PT634GtKYZxg>MEIZC9$C~l9dNQR3F(OjcvGf1nc^chj_vkzcqcP}XvRwI z<#cOR#x~9lCcrxvObU=enO@RPnK$bXeUk<8OvqX*Zx`|MEMz!{45GS7Wm{~a)_#8> zh+%&fc_btDA(Ze6i`#7=D^#nwLuMN#w$mfE)dA2HcYu6}_oY3ufF$RTwVDGSD6#lq zn?Lq~Je|~8=YvKKZ{^-&WOhmjtt(aYy-Y9Sa0l!%!FeHj>`oNnl}VO##9T;f(>8}k zrp~JbzOZX=G6*8?Gy7MmhhAgg1M))S>in?P6XsPGk&|vXFS`0d8__C>b4zKU#!&IKQ~_J zOZNS?AQ1;oF!8#Dn;EOt#fVkzK_Kq;V?|(oQWJSPF!|xI7z!|MJJTo3Rs3`Q^U`6a zFk7u^J_Ei#a(G@+YrXFJHP|ax3)Av8n-y$ImP?gyEbO6}c3{+!S-b%ro=FY*zSR?; zM+~n?XcC;8IR|{W`-TMk@nfCNV;YJvvQA8%XI(!h3rlUgXo@x9+lN5n) zmir~vaFCL-a1&i!)cMk_EL-t@iH71ABEBZam1|x6SjlrZ_gL!g=iZAnrfL%a3%NsA zxG7MZ9}QD89)PN3n731f1=6dZj2y1ZC$z4eQzJK`*j z;n|bxl&|kJ?mAw*P1AajUYM}r7OE|a8#>w%SIX8(b{7|TFbs&(9Y&U@DwUiqqEBAk zgVq~QE+mO;M%UN*5tTlGza;$jH17_O5%HdkA1QBU+^}?IYvYG=;LZn!otD_Csdv38 zvSYCRpG#M#`hoPb+F&glexqD>x|H6|QpbjN=iX|0AtIKTUf3NgH9JR|o$+#><_>^d$f19|54R*P zUiMnSC;MC;7T9uE%w;O+X1gNKH7deA)Cxt&g$7;d{~_uv9GVQc?{N_YkrGglZV;tW zIz+n5phH?px|xJ@cY}a{NOz1)x zmy4c--m$cJ@3wOq@~QVf9HJQ3tXEg!&!L~*Lt#ty{yrD+zIN{~K3?q<#%)$kY%X72pY16N zZC{d7x)}TK^wfjBUE8Ez-ibbpdWtgsj4H{pZ2$wFgha86Ge2{r{uf|tM@67jY{k;a!=N} z2S_X@>ATOuoM|tkBLUYA_@8^ny8b3OhlGd68$Wg;B!F0B3@PvDDXPrQ9k@)lKWjY4 zNK-XCK(^j1*!E2$wJgM7eLQ0tBQc)@@M zEski>4nemw`T6A;NU(B0k!dk$qyuP@H{)&VjGKS8+2lC=6nY;2X+O^xvA6oD#03oq z&>ugFiO%`EAOdLhlcjQ~WC5`&6Xxcu5dk3W6jr91MIDWy2iv$7w5MU89b#7Cvxe&r z+E6Tpi!VM0QwozI74kje++FBy8=HL%8$K%y>+4hMman_${NKs-V(wG{!$m#YS8nFX zLIOOFGz05scBdQNtAk?pm7E>;4`%<|v!A7ahNroWQt(v7?hoe&hZV z9EW2Wdu6ZM4{2i}-nT$$2(#WVlt4~Pv5076w^ZVOS)`I!##|V&!z3ZV zam&T7Lkw(xG5d>)Rr#r_@o>xBm854w)~NM+i}m6W?R|((rNaM={s=cKbW%U+vkx}R z&vP(ZAfwTrX2g!?W5l{Yg_v>GFM80mlCmgIdge#MFflmmV5-PzPRG_yH6;p2UC2e- zQPjFgyjbMZMoU)0KxQW1ZrcdgKYuz9jd^|rN#d7(en2g)tou%oEydB%B#N@%wx?y2 z`^ngVl|Z#O?e;u2NBauVCG~DE?i+E#7sTx&uigx6VYScxoQsW4p75uFzim-Ybyli? zY}C*(@sXD=DW%j8Op=l7;g5yB?~h?@!1b^R5%Qmh@b7cmwXNA%h4yBW1d$}`HO!;; z|I_C!Yx`f`#+GiG3A8kGK*b#1dF@j>4OTpS+@CSpj<6y6hfIpQj$4uDn{`>a-#iOI zVX>zoG}mY#KKqX=!qG}KX9x0te98KypSSMr>z=aZ_;l&K-}+5o7=_?&h1ss>^Cu?T zxAk4#l9R=il9&V6q>*14eAeZ*Ji2mOKAd#`B{?`)w#dZu5 zK3XSny>?yH&zmm0Qw?si#JJ^t)DIQ@y% z(PJ6-#nFUxcL-miw6@aSe-3)8=``)WNmm1M}3|C36O zkbB%Iez*SS)t5ms)Igr?q4lYB<#1-^ivKfxqXWwcI34z_36X|W7sQw|AKDoC*2GFC zAV2r*ziDuprrH)$a7|NyI@$bwcPD!G{#6?EVa&@>S6DvQm@XX( z2ZpmuWZ9$&$kk zfP@@v4MnXlbZA??^zzQ($s)PB1tLDYhv@d_L_Sjr?2?V6EJpwOyYGiNmb*;vIYJik zDF{bg%X$8tLOsjXhp;J)4cALQ_AOLf9lnCqjAaN(m3jFf)UM7lO#$9z>WT12$u@QR zzc2}lxuXQ7Rn}|I7r$bP8S79pO`ZfUlx9VxGHuo72{f&|4QD>2pGLhFJsH@i8}q^?aZ{TgAUp00hb|t~jJ7 zDgxvMQwG+1rx_$FXBiXr$N@(QrqrHHzWq{}ujAJAMs)j>iiab0iCa0sw#r`xC`F;p z>sCnNvxOKT;Wz6aX+#iSWx?jJ+;yB7~aB9hu~ zI`-b!Rm@*S>+A7z@g-Oj1PBe*PjOtc8l6x!ZFho-smi z6g_(HcU8reqU?&z<8%XU9H|tXe z=(M`P0N`%a-_gH;Tq`>Ql6a57g0!tc_t{MsR9O43n;Rr_Xw(ztjxtQ!*B?mbGT1o} z#)qI?s{D>O(@v5mpN?Gt$qZ?t{{12CG0zP@Z9}il4X(yOkgxE2lKbU#8xtzrsdYJ- zIuS&p#Dyn;liRxk^XeX7N)0*gEJ22k;A0+1+?8U^3t9t5&EDDOPsV(eA}OeMK4kxh}_izD!`CRA6m-|2B*f!ZG< z(6akK38e3pxtcJ&#dgIJ&$3Wg4o{`au^dIqtzM3QvFA9yts0|*kdK7TA-zte#r4t5 zEmsbG(57#R_m_4_ZYc2a1T&%0*Tq*nv4uq5j+`s84aL*_t6|pyq5uQD1bMmqFc5UA z29L|8($~`v9dLN5Ves}5^{M_HAdgxOcx}Epxu6FG(1iWF6BSS%R_haA&rhI^-vV|; z-~fO?z!zHo+D@QlJ@qes_C0YLci3rk>j#vDZJ?Z);@^_6x;`V10oJsBK!la})pd74 zb;f$M;r0x4x1Zf5u))7AKs7T~j!^NR+6^JoIF%go34qOC#rj=fF4W;=LSFlK8`%OC z5j&pRZB1|(V@bA8r0;UJi?YXZxzJ9dq-SNsSdJawQxOJCNVS^Ou5>`=$d`7H~>=RBSA>0j)Iogn2K^?J@s4Lgp2&7~*}ci8aERw$1%!p=-)PGt7*V6J7d zVI+fDq4Uksv}c@a#liAT3d*}}s-?s@kf5~vrhn5ddI6a2cveYhF#PfyV;f~_IBZqk10eAZ)QwT zGbUY&ja+Jzs6@rze79#hze1btDEmPKRz85g{-<(7Cc^%|`lPTMkvHD_U%iRyM4OrG8JMgEL>cSrQ|C-9?@T^ytBj8U8O(@Q$&G3MfI ztAhuYOG*weVXrAb^+=xGt1OGySN1gL!})MVnD3px$olDI4&vKM^Y1LE(=`qKxzH4d zZWwM`hdiYVJM`9W!T`!MQ8?l4dcgK1GG+@IG+#xz4_ClQUD7fPGgOSJ*?L3xEbC)p z4Cd;7Uu9M7U+0}+7Ww-sZOz~gNf;tWF?yfd;IRucxI21t(wdJN?Emqobi>F> zi8cD$X11J@UM;^v(l<(i&%{0#K&dnAIt`i4HkJ1;nyU{p+?GM|e(ZJ2CL@w=*(4WF z{9~t6&8Cb%o)Kp`ubB~nX0y4eUEX$+Ag>>jw_|R+I{fg>VR)HrfP&I#=EO}P>cmR8tpLZ^*-*1VI8^!?3kP^($la{U!w{Q9cmV_~X29MUyXxXirk z`FJt@z^Qx}6EplVRM%EtQbGzYYL3YT-27?!(2~W*g!#N@29%u>r^so?ZO;MVh}4G( zaLQE)zUq?f%DBohsa74xCcA?hpbmC4qRxCP@8&Rh4W{01u)FfrJKB!a*`lidrj_Y zIhqY~sxlj6zD4bTLL)9N_%on{#r7TW;y26L8Zq=e*W3v^@aSr2M1*9ORk!nQ1GOL9 zthuG48P;}i=M$NHv-K^QfnTFOW8(+k(}TkbfxYiF#^l>F2S4r9^8IfhqNNRgMyPeI z{4_=ua2rzf%FO_?1<9MSkyg=}1(iDt5@cS1aD+Nfl55?}oYC8iU#`UFAiE7s>?rLj zsO|#X^(Q-FY_o)-#F4lPag-EVu0!@R@VASA%Z~4Y+OXNLXRwq%{V<~jM5|+`WH-LE zIIaIm6+Zgf+r7Nu)Wo=c4yr#@KY8%CqxRWLh4Pa9J&$QNiQwxigU7soW0|A_=?b=Uuz|5W%fC)~}TA`CE#kTo3u?mDi|VrBxj3pqZ1`n~_m%nknS zqkv3|u8~s6$ZkMH%~-DDx4e2@1#V8*gZ*nR1F5x&b@3J1L-;Og+F;DE&4+#-wDRp{ z{>}DKbk0`5vfZ-xqpx<2lm>N%O?+PcyiQrScpDcDPoee2CAhswVXdkJcsfWK)J%{taOIDGqIF=x%ExpbUy`f7Q6L+vG*`}vOTI3!trQRXwQneC#&E~n|?zEzV@ z9oYa(w4L6F=xvQLb}Qkc>3P-c53&`HDyO?v(TZ&zT@^mAyyeqd!uidqfGMbR$H^`R z-7qWE3ZfpVm9_I?zwuNgJQEJu?mGA+RTvy5i41UbfSK7gLY1ttL}xG=jF*iEQAunH z7`0g!VH5s>FDh9MJHizgB6os8lh;=l@Di9W5@Slf4~A`z9Pia-ffO6J9GDua(2KLX z>p%~ri-O@t{|qpW-H5j*bi&-&4yNv)QjK+PXEKfTnk3iaJHL4COTd@lZh6-i95e9; z?pBMbR|X|x+UjI`u$sL(>A4k9y6NW*th3o9o9IQwk^*IRBL%SYPqXYb}7)W3J zE`A2Na*HC_zk;0`^+7>gad!b$lD@altgTD*dT#r*PZnN)eEUO4ghd=m5vEhGYkP%Q z5z8)Ba){`;+^LHN9p(M5_wzt6o@lV6!09AFryP4H4~j$4CpqX>{f)j!&d6u1X2^(kCB?bm$g!L6dky0|99&=WV)RyO$Dh zr->UO9-D;Jt79JZlhMF{i&?nFFt7Upfm24~we3yYuUYvyr>`{u=4VoUKR1Q=b$Y$l z0h@#9PSt>8<*SBSm;DAxL7JE`WnDe=uKFhXMctQ5dRJA^88~S}Px;CdGy#1K9Icu$ z;i8TPIgNN`%7VtG&ELp3&iqkraXWx=n`3&BChV6tll7DW%kaQqN%wM4((uKzlEfbA zJGfzz25JXT#>iM4^u^H4%7sSXVL zjZr6B(h@?Sad_Q#i%c3Lj71xX!cbr2X+D)(yMRP8)FlGR>iaops-z0!E}(k9#U`i< zi&GuSX~l%tntFe4{fnLhFMPi}1k7Cv6_&dXQY*417HB#o|ES4>=mprggh#W4nhcl+ zKQ0=;>nlFCr|SH^dHeN^ld;{OVZaevF~7|!N;VptBP_KSiBg!!qj`hT+$TTT>8dE4 z$e|~<=pSFz+Pw|&t}9W2_<3WLeoT&1DA-Pvf7o#`Iy$bDY1npOFzIS*p42Ept&|PS zn4R{U{R_>~8bbRQ%5kB^e)u+V0|dFhS(MCV19W5WP1o!s$_2E%Q1Ir`EnURFb}8|0 zEBtUbD<%a2Vc)vqGO3Iv036GoP;NkU+yta+on#Sj)ZYDzY|df*{ruTue|3}jt-~2p zNdb6)9vL8uN(gVdKsTHl!>a}BKh6LGdXJ&1hBIrmyw8Fhjrzo+6mJGh^5fI`1o{4# zk$u3E!BOnQSee4sK5<&1TbSclIWU1ustqRg12LlHfpSkYStEvgK=MCq%md1TB%FVs`(>+w!t+_LWTSpGtXem*42crux8ALkC-zXdUlm#UV~&iDh=e^T|Xntt%cTn^}-e) z|F)gN5mi74fVxAlKZ)hZn41k;{#w$t)G9f^@l=xBN%0&R?>J`ra;z;b%)w+#4-grK zaK$R~7&R8yZ?;3l!}7d=P)u$sE>4hMLzU+wPaACCS8{s9!2b7Yg;W2h zD!oNt9GjW?1mKg*81xAy{Az5&B*ShqfHQUpK?$TbWrA_%~=`^qc7#JMC z@Z0`jz4VX#ZekZxu>X|DiofjKT%2?VbKt+uqwR~K3o^TzP`NeFxv#3j-*{|^x~4ZBAmEhO74fp>5d8Y_6r7}BPwARVv7=M5$tUPfSY6a zgDd=NJNPxoW6EmyB7WuM)DQO}k9bia!yD4%Kg;)Uic0g0&VI)q*KVRY@a96#PC0`= z3*6z{g}XmQmLt|Wa67is`sURVc^>*|(j{BrBsXyeNpx%eSl#ZhS1dBcE$K{8BkyW> zMd}mqP^2F|=o=8{vFb+3Zy}ZB1LE-MbiPU!LXvt~p?i`7YjkVKC=o9jD`ThuA^o|b ztP1gNJ8r0v?=|Kcum=xUuE%2XY)7(FkuC!P6!>k$s%DcY3|l)0rAj#=FtdGErjD<$ zCA6@ObDjny-i0@2dRZS&z~DDM_}fa{C6>U{%7yDr-cV;iY0M@UT#9bVsgOb|H4CD1 z6-~1CTQ>e5nI4;E{GItg{l*cI!=pkj#BoWme@8%H0Y}%H+DfqBBDvQ4ViOBIVT`xd zC%m{QU&lkyv(2-ZtR;KEBTwjXj06V#__2ceZvwz^jVzWp z(R}57`6Cr}qC57R(%rBocc49pHz5&_&6sC50_l1hgJM##`}sl}|56tc#BQ1HC@7NJ zls6;QkIL#7HwH`8$y|XnHqWfA=}n*DEOBS>z+P+)m8xU)0?Q2G(L~`^hEJaE+|_xq z9tnOh@%k%^$-GFM;(wnl$`B1;Gg(t6{>o#YIIqjaI90t+PVxC^M*4ZNyk81HS7)E7 zE~iX=u2O96Rr5x3QVO&UJUFY}fRseahF{asIS%~?;zBQVaa#hbrK(+WL&53G9u^++ zEz+eU)H14{gf~pD`%6o6g>7qfW=D)KLxhfa#VCy^zrEB><|%!d`NmF=keY8Ig@@DB zoRK={;nPyqz^QfylQ(2+eSrD134VkdC8-%f5H0E+@Fzf_@8SK2m(rHwzQqLMr$@|0 z!9aR&@k*W>FMa)^2WPC}9JbS~F)Ct;jvxPej|bw4Ht=#wWAOysc0Qi)4_FqpGMBbg zVGDe+DH>FJy;M}g!NS&`+J=CH0r`A(GT&}|igQ1T78GbbLyM@T3unOm_8shS3VhPF zgkDT@4M;!aWF$!bJE-uX_B@vN_G`{Y#E-LKbjxJ{eyL;5LZdjuW)0XU7zAB3ICru& z&%9JffehY(Zdc4ZI5^Jw=gL#TUn7mQyd|Jt+MDZ!h6rd?p+m{H`DjXd{RzUoM#i`c zpO8_Jebt*-DKlq2G1^}}Tum)%_cbLvRv)_WdQCPy=c2;dZ(8bPLwoyNnl2(%Xi1Z| zb#i5-)|bDjF}Ah+yP29FMzj}|H}ri=z01}^km|o(t^mg)}N^C7lkw) zNOFi4`ycO>64i+Qk&L5IeUSZJ=Nk6l_7IPHqk|bkwa8^U-fE2V-u!lslgx;lp`*Im z#B@wmlOx}G*YiL*ZkoXNl800JAw3Lq2Kmc=E4S(#)R%J=epdRUMFs)e5N;lxJYoO+ zL(hI2WOZ+UT$*wEd(%~R>P~99T$k&457qRvPhZU=pkK*lBh3q04y3n6Gj^wl$vtn@ z;$qg$l>`}WZy&@^@*2N|wm`Ke!<}7X_SmW3{AghYizsIlchPHXZKkkCYsK-+aCmdJn8ZAeuZbXP}+&$OvOG%r`V6S%$sFawQq~+T!m*kOY`G**QHy<{1c$BS4 zG3iKq4xj!q_aJoNlA$*CI>a;4ja)t}(E?#gMt#B_?H1xLRdsU0|6A;cS&^TTXY2EF z_zhX!(WnO;c)rHW@*%Z^RR|~z%9*fh)v$GYx8ckLcKQIy=%aM~* zQ^Xb%IPxhH(PF$(k4cB`cw<*v@<56#SuKXet~J&C-!Iv6;7_gKv?11WLVAL{Gu^!g z6+zEE?B&2nI*L4h%kdefD#ESy-}^h~g~M zBvCv8Ma~8m&>Ks!ds%^O4bC*ok{YWv#at-FKIGTyNl`}7Cg94h2x>=SJ+P%jP2h{(;4DQ3sV8Xi$(TV;e@hK|&{`I2q(A=6wSNTde2 z_o@w#89Xmu4jnFbGL=e-4SSq)G-}z}*}D-PEGeY@V7Oa@^#NAtcheLqh|lm7_Jq6< zmcA=o&|>3$dI1}5tXf}Mf#hG#2{EZ3pLNP!1%Dwl$rq&9OBI1ZXeUiY1WGITXyqd* ztj60auSU3XXiIleK6~_t2?$&!$e;TkG#_){7T39MX{$Y4vl1i|76=HuejF?G{?LNS zc{=Xho5x{+o*NN7X(oTi=dWx=`$nI8GuC{kdS_yh5vWg_0fHFaH4Rg*Ahu&NHp+vq zSz@JfAw$RubDvtd!6B~sZ`5FqUUQ< zJ$?4sig@@^`H^{x!lf=Lq~sK(W81GgM_mdRmMLs;(=&{Ij7H{dpHMw+v|tP4CTLEp z?JnitKKfgH9dy!lNHjs-q;~mr#(tXztsw5t7rgpZ45Ezs(IsD$UU>TeZ;vVMr%KGV zU?~o26Ly2}47fVPX_FHV3^cjW+_GGZ@H=NUT2@;K>*xRGrwj3Gq5tk)>;F!Ft?adR z+}f#(Hs)#PTgjWxuN+9rWHYdxdZ-KKB`W>q|~3~sR)BDjE5C>rdLi;}dinvf*a zLL8k|y-PILHeE5t8Eb)_-{h`2nz;OB{~ov`O>}K{-FXxuVzF^86qq z<(J{Glkn{yiN!0HwRHT|lfdSC8r=Abpz(!x8Uf3?-j{ve^W_OYRqDV@xaN)d&K227 z{l^^US~TvgMy+1{SkCyZ0r_f+=iJDR>pGLLDDx2>rQCjRmGNt$wAt!2!@7jwRnpfc z4-mCd#QIf_hNxR^L;L~3(&O z-O*{jaswC5Pcvh}X0)v$O7|IY&;5<+MrI$ggzryXeCnz#Q6n?+ljQ!LxELa=>3d|Rj)uePJ5LObrEbV5lv)){yfg!=EpY>HDnuj!JwAUDCLUXTK5*cOFN3g6hy0>` zpm(}d^mCOCqDWfU`QZhl>ryb4+hy>n;^!BY@M;R{wMV`OQfonSGYsCxKhHwA5PN1F zn1*ber9*PIL<@pe0k&aU9^RFR7-H%Botf+3*m=7Da4WtPlnEn!y%Q|41noSp9^b;x zx(ddJ8_zir=*|(_1Mg-14^m{o#0h96ont%!@;6XD-n5_#+rH=nk>uNNuGd{l%%FK9 z#G^DO+-JLp#5A(5e0~9gCEKV#zxed!L2mA5bA`uE*thTS*j#AtR2Q(~mkx-6S5;X! zxu3Jt2XNN1VnQVDay;X&cant^XIxy!&#qlJ%f zUgH!(T}E-%(ojts2EqosApGt=SYo?-juVoy+ux zHyynl9H3##&Io5ITYmy$unc~Xb-&sEidr+V(N%i&xM{{=q2;Rwpx8> z^$8Rypd`Cyyx>O4&f_n|BGTP#dBXXnQ|zT<>7)&Npp2OcTSya4T5$E)dkX{BzHyh; z7mw1KyP=%c(zloJ@v5s%{~Q)DA4G=VN%GqC&=}`6*f{&O6U+zEHy^X^3^JV+e1H)WEd(s7BabTj`dC z+}}_J2Sem|iDJYeamWa5;b1O-0%RygcrHIZrvG|(Y;(or|5fPYI;8xP07-otSg_BZ zC0F|5`A}=-A=PghyUom_)MM;kX=<%+F)f3Aj+YNQ@^+W%s`r%N=bB++P&mz1f?>gH zk>y=S%=ROd)idbCt%&R<3mGqL_Gtg-Qg@a1rbC#QQy%3(EzpvSKiY8{eM#JfUa z$R?=V^cahgZ=&rn+}SInCT{K7#&g>9&CEK^E-iCsse^pDmcPuzv8EXbOnUbS+N1Vq zDt+lf2jpb<=*22+f(aSB#5NTZXxk~IPLp}Q=8BAXSCMFK#He&sk^3_)p>~cOsxw&Y zb@U^OpzltOy8cWP`cFU`7A77A{<#O+=(p?^?;S!&w>#UCC;82=AJ3&(fTm0}u468Q z$rR?2EO2mYrJ2MSTK@jHAlp^)y=2EjHOG{&kiQ+4i^z^v>@*-fVessf4SiPssLzrG zcv47_5UY@)U~dpfzeZsArw^@tVo*o-=-o+ zbDv-1`yAl7hob~`9ta-{cyzG;XrVn`{B6I81yYb)c&&Q9uBd6q0=qSxyU1}BxZLYi7Aa@$8QBa~uWCTCMv z^Ojazo`2Y~Y4xf)knmK16`nb{I1#%XiI?F2q5#%a$hAU{2?nV?cxzjM7Lh zR7<&Qp;Hkz8^vUG3#1`4p1@oTx%;f0*MdD%i?v!4Fu|%)(86qUM&6u$mtM!hT z>0L1_U=nWiVnyhNQi?sEQ;YO5WR$phV*o>uQdJ=!Tbetbs-LX7(R??12`BG1k?OpI6!v0>WalT=*afs}I_T zAcousloi*oLiGUICe^0g7qu19M`&FOXyynF&22wAoC-eFG8%8zMY|i?d|h%R5^$0c za-^O)+G(R>;roDqKSNB>5bY^4nuxkTm*kzIwtU=m0vd9D+WTEP(QWxrfQw$Lxv<}u z2OVyc8PjBl=rh*ZyZE`GM%FTau1LWLSw`Qe2Wf6q5{`$9|IJN)nt-$t+8W8ghE!G7 zWkaId0g=Zi0*al3$!=MP=wjX*M1FnI4O;_w0YSf{bZd$o$NYc5e$w$V+!Eb6u>#P4 zo@O&CZpSN@wA7@XL>NV=1ub{%8)%U?)1qd}$1&Rtk5^wLJ|zZfoKyc%KO|c) zn~PJOm0?+psqu(cba*B_{)}a*xc|23#Z5+$orbjSn$yW2t+`^+xwGv`pmn_m4VZ;W zOr8-c^~v8SbZCAj;QoAmSXckU=>UI1tOYY)j^SKZ)X% zGijt4H|GP{h$S^b?W~7oZ>k^l-5!yV1yvGD*hkPczRH^*BzNq*tpj=C9Ofqn40_IA zeOIF~%VX!(ic)at&{pr4ZI-=N2F=ICWE{no753A&@O{U1-Qg5D^a@QOp?~$uL7M30xRwH* z27!M;(lSuN+*WlWxh{pWK;vHYxOeuxZ0K`&^$xt(p5blI_T+2+STEvCQ#_;D)0SEtUn)Up4jeQ&57GDSX3x1tcFBr+CGBXjxSfVAvierC zge=*l{^Ad=1MAu!%jn7P*3VaIQ*dizP;c-UjK2<!0Ry;( z1ACKk8T6~H;f015vq|s+7TVXr{SR5{u^Dk+Ew<=BOQFkJT?>w5cBfEW8q>{&JT&sk4SIG~_S#;_A zIqxDFi|9M`DK^rH&K@~;c=gi%R_tsOUZ6lUbuO3{aTw)25R+!=Bba+c&DO#Ku2T!6 z(NKQ(sn@$fgrvvv@$8vmgp8N9z<&{~z>d)uVIRxA*LHHT539%R5P#im`{d=q(QHE7 ztU1A5P8sw8Ie{WkEMO6*OUl%^19FY|Fsrf⁡nx_pBffp~aTq?jBH@e^55ED|h%e z?{MK>EPd)7we6ZX;cvLoY|In33-;%eDgz#L?3g3Cs|0)07Zca%(0sbj1VJKnTbetc zFe0McNvc%`xV5&}|Ar(kaz{m&{%=!GVW&v_9fm$_eb+bFmmZ%|?n5OLjEyW_C?_iJCCBRhjPy_~jjTF#Ycf*m`T%o{+Xazldtu93NEoyi< z1I?9G0&aav^m4)-Vps*JmDi1QLxj*{#kAkVo-F9|7kLzFGM5uEa6fES%DKd8ZVjtXPSQI*Mm+WLkAsi4++@$Bqp8!d zinz!IfWKyDfAMcm>Ehm04!Bo3tuWF zm-E84HOE8c)u1SNO!H6zMps=S`iVw{X`P*_ClX|x2FCMAyUl)OT7O69coT+>Q`a5m zI_?WR@bq%Y>A7%RrZ-~`98(?Ai)YVPD7}Mx^9n3eQpa^FQ+?+(G~%1>NcYN>`h|v! zz`*D+L77J45jOxpQJhM78|Gus^;k+5B`@o&SmE`7X*@ineXrV56JuU2E&OW$zzf0F zT>g|T{BZcTPm~vL!NsC!gx8@Pd)V(`IEq!55!{6ozQXc#sqAuIwT{b;>rgK0_x@ZQ z;>kUs_uN6b(KYNv{`M>HSYMPW3!K<_1cK%AuA}7K= zBu9N;4Rq4_iW}cdBV(%>^LB2Duk2vp>-e@xEwi%u<0aONp$`B50#6P{C$n62;k$1Q zPtTbbB{i;7etUX{zp%${GYrPp&r5X1GcZ?cm1pZZg6n1B9T6=m9<(Juqv|UAy+Yc} zgQ;g|y7svD;;Vit&8)8JqEs^3U0{hovl9Qo73I7xcZ0 zq!6uYY_!vo7&wZ_$n1C; zNlkoT*-K&L8RB3iGR%4oQTT{QJ33L_kR<|x@I{tf9^S1iPlpR>3~%0hWz@%wG$5pI z#-cC1?%*&H`D!^VnTe{2N`2i|z)ZEAf7iU2GSW4bC9Kh-k{!WZj$$&NGBxDZ9mJ~^ zUyTvN3tk(8R||(=m#5*N z6TU~jx4quq;KJ^mq}G}p9+V~35oTtT|JWwN6? z;WQYoB_%wng?YA^HrB}ocKtg#aCwb%2@ZneUV3agwd^mBSV#6Ds z_?p(aH|cj&;L#|+;;cPAk0slGQ3Af&hTOfj7*LA?&kY?~W;GsGPy+_8oM2|)C7Rgh z8-}>YX|j(3%x|BtDm_9P7S-`CJXLopcE$CeIE`kaUElcdei`qvMnIH6uOhJg9W_fO zxUy3cgRRBlP#BIu5k2mOCoLaC8_DRaB&G`aKJ+;@)vCUljN^9A>I&dgL zV(xC|WEZ-8Ct7IG;dQZ{>6ex-DNtEO{&W-ZcEF_G{v7k65v2>oBkO&E0N=aqZ*jO# zpd&L**e11=;TxqRa(COROb?^z^h#F@t5k+67$pksGYb zs}vP~>g8S*7`;wn@#_*Mk6tAaO7>1IIIy6Nfi%bJWD!8P1A zlcMf4*kVWbyXElgdc7c8=>)I!pdJSM1xL(JYm`wpIz)HZX}^ zs}{igBFvQi?|;oXQjCgIE-`$rk|soIuH>bXBH0hM|30MnWR;8JQH?N*m=O_iu+n)K zJ9x9eeh#Cqy_qkSo)-01Iv87T^dEOCaw&(2A^o@=QP0oFF=SR;GA3Lu9Q?R(2s^)8 zZW%yqBHZZcFSv8+<76+t88q2$55_|0p1RZ3fAl12DMuO88V`eq&i4>W+6bA)OlgvtlGRpq? z37tK2r55|3ZVJzY{__wa4cLm*Aiy?C^1?Jr(jxl#NfdsHki$AzjM|49bMWzr;m$r; zLr;0b#c;gUOP?mM6JKU%LC$|eH+I3P3f2I7R;~4!oDo#A-xvaKpz*>eni+8NE zf|(Lqh|tHIsLmXx`&AGd4!BLY|LS=clFX83 z@=WQ!r4?^(mORg!5YoH#4tPNQ$U;ld-1~1%f*0?3-&bgqf@a3hmcPu<_+AH6+Fm(g zM_d|@x>plnC7A0@VX8%*9su9hH9@_ban|86njq25HH<+B%@yeuo%HE9vtZQM=Vlu7 zqu*oxLxtLp;^t)&L{UJ1x=ONrM@_?8D{d{ouCfqn>JWXSDIR>y*G->gIhU)du@vxj zv}(g$SIaW43oD}2EZ8jt!?&2ALHSY8RpgR&Tfcvd0q@oSfym@kVffk;Sc~qz5l5av zZUft>-KZLoYYEnKo&`2Ke1Sfi$Lo6BRfSy<;esr!n^C`8iVVMiN{N;7%3 z^?!F!VqyCCW?!9n*R~tAVl4c7oas+iq<#mjBhktDMJ_I^6e|ki7y<*HqmVL&Y~zl* z>Oc=Hg{|NN^w&LM)x;;T+O7zFaOZ>c$v;_yoomcc)Hed}U@~s$pp$OwkK(MWTqueF zIj-1sN$PICQt`Sw>5!%a-M8EiB9=Dyh|9tV3^djP-{O-hEwM9#)2ehyHvBh)M84i5 zx};^%6lx*gX|AQ_&ySJVZy^4;<4ea*-z&KG^o6^k_CHd>3`|Gh}v*Irl^pl>z z%rNnWs`9>6Y$t#z^!alulYlf`X_S0C3uDoW$%F^@3-uG~M|j*$rKp`xg%2E}z}92% zCn-SP^*vRzqe-wNaPXyx5M3p_06)UwtbR+w8eS*pe>95Zo5Iz#y)_S^1 z*k)FyTgM3YUd7v8bBQi_I)e^rJk0f=^D}<>D`Sgdj`Dv zFN*$^+)ZKr|GPnq%%6LQn^dv_V`O$EKGc3jH>ol~?kBawGxI3+L6i9WVHtmZLE(47 z%tg)n%9`R=)V*6vsaNYhjm$!``Ec_D_vijE18|khDy~=u$hvh#4T5G;y$mVpbN5Sh zDk**(u8PvtJ1=-rpBh{XnQw(je_VQwbxfci7s`y#(6st_IYEN4 z3Clb;`xEVUN_4y|*A3Pk-|jkZwI<1@>gsO95dpOg&=3Pty6K}pkbJVl=V1w*5HWOb zeh@%F7v~8*S5@K1@$jelkr;;gb8qM z-8E^=xo%DF^JVMzA~<+B1=%G5z8#xJ+{rAqxE8R__c!rR@+LRdNuN?Qhc!Nom0I|j z?@>6GgwZ}Q4NL--K|%Z`24)d-^=~GNEUKqcu~B!)vlREz1RWL)-mI5~)wZu3Vg6y! zK;9qMQJDBoT3`LDkA}9quHExc!^C;Z>m2OE((GW%oX7tu;CTciMEYvHb;O4dThSWW zgR%n@%l74;-xi2gCmymKbhlGsJ2Ixf?x4LbBzUMa_qoYQs2s-Xu#NtThVQglm;^x> zUW}~}MY-s{-+sMLOQyt;L7MO=57VW~5zDLf$g_nB%z8`=Vpj{05ePBjJUQ+a%YfKFaxxt14ERdr|6%jw)~%7OQTj*oB1No8qU1nd z7O0O_;E!adwV`uGB#)-r$d;EQcZkqeMN!Cc4Z+O_ijA90`Yci?u9|k^VQ-{zP-+pvk;a>I`>BX+Yh-f z*jqCD9{!XIcvU7s*{M1a+K)|jD^3>x5KUZh{ni}LH(1`9Z)0@|2I+`Y`pPw499Ns@_gn}qi@z#K%YR?1!!AEc7p4hQ;r?zo5} zuyCWr@@b!E;*CQejvQ-h=vh$4>UXIw{97LfMl_|-c-JzHH-R#Sz)YU#(cO1lm&<=p zp#C!95vT4EWsbY!@>8*qtYyUyaZ<4!^WN6iTZ>M+lNuE9yPCF}gDIo1BD^u}o-Sfn zsM*Kke)9w5^%-|Tr4ho_>cny>Bx35M@#FbqexU!aEu}Zvd}}URm?0x!>s)85977b# zg8#*`%}jZm4Io+C*$0dx=le4vajl;F%B(7>ZipCt(lXHev)7vMyj%FlMQ>qd)Y zh(yxKxv#T_*YHQOMTRmrH&aS#3Kb^Q#SU#ifHi5+AVH6y9-7ejs;W6VR`UAAjc=&4 zwc;C!rUd5D@#8*PyOEdgdQA+#D7$H$UPmwi21@#~679m(ZXpW!*z zmD~8EEN5F|IdUU&A~E0}U~>UF^-W7;-~B(kpVh^p5&>%s_6e%|4C(lY0$Foz3H7{Y z)BfJGVeDjLdOij@;sQO1@i4+gqXD^)oxVx-J-?F57R-)af;)!ihFZCNACUuGI zi91FONPZH!h|`&bIyhJ`Y?cE^xJ7u?uuuT!Yd2%`VI2#v+OK85t9cgPx7R%)yk8;X zHsw%>@G`5<5D!Q_hrVhZV3MYgSVpP_1K&BU)Av~VMy&^Cqx+bJqba=VNh2M)`GrBk zO_Zx#WUA1wto@ka3SjN-e4{k{MJo2wN+?{JU&Eei&JQTb( zly2zj#+t(ZvG_hk-#$eg+Vl!;<3&qdhFO!OZI9*U82jc`A%1x;F#x~Blk(#Lr0vjp zxuzg=@xvu`iKYyipj`4X95KS70!!jIZc`6XYPt%2Ba?A*e%wB6-U7o8;F;(9Zdg$!aZ?G6pDuW zUkAHyQHvW_?#sZN>RRl2ctScSt;)x1l3`@o%tTYW={JOBV7mGmq+#i2z~i@&*>D%*CN7jyOC5h>_oDdqle8HVBH_d#un zyB>KWp!Nu>RLQev=nDm|LK%Ib?HM8~nbzzhI#=2EXhKZSF(AO{wBl1djjVUy8Jfxa z(BN!#=ddr*yi&pZnU%O(1C&BvN4&=1$=85EJOQ5kq*UG?!V#i*_cZwgGF8&-t_)l< ze$x1~eco?6%6u_X))PSoTpp479?RVshZ8R!j#cIkSRS#QedrKA9}TcttAAZR^-(u> zc4R6N3Coa3)C1>U9MvX()T{D9ITzF@FGRz3lQ#~2@63dX@Ki)Fv6ZBGR!227V6mRe zotqP!gO>DEdLVdobaEU}ODgyK7#3+YVVK=LQyQ5eP7=L&ka>qR(?!uBH@0e6AdN)y zoXjHpj1`ziO1vh$tS9>f2d$@TF0O62(K3P@Mq80XM#hLiq-77Chok%0483lRx4b?L z@_^TMGCDq$4$O9hc)m1CaH2TF;-qtJvH1GVVp0>&?r#H?Ph6wZ*~6Lzj;xrgeTVbz z6o)ZDEJnSVLF}Au;p16P&hB^KbSWB_USpYg#@FueP!`d4^NW$@WaGC&!X#$v3Bk;C zu(n8n@xxV_gX@^X&p)R3A|`7c0Lgq){`*yrc*{8jZv3B)rCKCHp~UhmW`Mto0d2qi}E=dpaxPjd5rag0@7Uk5Mv_*28U8ry#8THjDwx65c(VcYfvRQqD~xOIHkhp26&f5Zn4}~CSu!U>yNJRmJIafmXE<7BnIspWats@?LlLCzkCC^oTh6 zPmNXOeVJ0@7yqc_TsN?H>rNGGRas#@k5~d5aJq$G)>T_dgX-!-ARlH^D_O@kpJj2; zyaMvjJ%d6T-NiH@CyGH<3oozDLKS$=&Po7@;_rm85F!y@IEM59$1-$#ne^r8*OEo= z{ZdhiO6FM<<+MjY0M6qM_mHISLI?4+m^pxxh&>$%JM z$uy$^K?gLC<}{j)-dhb=-?Xaa*y62Jj1xy~n0{UA-Btn_z#k|2cF-W+VvSWFz4#Q> zy@Hp*0339OltDu1<})HsQTIXFB3}*cSA*MX1G#xc06mYvcu=DaF{Ro_B0E&!hBau&7K~-;~&?BC?Mp2 zA@Vay8DuIIC-P=Ru1P|6KH7`^&8IlKMCRtz8Ky|Yv#KydZ{3x@JE`bcvBm9YWnbRI z?_3dQZv|h8%!zm1r9gc&Ke@K8G7gdZ4&W19lH17l3p|X!`m^N!<|csGyJ>Ane$wj% zFyAjG-}u224LM0%b7FD)DXVBlU3@<;>B+n zV2x9UlN*iQ)84Zi?MRNI<>ywwO`-#AaxgV$iz5%(`r>omG%NcZ(Mc&MWv>_r8ut%3 zf8uc>{a)M-3lGxLmpEKlrmNQuJ)7{LK%i->Ze1JApqFseezq{ffV=&J<+UzDwa()5 za7}(As9Gn-VP1J=7fplX!}p$@3@*WL@;)M)s1VoQcKmi$Gm~!^vh_&l?%(y0OLUbO z{(I~pP4TD7r_HN~N<8AF&)dJv1C|T&gMTL;u~8OOFbbhNkvNo{p-p~rApUn>L%$wN z>-w5y+bl|d#?SF=7`Nr7vF6X0;AiQe*;V3c^+yY=jU>t3r;%dO5Ebhxq~++$GzqSC zjh`E(V{U~=xOZ%=IQ1aa0zn;}cm+gUt&_OnZ}4ruSjfqHc4jc|Zg_4KDarA8tkmBX z&HkGzhzGJ)AcMiK#s^3ivDLIg5l#Y`8(p=*K7m1IyYCJ3U4wKfzqN4FXLEi?5{ZCf z_T}WZBaF>PQX6ycJ>sDJt}q)sz**|B{fb*fC^3M+Y3sAG*}bEua$6udw7s^&gG4XD z&D&G-!BGy7dS1ozldH8Uam+9czkdXl>jSFBj2iO8$e%) ze)ki%Vb^ZjXqZ1xAgrXbt7a>>-ZbTJiHfLW!O|>izsyjDXYCfjbXR3GP~>=!52V6l zN<_Qb9WK8|yO&0iwzLy(eHuzOP$4eoaha(-T=MUHPd>@GKHuX?gH{{Ge7f+X!$wxN|#7mFAI{H8m z#(V95k87eJzkuBgsrvkEu;BH+V!$pw)jnvlLgGFimZb$_gpSxJVmE^i1rXTY-jl4_H!^;%GXWU0FQ!&4NqX%>53|q^ zjdqgp3n51c%VI*$ouv!k;FT@Kc@O%kTf2zH3uDRMfYIe6YRFWxF{5_0ix^SH& zeH~#BO**VPC^lJNLLEnh7Tz?y#F#e1PU#sI9X5z2Po2>n*wqlv?E9GC&be^rY?$hV z*<;!1q=`hfL!#piJ~aDER0eILtJ|$Z&Uv>op92!F&wy35BZ}r!_Fgc23UCg|Ppd1jipr@ZC>w4@UWI*41R&PprO*7^G658ats5nfhA z@}W#PRDh(w!Ir`Lbd6xG*FM2F18ZP)$)-tNb=8(x@sVb#FcPsx7jk=^!KG2Ww*lQ< z#44DMrEsI#{U&c+prh~sPe0SjV55e*-bR?BeaTyOeXCBFN8gfii11R@;q?n+kg*97 z;?~}&nQr87vUTPGDs#+oTr?8sHm7*IZ|6jtVTW(O zfqy;(@{h%0Y^;7xJ@*eDn?3x5GJ9xZbUyIm4O91n4>5U6!=B`R;@fN*mGTn0edbS0 ze*fuWT&fSj44C_N+E0j%xjR!I|A8~M*QX!8ABtY#Ul8;JQjs@KTSBTneJ5S&2B@d- zuf&acc*cJUv*+?Cu#(Pmc~aq+Ut1)o*MwM(ZFHNYd}?!Ok??qV1=4l>{iNXVdV!O| zWaER#RcFCSB8A9ngIO0Is>jP!c9Nrb_@y#NuZT_L3#avhlB$XCbNGOLx=;*=KzUO( zxh>}dmBxlo8cCnUE7M@ihi6e4i!A5{;e`OoQmsRGBTIE{>wQ^yLRvL${QAD(OKV=^FBxn49 z_` zx`=TgxU+x#DXfDUXL>^0RGb;wa~Qm$64UyWa{cL_1=}E!4N^xjp5u+AK`AN^3 ziizjzS#TN_oabbR;ZgZq)2t>SHVtVm?aDyxrO9)EpPs<}q=vicsL>|6XW-@ClnL>H zUJdu+d!~^*K6le=3BrEvE)IsQicm&^lLZlRlMdY_tcSkpv(_h{+&r}X(pFAgOKnL`He&gJ2|2`%`B;xH?V2f7-0 zAS1=GrY-Npzb;*r-YL!DfQU;?KN1FUn0m&f|Km=#QA%-Uq(Qx7G+kttSQZR65Eq6)EP*!r zqlX@4go`ev^b6d>yy?a>>yMLb;IuXw$xvK3OkGb=T&Fe@-=!7lhZFbNN-9_w57LFw zzy;R3I)0ecp4GcLv6l#HmjRaTiM|&Xpz3(Rv6jfOOwCVP^*Nm!iCe&x0oat)PF@|L zls?U}y4Zy0iGJ*%GcT`z-hna$@T&&tplX-vQG`(`^N}w^vYg{51J-gn5jpGUAuN6m zm|ISNGkk;!CNE%FIIX9zHA^YwPklkQuFMb4xjrW&Xu^5it{KJHvH0F@dDP1E>9=p~ClP zdAJ8V`0w97Admq7n~t`VQV&$`^7GzxJ#I=eyTnV^TXc&=H#3fAIv&kgsO8E7(C5Zc z0q)Vog?ntf1|JQksosDOz(B*Hy)mlb}IC{(awu=Mut~WXjnTpy7G1)!tWUq%^ z{UTt++5KA@5tX)<{M#6ce>BGQTslUPOmRqeQGQD-qxXqqml{kga)E>G*{h-~Agru7 z+RwglC<5M>JPI?YVe2%#*dls}-ZSyZ-+zOC{Ws|ABa%w~&s0b+X4p-TZWNkILr3=-$gcK|?coo3iddNLH0DohP}Z`3IR z1e;hqk}oK~y>*)%$>>yXxfzxQmL zfKdg@o#QcAX{K4~J3OXI3X%cJ(95(g`V%Fy2{SUt8KmQ4^ zBCxKu5SspaA@R?y`4`Xs{|3i@eC9v&;P&%oP>GjIeuS2H>?S4;%oAl%5DD_s4Ujlxl@6*~vP|rT%IR-x|^idUQ zI2|uf@g%s`@xg^RP;oQrR{Xn0O-J@o^o*|tYi4Q|V}DVzWLQJ>eaCQJXVU5{9(8XW z{Vj|A)f&)#a)mrp%1{4bll|qdtR~>l`#P$$ZaKO3$L|L?G>#vW!YQK2w+wg3a}Szs z@a<+`-8VpfZb9jm(=~QG17djrgXosaC$LK7%HyIBgP%jP6#?Bu#)Y4JeM5BP5SZeX z+9x>2&DD%|OU+Y!{<~i@E5-}4_NfFSJV*G{uk9MbuLC3SCk&cwYn)z4nY6S&1H^2F zgz640s?QVORKt&SDf*MCAk*jhXd=`p6XSXr1Nhew{gwj25e_Ny-(dAr~6PO0JwRA$WEnwGkjY+esjH3Q254^1fNpKu|uZ7?2W8$lXc4{$& zr~W4kp4+O1p^yBS6Cul~22JStm-3+5T(%GYN8@Wd?e% z(_eAm+$JkyGh-pkT%V~39sY1x?7Ly1hr|sFot=-WO0Z^ljXcD!LZmI$ zv34}|bdrvv+-!uKzF{h7$(G%2n3`3}szooFouC~gA23>XdEObRKW)ZWL`CDZUS=zy z*N6Hi)bLrelGP+wr3Rp7*TKq-7sl8?2WG7`i4JUkmg~!KoSQ zbd&F_Q`>d%chBD<_9u<@c}Hr!x!x^N@Jl^zkm}}d^5t)d&*cam+#{cWkB?HS7wntz zq&^bD9|}}rZ@+=@nu9G>gtP>`Tcm%Kh`xLmte)Q@Kf~WJ7WCbQ!ypCgyi>S`po3>W zl?>Kxa@$XpiFAdqa_0d~7{7EKns~cx2MZp=k24tRk{zV>5IB65SRrj{fr6W-6IN#EB$p5K7w(??KR46Ywx05#3(o};D-@K!aTt&lr@8RyXKeyDQ=RM#x~km6530OPIF z%&NQvao??JIuMf>a>rEHCD;nkS0QAXb@V<2AQ4%on*&V~(-KZyelrXZk;K2b1`?5# zTg9!P@bGS+m@VBaMALJ`@FsA(WzkqVfI)7QE=k?08HGCT8guH^N0DXRLBPc0?w>3% zafVTH@msqkj9xz=HvIiNgnBn9i&8a~Rx^3AUm{;B z0yAlUmAo^89I<4~!o2cQOW@#K4K!ZpU2FgTjO%C8zvY1EGFb(HL_p7%I;=PL_oj-ifc-CeGhu>lzv28L zYUH9qV|U!2WziqBw^3;}V5|S03b48@Ie)P!`Vt1HE3+g+jcy;-camrnjz5@IqV1sB z%bvlY%WDkh|2U<2tqU+XP5wss)KEm^vt;x_i?yfM6$#*q<9%*@Tvhx$TO;ClGnzLw z4wVO$g6VoQCs^oBmv9#4N3E6MIC1P{C-y#3%3*hDXJx9R&k(rY*>c{|i{V1c5MtMl z^2lr9Q1GFNhw0k?#uX;?!`hspQXHVU&lOth4hgI651J>Gr@$+7$&N`=QueF32Y(1&D#9umcqLu&#%{dY|M( zMSd-C(*jnOS6cg0;t_)F{elCH0Rnr4HoWpV6q^5)vwNekS$AV|qb?GgW| z&leo-i8+G{&0B|&D=WPQkLWjtf_aO6AYnE%>`nXAc}mKCd7s3|UL09}yDmR}Z2HS& zlPKWMkxUO;XNw2N4srxilKTx-%5OI;R8)Oa%SntX>%4jP(&)M~^i8$?Zj8q8AD@h+ zPPF5M{y4lC`?2Yox!q*~N4ae*1ZX+T$Bn*2Fik~lHKD6ujevd_@ z8&~Q6`T0=g)ZHKMFeeTnPxy!Gn!Yn`!e`CVoPX@bVxafkbyZP3x)b!mc8|8+`@><^ zm!Ma`SrPMl&v4!)vGfPp?%CO`dZzL^J~PV&0?ipHMw`sWeZggsly3{#)XQMoOq%BS z34C+=X2n5>a9|(7^)+TQ z8z4C72P0zAB~2n6LeR#V3^XJq5Y{{hG=ab z2C0`}A@M}M&KNFn?DN9Bu?nB6=S<8u(Fj5Az^;&ET!_Tl1PPhmL zP95ODQH4wQc^=H5P(5RbY&AgCl@@!wcx8NevG-=*@*){R_=dfRRN`j_xFr z{g=11Ka#+GkxxJqI9e&ZNY}^~y4b}$xnSUqY7{^H6v`0=2I)aw0RSmERtci7pI0)U zPHy*st9?1M31r@>2Df1)8XWp$EmUZtjdnp=0vvO@Yc`R=YL)hDJ zuFq$q16&slOLwMbmMhG!^WBOUc9U(Wliu&g@PXHu_k1B*+mr{4uVxHAh~GF&$EDR< z?DHR5Zt$Y=CAWu@lHSd?jpOs0kGG^9ftdG2w(+=*t58ay(VFmbl4dj&vjSlLYk74( z^I>rnUpAuF?&ufYeH6=ceZ4OpH~s-1Dn27sXBLi&ICi9C4J9^W{ydP91HdMuXI#y_ z!-o_sb_{#---5nnR1}A+!KoC3b6xFMPUpXuzLOu~ zRP<#0mzR3u#n!cl6+W^@T-d!C6FBr(_~_^SW7(!_Fo49f!d{OY;8Ao(4P7=|7MB-J z7h23`5Kyiv^yQH#JX*%KOb_WS%8!VOcj29&EHYF9g5cN$lT21;d>(!h3gP68j z@+~eXPkf{6VZyIU8kmas;X`YWk);F;X`g!`*~{KYiJBb zFkBFRotVzfprWT%IamMAvXS&aawx9|uIJ_3Jj3KVtc{x0C0b^X^UM~>_I)RodJ@5e z5M@EI@Q2ZQydy?)F43G;7QEhxu`o+Oig+JV@1e8}EI zu}I42&(B&ly((DUZ5n%HkyjgCF;$OF)^NY&ksf_sS3QQBNyE zrEo^HT9sX(qnvXKAOj*c$jdSwB+ z6Vf+8OBez7p?bDCmWS50Ra$l4CyGbGfLJo_9$x5@oiIMxSWDj?|}H2 zwWPO#1TDO^%87(u zRcCMtducW9$Fv*>H^sYe`k1J<>ZTe>gYWqw&nmB07t~WEV0k%aG?GzS2N_<25w-Wf znlhl~Lm%F#PA?d$MGW(}e`Q|q?SC&iz;CPE;{wOLBaOg{RoJ@PSA&$&Ff5JtbrkkV7Oy3>Q=M;~9+>$HwVI-+z5lH2B|;}S;BOD_mk9^jPD;?(i=XRyyA8vK z9h!}xTcwl-a>U9@;Sp8|u#f&Z3=@d9hdyfJ%!q)aSy*i$Qy!O!(^p5%@*0ksX3KI_ z9^Qp3cwGQ|$WX3RF=g0}pkGI;s_)LT^-z9zoZ+ zxXFSDOcC<0H-W!td%jPvQ6aI>!Uy$CN=c_53c_@C__9-U7av-ZygLA(W_OcMdQO5w zL{7y|qoBkzSiTxNNIAv}z@*-bvIY(_1VGApaeS#^5PWhLvEWdXk0aq8EJt?x?wZHs zjzwu~1H+K4;HIeSC4u!!tNKq5gfKVlwFblkhl;@oW|wV~Ebxx0_fFIRlg#}r;9xvM z1=Pb68pX~l=JZBkbEMw4)9Dsq>TTc8Xay2DHRs7?abIGP?JVLaUWJ0IK8J3g$BbA% z@vnD7P+tW24Y4pH^N?s?MILyiC+e>wZOsTFaEF55LQW350tMOzF z-j{?9aSUYM#?F4R-R`l$4iV~EH6u_~9xQO_e<3YCT!3CsK`nK|d- zsW8PdI+3daV!D?G#DRY~2uChFwE*mb)gKR19tIy}SC!E`vXN|0a-PjZ*Iutj18($2D81 zX5_i0oIQUHCEM*8l=IX3$DL^`pVDK6}Hn&h9?lsBccW(5KB3vyzueJ zCr(m2oCUhUBx0A0#ndg^*^v_^a}3%njFWlE>h&ekvy``*}$8u%Q^oo$V%2=*BQ1=FC>!Jgnl zd{HO)bd~)&>!t4WpWj5I7Im-l-BAGC+0{&ne7PdHlEfC!jGZf*k)F@knIM(^C>>5K z-9xYb4MP3yfV9ZN>JBBz-k7#Uy0K>ONb;MMASye2y@-w@O84{-O0PY2dop8=n8EHk z#t*4-T3$G`dubAJXSDm2x z=FDHzGnA7B;9qaxD3;IST4u!FAbA^?s6~ZaS*di`7eC4rnMj3JL^tjE;&zq}CJ7Z+SpEW<&1fmtl z4`+4xKyAGw6X`Ww4K*<~#q}^rQl+gDqw+VkaEkm>k<})oh;n ztp~!R5}TQ|wSi2c8T>5o7@l0T)C9TeX_!!UuzgN#Ofy&RR^~O9_;L|Qf=|h~RchRu zfwvWh(sP-o{ThonZu%t{-O^u>$6!OyyS;qVmD~=eilnpekwFBYm$e) zHCQSdU&$9~;jvcxOKm~)d>EsCeQDG#--w9*h1RC{rxpO$T!UyXI8<L1?fw(v>_ek(G-P z3ej#zP<@Pvtz;}oPOS^rEAqI%Mtl3(rl{8uR6XxfuDP6$X64FdK1NvK&CuUpB zH4nYUDcy>P358RH>P0bT8V*o%Wnam&FO!bZuU3bQou9BUG>BrX`JQHCTR-#Pt>`6| zxPB#vK`uyK5-O4P&{d{sK8k4dUvPk?t}vgxpYnLB*T6kjeJ6H^&Cl>Hc&5PenCB3e zf5TDqYl{cH;LG>8-cS1O)Ba7vZ(qEUF2>;pWid$7dZQ_J6CIXRS5FMNyx(eh~@DmdhJF#e14_!|yV#8yabnhtWB7&uFM!SzLGL z!NBdS&u(gSlB8>{jV3-i`Orz%57h58{i_Q5TLGY07;IvqIyNy)O%RFPX1o!zav z-?Y%EaweXnv|w8mj(Ih;wI(@QC=iwHOZ}LcK_kEFGO$kNSB0pEnO?c2qt(6NxGp!7 zRgjJ+cOUk$6;K7qx=mfuN%N?Hk8rKQE()Jsv%Bd|Oof+K!mshAhBmeMs{edk651YP zT4Jb_K5hLJ1@c3S3@aZ{Aa4w&SJ-_FU&>OZlAHc|^OEy}5}!ykOVr$%%9fw}_EbX@ zWIXV9_M(2@C99KEvp3BzkYPtdwitZQzAGP!TuLJXsg+ z-P%_4b=1)bt-hmZ-g=W@L*v^aEL)6&ekA2N5WCBWlR+;Q*q`vKz@8kuLH85NhL^&a zP8|igO2l;Yh8`za_41`buTMEN#qx@X?`<53ZUu4@&41$^Q5Mx6$Pbwtu0NqD*gQnp z^E`EkC10MKHkO;bva4bCi8jm*-}5zWX3QL%ydX8|H_hmqvgoMZ zKSh>g)V+Fq21o&pe#ysM$e7LSG8O8-#gDSJISL_-RZeRfZOae(WRg1H(|zyz;;peP zeHx#dz&FQOSz49|21#B={cb<~lG3aId_05auZQ$P#x5IaO459UYjZo<7(69uIGe$^!wXca!&pQxHloKXr8iGFr-=%N%H0P< zEZmponkWb?*ccm!TP*xJ?zGZz7SE*~Z#gz54UtObZ>`TXX?s;!twL-c?(c4!>1Km5 zpR1g{;BH=a_)rRwa}*f(HwzaJ;pCecnU{C!>&qsq$n9y5SL~<6UX(Q4zPq=MXUJBc zeHmJ2m{VXsoN$a-dP*zzSQ34^nz5qUfnQgTgDr$U^&R;A`QBinAKE5=fifmV zf~5Rlx~FH{x*o^EFFC>bM;09&z$#y9*oO4aQ`CHr{DwIq31OpZ23B4ev2}C|!>Zpq z!iV-&KOb%V>b{|`{?WG-cf7UrUtZy*7>#6=;|cDPlxvvjqwbK@-1J0bkXcU@D7qsC z3qM)FL&W@+kYgs@jqua1DEs|AJ;&cQTATIOszqn(`D; zG}HFXu0^_T8!f{ip@T(FYM}COmu-E!)Fi?1{XsmjH?w{+zU!fa?{^m8*txstSxg&z z(i_~h*|vb*R{3^ibtLV9Xms^s?MnZ1(gaV)`FTQmIkdSJUpRp}6LO)R_LDIvGiaGl zdhS=~{aI-r)E^p$I-{%yND?SMtGH9}Ma_(`kHjWZS0%F3_LHOiOo{$}($v8|E)L96o%n0WGfE1D~4#}$#M>@!8xGXxS?LEJXGb-H-Q zB_KW|9pq-t`vpUf2wsQnB=6&tW{OqOO}=82oA3gN(Iqf&3a{m{Dw+=EhY24#K$q4n zu>}~g?4tlbjt_&sz^QKbblV<3TL3a_$Gk)%!|vRGw!(*CGi@)Rg=ra{5;Q$eRDKEw zcKv`n8G1ZCj3E{2&9H0gI3z#jp%zR{@_jweKb>{$EWj}MlP#Rr8;SWWu7(fjc+mGe zGE?u*fRvSQ_hxfP3nWkM?}M=@%EUYxd+P@~inR`aiqyW4i++$xz}2(&SK{{9QdSx} za6|0zOrt!=*Y>7vr$r$9c+X8|XOZ)Yv;Ka`rC)fJi{A(r?%ul+ezVuXAYi5)Izt!iw?D|abD3lXnMMYWuT5(2W9qAJW`=ia z1fhb+L=SD#D|L&w(`UbhFl8t<^ZaF0;RVyZ&XHFA&5m83%9T%tSz`*R4Qev!Ymih1 zwG$I$WQ3t%#Lx>IZ0f0$=!Ouq?T;WuNPm7uiN$vl00>M%%YSMywFkk8J>T!I*{@H3 zigAF%B;#v3VE?Rr<>7okeWIIwDNo&id}DLt_VZVXtk-En?(oYJ`}S7bfctZ-Ewbq0 zU75!8bb39=YM-0vbR9*NxZ&=#lFrR@={ixW9}nm%sHd2Xej%nhj>Cr~h@ety+(riSdmIL!=17eY}qF}87tO`lJC%&%>fqvToG;NKa74=H+oWJ6JJ zy^q%vxP*~nC)P)^Zsv^Ww`AWQ)Q^w2!kq{lGPnVmRvx@cE{f>-?x)R$ko6%|cz|Q# z#>0MzpRyeyOL%0vJ!pnON3YXU4)wI?hO-$kKI#SGD2hnT#l5~HOYi3d21qTgxUTrb z$vP9K;Cy0UU20wTyB0iVBXp>~L(9L0nNV2`D?OlE7>NUP)r5u$!IZE~j%f-rJ(Y^F zTPZeVG9j&L{ScX<=RAa2IDIuQ0Pm=+sA;-;=K+UsoaNho;MZGJSE8& zEIu=`r%UJ7D2$LrzBLBuI59i^3=+;V(DF1G4D(R29#CYJ{>~A`JhtJ%pQsY0m&(BS zU7J}h)H9HffZzQF6|*u#FrXyaq=C~O7#r()m14>87Rf@mqh1{K%)lsCK8WmN7|uQdT3o-AXf!8XYAxX5NNcEu~5K zQ3Wpm6H+F&&5t-dNSn8nD@qNLw=J*?dK=HFk*tz*O0HX@QgB$Rwk{=fIH~^Q-n=y% zQm^X0i%_Z%)l|cW6)p{~J9i2^VSv*KVC+E}7WM~SrLz4{m1US-BL3nwK7&2v%4Vi*~wwu@yNss85({n(|w_FL*w^-Mtl8Kbq^=yx$lgtxD zJMG<@WR}!-rhT0su)A_EzGTWKyb;MSeJH#Eus2JKwjK|qkZXjxEY4F;EzCT5v`B_^ zZ%4Tm_|%ml7Anoe$4{nC_boL9{mm8wNaFX%3PW{@_@HNqgg~F@&Ops%^p+p`{MTRB z{f*v=M=kphprhtuoiq&HA*ad5`uR*0=1=l?e7t!dOs6&n4hY>7anW*fbGhNU%U%*$ z;3+_Lc<)Geh7+lV@kTw-QLw@?Yg(n-(IphsXKN&L@yy^HU)<+mGdHXE?=)$kE0_cwcMaIqh9!EBcv?uRDL;UpC?6=zYi8pf_IA8S{MSBe`#F3jdF;4O2k z$6ncmJ&TkpjRerN*D1z&Ivbefao!N7M0`o#rluX;%L7Tynk@$YL8k1PWws>{F zZ>+r$6_SS(RP;qVNB(TCg(G?Y#ca+>tR|G+=xC6B;VAR#<3Y+rLZXW*vc>pNVfFRz z8k^{iq0Pd*EZiv#*xU_`x088m^=R2#fifM}aQz?YY+ zTlS0O%qOHDWQOcdnmZQnU{2OF5fcL0zaTfFb-&>>$?|6v&9unrkxAoO5B}l51<_S| z14BoP>lH028`JG`Y!{ZgPeurzYIF_TKB9eZI5Kz0Y~>-|uHXlV^@W z%~3U~>V2z9cO$rF%f+8fWZe8)md~b4F^}n((>et6-4F4iN^1+{_&POrr|DY%%IL#o zBB~+Zak12cxSOz>_<`MJGJSYWs5tvUWRI#o<99Tx{mH#m76gr#vu=58F!+kiGZ582 z2FY4AxBngD0_3=+YKNpQu#~0T66osbmuV)OohwohW_U67tx$S zKW98kFV|Q37wJbq&yF16T>N=lS*S8HX4dX^%QtItbs$qM7Kq%n;z@|y-}2jp|3T&O z0Z=??bs(&k7^#I7(}q!eWFdx9deNH=30*B>7J>gssn!;w=z-grV2XxCXjzXP#O z5vBUZe)K_w*F!4voFSg@x7imXkE7Qk=cUp*J24a-Fb0*ZJ@9#jNO3izN2_jAt-q%0h?Xa+I|7w4z zxoo@_uHl2&+u*G}K`$FIFRQ(jQi3olmSCNEOY#Dd9Us+q?%Uqq8BnCmr{mOO&#jz9 zV`AQaqhIoms*p6_)wz7v6^~v2ouFCacH@lr_H7HzwIMs(Ob3j1zCve2mzF$#fo1WV zx{m$Q^ZGNwNb@kz#wbwLV}}y9Zv9Y)se1TlUq;7|k{fG_+p_Z-rO=4Cb&a8Dp96m3 zfGW%Bs!|OfJ?X!C# z9JyuullSj-sR&Q`XTw3=Dt20fnmTq4_sH*&7DYdbQ~AN&T^JpFeISKF7o4mH?OX=z z*amDeVGn!)5Ly1HCY1k}EY_nxjgj$td#!OWGUh%D9+v$E+3nbe^dA!&gg&rxyJ|d} z<7dG5Et805@12DhOxlORzzpVs(IZ`qZ|3@16Pob@nBOMJCa?-nJDWFO#CpGi#$l96 z>H>3OxOX$JM%q2+CsUrjLzKJU`#n>dJY-P7tZ@``wUnn6E8P1FLZ>5_t1QzgW3w*- z_FY*Bt56!piP70?+`cX&hUxff8AZgG*mmX19>B9m{IbkbXMzbw1@H3h292$ivs!u@ zwE2XOo8zM^!WtCtTxgx2eyg`I+g)H@Gedkgu*bvR(y6sg6=t>azBmWn4fc%jN^;^s zK(k$gjw`Ps?T92PGDQsNq4At6yKW+SIu!xx&p7Yah-Xr zJcPSE;8=gswoJq-#}eN_Pdd)5Or?DKX}nA$(3S|kdoEt5LJah;e=P*^3fg5B;rv{Y zqfR*B@Gu+y>-*a;@m@S3d358SKR({i#=|~(5J$^*@f+X&7Yn-hO|Zg=?8|<_Bx#kr zmtIR`it8#>xONmh%cnPg;GOgcKu1W*Dd<8#@(F{%YAcTn-Zpxs+$9YzKUDOtL>Nnh zf7rhk0sAXSd}AX11LJ>v^X>V(a4P6JhIrLqE%8SPN`7d8R`vZX zI^$HyxcR$~18J@2bA1f$rRQLb@zb6pK~O8Uj)V55mDN%%!!W1L&J0)O$u~sM2Mc^+ zUWvYbJeP(xrYJSWkusW2JiJn%*6i&)?u!VqgaXZ6?`@IrjjUh#_83(-p~jXfK~_NZ z2M0U(<-gjrD)F$MGF(F-q{nctK^Yac8#`6P6+Y31$?v`W7fB6DygTYLY6HA}qW=m_}w2nLNN5-E2sDU?}eLfP#G-SWyg|f;up>m2%VP`fB{hMEP3BY=#+D?Q*5Yl$(&x0@zv5xn zMQpUwl&2=^fho&R9_}G1QqD1 z-rrlHG{~#TuD*|bEbUL0mJw@HXx_)owUjO0TPjSDLK|u-^}rcPwe3v@Qz(+*5CpiL>`4NniHU36%G~mH+Y6o z_n$F{=vF}2V?}p-xNNghhf#)Mv+fCXQn&-c$ew=LBxl%?kJRd6dKgPsL!W=gLZqA$ z6#gYqW;xK+48P4PqmOamW9{QvfNQ7@9A%OuIY>I z&I1oo=VXOTc;P?mGG1)+WZjKxlN%NvtPhtrfVOBHXvgtFVqIgwG%LpWm2J|aCtpAR z0nGYBOjaj%%TiFuRm%Q~)=m>-^o?)ZeCDJ-_=f*w z(0LzQsRkUgmAO+HeB$}@NM2pK({~<5C~NtK6edZz@-e{wXXot$6S>m`(7AVy37`)A z-RTN7Scr*4+`D@Akh&%KVZ()|qvkas-MaDg0}tvjJqgxW4%P>m&1r0mPCYq^h_|#^ zsR?g>G|Vqt4|a|*4815z{691nP+xXxyiXbLuP_IF=jq=;BU>h9_(GN|js=Q$&Xp%s zW|5l+va$KTD>&j+B*-(qLpWV;jr>I9Rb<^{*T$-K)`&mjo5(tsXE^@vb<9$>1_cg# z(rO+1J2TmWfxkN+3oy63OnX1M&KOzcIr^lQu|5#TZ1qL3UuSYZkptGgNQ_bv!nP`2 zyV6_Y*Wj|%l8$b31ahpA(~O+YyP9MyzM`w1v}U8nc%UfVeJO_SG2hWoS`?CkS+2XD z_)`>FN+(fZvoFFcfDzayJ7;|hp@A$?-nAz-bQNS*NH%B&+nP!6LQ%oCtmjj;gpbaf zMZ@_}3Ebf(Hira1)IJ26C*zp7PNx-!l?OKdWEMCQzB`T>eeiD ze7VSaQ#qZj>IrfKL_|4~g$cTzZRjFfQ&~MG3(^^#kE!1jd$+Y;?JzEoVl-02SHPhh zQ?bo7C4n_X+BP;ySYIv@KU5x@CkK5E43gVf<#|)&y!hcVQxvkQ#8rMC?%^xbz?W^I?B{Ib^qzc zF1qPJaDA!NAW~qzTWRA*((#r+Fj7z^yA67v#*~pTGCvER;sgSQFH+XyPEs$O;aBku zSh;oO6IQ5(%oh;D_>46|LLT^UDO2GFgi-ZL*0-%c$54wx>*41?la$vvd>$z{rxSTc zc&wKX(vQ+5sB$>GUBo|TB!isg_oq}L>?>e9UPad+yzU$Fxg+x#8~!iyGV|iarzkX; zq$ZaA6Qb)<8iAoW7hP*W``|?tB5R60zn;$W@meW#MRyh-Tg4R5nR)s7;!dV=3f-@i6*TRQ(#3LIJQyz9K0PW6d>1NdeBpd{ zruD2q=|Aqf|GfO-*`KI}ooNvO=`vzUPWALH+!mFI%J*hE(_d~^gtr(MIFCbH+eGk5 zfR>AcSE?Zk^X$7y=wKl2$s6WDDIl;b!?%R5O#4#NC&ZQNJpsvk_KRoq7pga2eQBQF zqVl@BH%Bh4gep1U3jeZ;!lo1O1i9gpz+y_@;naNu&RDIPUkObVgXP!n9HcU^+J}I` z?1KnVtymuIkh(jfi;`|^R?*J3hFTWXXEhMcQLQ2>7h!KE5~=r-xi#7Qk~rnmoN`4BQnEAVyvGHn%8duyOBd=256AlsJ}s zY3kK^=}-4ySIGw^y)vAwJse2&LG7lwAX)GvrUkgRF3%m35%yxNUk#!fmCy~ciY9QZ z3UEicJR5=Ax+M2Il%c%F(cTvr&fShGwR>H3OdqCI)RSB$8o`*3ZGY-sM((-Pir)zG ztfMQG^Ux*Y(JtX5ZhBuN-|bxys?$(PvPbCCwI@t8N-Ccv^2(h?W=ev(^r5CfxalS} zm=^I;8$B(4$L(|(DT~maV7`!(K&?Q#4&kfyr_DXLredWAk`&9kjvrO6;B#Oc1#ugo zRTQt0Um;3r^|;D>&h}hgmR$wHwHtJFGuXV^6y`_=+!ptDHIHQT?!Q+P+kRmT#4tKl ztBr*OqR^A8Mf@QnTHwa;pG+p|zB|+bABPFh1~hAZ5t?9vU{hpU2}ezt@zL*~K~qQLO~wcuP{L}fFVwOedijuv!Zj2h?M@9bHEa=FIZGkJN4ejl znK{cq^6)C)VKFBcmIh|p4Bg>bb~1B@NQ=x&oebZ9ndy;sUSfe5)#;pqyG}dd0Mquuy+|kYM7M+MSIboAUR~j`iWe203c+>3~Aw7I|oNf`~}lyvO_M5bh~j}td+z?x2>qR*J-IczY5;Ugt*_D2jRfJPw38q1AEt=QKpI(P*3KSz+f@9K z(oQj^RgyYdG(4;|wOM}M$hgCCD|kfbt1`W>uH?>{zRkLA_iwkA!fmFxt!Nn&1|CWi zf9r_ix-GkjUFwlsJyU-&`2k*(lcKP4s(&mdcz7g6Dj!n5=U_UQv=z5PoVMX{pMqY_ zCcOcW#yhe(is5QsUO%V)nU5z`=T)<&ydOY}iGt`^SCagWrp>+Ot!XlMYw5v~=2C z{%TNR_E=G^Jc^#&dbqMzPi_YdaK!{H>U$LS&AF2Wjy~Lhh3+p)XN~mpqN}J+z;Ma#8c$^J;(Hh92EJdZc6sJI7puj-y)2>gFYIz<5!>*y0zI^3wnGd#&pO6KR7EJL#~N4c+Nh*j@rE(D}WX*50Llw z8?t8l+Ht^Pr?#sCu@*LJUl-k1^d02xBXs&;y7{|ZW0S<5g({nZYq9W@)m1%iZ_=7F zcX6T`#5DLyBQNt}X#FUh71Q{TlG*MS93j{g*Hw>}EzlVz>iKp z0|b{8v?Nw#D_s2sLvDi`ErC63`d*^}!$;5s(Whv6!xEuxAuX;HgUrcBGl zG$7nkhROaWcyz_ro!fqSYFfUVV0`Zvw`u9ESx+5>DX4AiXwR)1ZXUl-%sz$S%y=|C`Tg#-rYbV=g=5LJtlt9?62WVS zWfvutXFY)xcg&u4jx`F~9i`*V$WFaXC3rrvvsdu0dnCE!ZrGsp3yc;}o5Y{Oa%JUx zDrMZgE9B6ji{oRKW5ILzZLJxX9fXo|$$jo?DujJ4wb?5o$$(4y+$=r1x)fawac)u$ z4Bm=Uzg%dYG&%OCIU zQ>tGlO@m?`xqIXfhZJ3;nYspfcO$Zh*h-aGs_~CC!80c5O1IdycHF4n%o)RcDn9x# zPcQVMK5Zi1H}O;T&GzP#5ebv`Hut>~%td}Zj~Er} zr{=g8^DpthTQ3%YBvfvt zsG!&>e0nISZmZN^R`o zUx|A%Hjq^>w$Fa>o-6q z$N?bQkXnS$&=TDhu`^9S(Vr?m(h^4Q(roVcODxi-PVQMeu(v$byK8;JpQdGy@hP-o^Zu4c%2Q%f%8uKu zZPaa%N?SHE%-%khMBv5zzL@ve*=*&iW2zU2+q0qWt8zjunCr6;D+p^Wg8&By8#%)% zv$5a$!+uk~5Ri-l{f+O>_=c-h743(AKL1L0){l%~VOXXh^9U1rUA*_jYjY4T^14*N z#bgoF3c}>jmau-JDwI@qdpsdCEUU!MI>!7Wq)swudTW56CUJCqm~*gZ+{kHL2XN8j zHPS%Cmw7$XEe5(DaQuJM)tNUjInsmFL1zxST#l zTBeqGr^W;+5t(=r>A+P8^Xrk=R0KGlw8Z+lE>>~wCNW^pL%UkfQ@7xIseYmTX`ExvAVuheUMzK6$0BN_gL6lNUeCQzP$`ye8+ zd61>_$T$}%bv~q?jTEM$@-EB-wW=ZAm{duBb1?U55YX66U1h+p0Sou-;u0mt@#`|; zjSDetQj4hz$8Vl_YKw5{`1^cmtO=mOXB?hU)&ioflY2IG_rkE|y;!9$7uk~~h_NmE zTf!4wvUb`lPkZ$7uw%>pKZ1ru!uNyRGJ-59QZIWSHu?IG6UU{d&aPD4%5<{8MXi0y zjs3WPe7eTARzt}VelRo)2W2kJT;iiA6i5}Qv7)QxHEQWf{N0&dbs-1SA4ri+FvMKdnq+ig)w|4tQkR~05Xyah6nN4)`DNmZyCljlEg!aE!-rq_>5lM-qW zN+?mUovDoSVcnev?G#4L73ii}4$XQRYM~B|iUsgs1U9%F&*m$ExC6pOnTomP_ghbM zPg)&J5XvBUBJLaZLVZ5IhIo$>C9{+Fi5mLLIwgz>5-S>novmB@RdvB-bt+jN>OZwa zqRqY7xv)w^AufGntK9G%B5F7CUI4cH;Za>VE#-I-me#iOqG$I}>?3KRS*Vxz@=5at z!gb2=;4QR*R{|A%Q}nq&Aqs$%jd8TyA~A3plH-mD=8 zjD@>3JP1CO&nL5R;g)hdO+u#S6Tw>w>pw3uDG~Pa&CdlQzW6{*NZZ<4zS@DU0Ob8_ zaw|t|u_i;nkkBFnGVt zc{Dv$%x-D_eU#OPkS6;(bY)BovRL!RalliBfw&c=UQ+KNQ6>L`fjaMFg z>I(}eUrkR)c3SX!1rskv}JOf`bu*AJ0x48bpj+D#qHXNr#|rM4olX$ee7vO zmcq@rg&7jA5}{j3Xd$@oeNf4EW!=(jy~?VVz;;4+ zt8?~y>>HV=pO_#XqS!NOKeYIUyhqfLNRBn(T~TN#e5E1|GtKh5+;jxzF13EzS$PB3 zJ=;|Kfqv@{_wMBl{pa+=_*r5f(Ccd6&mMPB$S}7AMo#W+iA^+otT(qHX_R30{u(PT z-GCIBVBOQ4qi8W|^FMgGJgup5N^Hm%-Hb=X%+C+rQG7D{UCG4X;txo9q@AZJYZ?Up z_^@GvGhFrm`= z2L3d=uN8CI!Z}^l-*Eh!z=bta#LH{siep!s?WN$vkNwP7CALLB`p#A{K(+{(M9F6) zA&+|OS?#MV)YccbAKE*(FVyySDfj70Mg{kfea-B0*2ze^XhF|IM-B5037JKeORts7 z9GL)#IZ`^be+U6zn)ih@m?q2Pt)|rI$PM}t_-0y=bI)_YW|M3S{?N{ko;PTSz8ADq zTq+4cpk;mWrhBY5gY}gKtuvAQj(L~z`vE0qDYG(*LT6T%8RsLDHIB#uH>~NkZMe8?hI@4kS0`ee6?yJF zb5$Vw>asT>mf{Y(U)!GS1}Wqv^4so@>tz)Q8Dv5tV%<-xYw7V|_53voq5fTrvu$gL z(q8^Y>F)Eu-i3ts?c`lcl8M#z>5!9S<6vN5P>SEmLHgHFO;m;tv~};{IhpbPW}o&-@V838OoJb^`6b9t z1jtn>9>ZGWPjZNJy?+O!|B=bGdU)fm5Aypa*Di$SS;k*Z#E--$)5_DsJ;SzB5@}*i z&q@r%fht~U$pJcz%=vgF|LHTbMQ zrGU>Y`KX6Qpq5P33g4uL;3i|1%aNp-d_^hXk5yb0r~OTC6~N0~U5N{ur73KIo!$7+ z|4$74%QAk`sRemr(s}K1xtndFt9`gbQZt?Ty&lkNXTTkUyFH1-b81 z>7Q-0y+O$unr65$&$@P9uT>sE@{Daj*Ey)5hif7>!@mpF3r#mXoBH-KZ{bbfwKQ1G zAtD8)PukmR=vZxI_pdv)wpeX-rOlA?x55^%hj5w0b$a4lJeQX*|4yl2u;Zv-hJe$- zIe1u!_ag@XL_;V@ZxkI!z4^ei1$nMK`%meyzs9?aao!)qc_kn}7k6FyMwd%#@d?_2 zLwAlNP-O-DO7dQO9!Zj{!JgA(78~DB$jb0sGHNm>Zz^$EvP4V`$-S zU-;VDRp8?38{5C4Hv?s}zEr^?Ad|UXh0+@Y(lSO*85qvi8i7`>pYL`LYJg}xkNv*? z>-Frs!~sn~prdaW93>)nEAq!SJKvO~21Wzh&l(Et_jJ~i@VKm=54kN~Y+?gNU5{x~ z>jJo(&d=V@;R-oE2}m61Lpk8SDI!}GyY+FtGItzs1l!^$|4{Ha6BA#m(DtSNP0D@Ti zr4y_9;$wP1_nzVvMlL62`uYJe6eabLi^c8Ph0uoL-yl=Kr;xY+&Khpl-!|3qFG;zdw-&zAer}+wgz2 z#m*JrqRi+s{bHl_Z}R+qk;T8ZeL+&#Je?Ia#q+<};uCQov4M5G`XK86|GmP$33E;h z_~chcBX9m!Tdb4>1Y&zY_hVA$@3rCIR`_>9Q7{nL9k1rp{;#(9r4JB@bQ*%aFZ1uM z?!T|_Z(ryXI9YY_bT$8j!v6ghLmV{$flOWy*d710c34srp;%_&=WGf4X4*SJnP9G5=T9{@YRczpD0ME!zKIRU1~B1B&K;WCfpv_je|6 z#Z=hqPR2B^@EfjG>uBXMOS=CMbHE*M=f*Gv|M2MVN*nVLs|+Rz$&dyQ#woKZaHYGp zA{Ijqx&LWZ*bQk;H_H%fA2}c@I3F%U`D&Yf>dv2grE2UUQCnY=@heMAfO}Nsee?G# zTDg&RuJ;?J%dZ=*i%kmTqW{ChxKfc6wqrq!fAs5?#ZYCVGwv+wbdjq}QNT-P6lZPc%ds=k zz!bwE4NNLMx|1pv$g|w%Upub$>LfU=QLi?@`aEFNzVYhZ!LIxFo>IN9)hjo<>|nMK zJ(CQ@KEv7yN^y%*v&O~ag`InwV=n4h0q5Kq;GG81z^Mne2>%n-#^qY+t-$bo>mncv zmtp9QINfbq9;r95VVWvY{=iARVs2Zt&emG4wC1qv)&5r8oSVAeRQHFI|#9tzc70ne`U5Fsata% zQg4Gw_kKxmX6H*B^Pq=V_vhQtBz{Bl#mI7|r!M0EO3fFDpkzYPrWfLd)9oOF(xw)) z-cS8ZKX2^9OeL;82aL{(v>&gV+#0XrO60$Ml4Q#el}MpyVm^nTm>^_6sU@{ASAV63 zxIV7*s)wl&yAe^cI9K6zrd^nm3W@^GM&I%n@KnhL{BYv6pAo9cP#>3mVo-8s#&Uj| zpv93Bu)pW6TW}om_(j;FLgS#H=a(mGR+a9sv^5>80@#C8@KG85*C#%Bk>`e|FALZ* zM-V*)?$=4~*_}p(jvZwGx&D*3FU|ObTTO`Hj zxxwSXVe9+f5?Guvao(OeVS?s&QEy#iuk9Fd^z|h+xH@3n z0vPyexgE%JU2X|e9q!GjJlY)n4nVv=ZX0)-*RVc_bZrJ&M`9;eo#rUCY6~LSx*Z+N zs2_yMq=(<++0`sf|4n*VSJNVVuQ88&d*mx_h)?;$vs;L7$O6d17Ln(74y4v zN>h={qPvnjuFtPdDSye~kywrTfP`A;DJHoEA6KTlDQJUZ)Oc^j@J-)VSq0m?2_zGN z`7DP0(#$&8b2*DaCS-|Z$^&@9q29k4xKDq2(abil=4OA{aY$`f4;y_%^trA6nax1^ zf*+S$`ejt1D3qBSd#ysr%Fp(YY(;-YT5GwnkWTHRRLVhrtYD3|1LDFcHZnFlVWRK* z9Utsu6|zZL{c1(}8unY56?vj1fBB48?);lGpZ*t3DQ++`WX=8`7=NMN zV#GY2O`Ko}QCCQ~bJa4hTxEfhkJdQx?WY|y6c3UULTO$Qpx0NKbv19Ra9&72A5=d?f|9h??;eq5lS7-0|#l(l6;)vVxB zpH3UP52?#o$(d!oo*(=7`(+Bdr5iKTh&vzAX|VGmp@T*3wOyLNpE8CI8(4QPP0m%` zsZXXnX+(%rEf{_~3uoZ6vOZ6b4eoKF&-57+y84Pk6|O}go5;e#?lv>PG?L?#rrS@uHU3( z&T;sy`?od)g}veb#TO@51pHXHN8tcgN4h2YOMBD2&d_)MLh`s!5wRA0y-0Z;)OJ$s zg!AsskOpP=??`&?t8$(B)_D7$?BpSOo%cB)GVxd78MiSCwu^u88#8=b_dKu8R@3S= zZ$-ClmfH>6oVyKkp0_PMM6DMLOL9^&*NvB9WLwK{D~3{+)jHe&B)Z(2GK@;A0p>qr zwge$L{c#+T@D6;IZ?d`tawYnE6*V}k_E#Q1&f<<4k$P-yzp0O>aDsy{qCG<=NDaEd zf$r`jSAq=JR+4Jl6X&$Ay#*kr8B67M%nGr=H!CjL3X2n_HbEdC9n{rMN1X|ZK0{pN z~`iJ3BHghAIAq69uq$;O;i0hs0a@$`}B z!Y3}EMY7QrItx?=FgUs9DY3mczTj`C*Qg)iU=%Vg_af+QSubp{u^%MeijrB3ZayKG z4ZprPsN-gme56-8x7ZteX3Z67edx0y!&xxz9)g7`$l^s&#-LZFj)cKpAOL`26fbmO zBjKWY&wDGAZTXH;(kG|s+Tm0|#E+SXuGf0TcC*X!Fpl(#@+&C0XR5)4`}eM>6RVrY zS&p)E=Rh*mCJhBp@4g@O1J%x#?6QcH_qA(PC^@dBGbN7)!sA-~Y(X?do5IYdY7nzc zFcCc1ie9%mUV8?m1mSR12wu6isLcOEHv(M_`Ik}H{lkEJPi<1;Om zB9D;xA#LgbE0P0EIaf~uk_`6jI|5Cv0kb;ZGKvW6P#(l7iFLX)xVu$SBf6?tsHuN9 zD9FwNobc+=LzaRx8aIK)wWk@uE`AIE|UhZWD)?(!Ghr*#|lRualPN$l;9mgIay0Sq@_&Sljz`2wc$Zt#76{yR8wgIXZsDq;TQX)wDg-xQUb{=Apf3RCpry) zb`o8JlH@8|NA;64Cp4bSiJ%i>qeaI{jUA9CbSAC}y~6Bm_XIb`%L@9^>QWm$Aa;@a z=a5fSoQ@`22PrNcy2$8{Tw4>yax|qnENbquU64zY`$w}q?f%$t!Mp?izBCi5jtT`Y z?fgU)C~w=DaS!(MvnJGSCos&e04Ggz)P*Cb`5ixf;Dc%1P41+>wH}@lwWnsfV%kI` z2%7QmF~ap_u(Pc*UbB;zsn%Yq0!qOG4nK;tHFHLvu<8=cT_3zI-l~^wJB%{d>X^1( zX%qh@x4W7ZT#K<k#GrFKsX@m|Bo& z>DAXM*{eH!-z2BoWvvDdY;rEx?1z3O)`adJR47kLfU)heCyDuD4o#I~R6~)*@7`0k zQRheBVwU0iY^tOr_tOw}@8_GEN{}S@`Zxz%Xn0~mja+T3uEo*nuz5VG$Nnc~pDX~s2i7j1 za!7_;O>mU-wSL4qj4lHIYGMcO6L`a%7aalT(?)Ns-DYIfOmGMc?6dC{k96kJp=8%p zxN^%0tGimX`b6X~nyo~>360fFp*&;t^W0SDe_NMlt+b1>nIwve>5k}a%u+XA^?@NO z$qQdsNqPC5tCrDi%+yH7Fka@=uz|e4x=7Z~=ifh0`5t_jXdq~a5V3_nEt!(BFEeEp z`QXg`_IEqWgRon-DeB+f6L=|pe!QtghV~KZ26exj7B5-*e37i&(Dj^qIA1-RUnzS8 z$t)J7>ThTNrtC#*s_<&Y9n#xCCH=Up03emxIKet2Eq8BgQUD|kBf8ZK0`lp?gt-_qPi+=`z9=UC)~8r?9na&e`y4Il*E7PyHtTJNu_5XX$5c3 zbO|)Z9|vS?L6a}%GzpQJ3ufG}^L!EwhPmSnZc7~({x(c;(U#r23!$NVNCCx%h9DHYxCq|drO zeGgJ}5z~E-&#!c0zdb%ufURoUzTW{V>BAk7GD=?gAgEZu>tk&s8I#x6$;MB$wM&OdBpykxuL z2$8r*#JnzNvGizTJorhNEg$Jq1m1B_)YzQlWbME-_-S7ZGm-3}nT8GpYu#C#Z(yyd zp#AW1MWOpJI)Han+yJsXGeggmUW zL?o*RCSy_9x5sJ*d+-jsd+?h@Kd3*a7}J^L1yKj$QC={yKAC3v>PZ9%ex^ zd(mjAm#|Z({A-^nr~Bp<;m(b5jcG1d>|P!|{eM=aNHSAIUj1xK*d_@?y&gxuLwx#7 z%dNiH0kIN1|F;Z&h@-}zPOFL5eLNT9N@=;_(T=sz#TXVf%F)tQ@AtTeWH)Ae8FX^P z?rNmgc;dA}STbGhPTOzDW?CYn5AXVPsfr=W%R>juTTg+pR%#QL1unCGF*%&BbDZSf zu`?#HZ&j@VhlbA3VYKD#TQqUfy&Hp@WVGHv7XtZOY0t(@JZd~%JjBMc;X4aL{JYwt z$cSMs+;6`LHQWp(S$T+stxKCn8bjN$o#}tHKj7k&ZlhOMc~g8wRbq7JtBy4bdwv5= zVy*s2x9|?1N^7KQALHv7cDt|`4DW8zt?Z~EfQa98_WmU!TS>*xhb+1Xa`%Lik1Asc zD0ySn3Z{6jzVb)+$156g!We`((EB&*vSVd`jRaTE(ojdySs4ZJ%eO*3x2Yc%S_YrS zPysaCwTD=Kr;MH?-}ftm`t(@n*;&C{rt6_+ ze%*WzLKGSTf}XhILTY`^M^ve47Jgnid~Ff#wYXh znJ0kuV1)Cq1wny+N;N293i%LlV-^$M*g1Jx_pj&1{sicvY8LG{f4=gbU2~7OJwTLu zRR0X^EgEIaDjF!nWKdh~x=B>^uFK^y>xH73gj4tLLRF|P8oovMSeq;a{3UK1a(J+_ z@KAU2d*cw#0^?b!Q3W13tJKVtKh8$KBkE=lyJy*J4xd|p$UENNHvG{n%}RtgW>~j4 z8M(CXMmo}IBM4>2QDDb*-xFN>!Hj>f0d%(y#p131b*u^u)H!qQYTDcqz-456MpCk< zfM-)03J=ybjdMDZ(XQWwScTUbKDMh7(^!?PD=_0CRHJ|4t zgQEuz=SKe{BS(wI~i6qOtPf;DULY|9}ItdKkFtZFuOL4uR#{aArDy?y>4@?2UZ&N~K?yX!P(e38*h&o}7y!SuWm+rm7SQ%Bo5 zzEt{H6S~?Ye9XG|Ug74sbe(5xu*8DyRF+k@lYS}&cj0>~ks#&4$|DX@hvVX7 z@54@WXIpLvi~Y%If~fU|=Vk=U=evDB?u3z>2E+R!3}8#Ln-QPF_iQ3Luru%TE>3GQ zdl8JI#S^FZ^bPjWIBID82GAy4*3KJGC6on9HB{d612EcAsg^vbN+`8WZ~C$~?OhPw z=DZ@Il6r^P{cqtS^t%=(_aGN-&f{lfXd!8SrO84O;eBd|hiynnRx=ZCiSH|1#j#es z=e?yB7F`L8G%#LXDp*insQ7JOJawU|wWO&X`X(f-qrcKl5LJ$vuWMtbn9%-^VI&kv zcwF%6p(Pd=f76H1wgFk$ayW*!DBt5`04QTci?V{3t|^m~qZkcOxRk%&@12*-yndLQ z%QKO;@cU1`QFinQ!o%dmltuf4mPZTk!qKW;f;Wj}9prU;m{eWzK|D3T;@9}Oq+&i;6?frrg( z8{CGGz70|bN~wr{x3D2cEr@8ak2^?80d_zY1ly+ET9OWv03^AK*1Lp+hyw{Ok%7DKgV#{^J}y>h0RaV%X=!v|oy6=7=>7mrLA)+$;`Q3+D0rS{m zw3+urcg8T)9k{6TEoJLq9baHR%wxb|H^1s9JSS--=pNyIQGGXY_n->6Z;q$~M z2BY2XCdWW~A5Po z=s#P+x!h*4R1E>;Jh}Y6AiebrW4rUXgGu->{IY^ZBSkcm6zvL!D*OJ@&ue7rnKfR} z#8Eu1!0jQk22M#5W+1CTSTLCDWC$(a6JU4s*`06HeD^7+*l2RdGr>;U)v>WI;GUez zb@CQHvt8MhMNHpzhN#u?{{)xikERd%d-3be2`{rycHMu9yjoDvjne3QGM-9#IkR+HZ-ZTE|>}|C@llh^i`CV0Lql? zj#9;_xsr8%>h}k72yys6B|cd{ZR$9=-!0eF;}Z}i#ItMT{%;N=GcA5^v!=!^JR73* zrT{t|lLqh1fH$4Mgy-gUyrqpn zSq7@TEi+}3N?qa4O(wpPsM=zH8tqzd#HhW@+Dx2QnW{4Sp}b6_Wd2ZEG&~r4Noi!# zyQkR{6g@Y2euGHl-p(cxRk;MG{D0Vc>#w-FtXnt)cL)S`2qZxY3GPX73nT#&JXi>> z1r!pT1b4UK65I=SC%C&5UU=c$!_!Z{-Qyege*OLdJ;wQ=2BS`KWUsx~nrqIvr)`|^ zJ+_l#>NqUn*ZfhmL;EZBxaZ{bGH;)3wi~4pwh4+qXWW$S#1M}}N?oa0@(i=lyV#r~ zT)9nM!eC2D16*x_F+jUN5~6?FAhQ;~Iiq;B2y-8Pey`oMJWMtAl(6Yt#23CBWMkI1 z)8aE3nL$&EVp6p$CRf1;*8-M2`2;IgD61TM-zDEL%wKe2aHi0%_hW3%yjY#W7|{{> zSTJ$SPHIv<-&y5)MoPG1OH@Krba~85O5+l7mc+RC_3-))a&ZpEVB&)Zpz7&qx_HAw=GISmZvWzT}PM61>uVx zEf~qkNi$}2J~QT*c>cJwgc~$Q^e*(4l?KmGbPbu~w6AnGmx}L)>ge%iWNbw8^lmJ0 zDR_-{Yn$^Y#W`4#t3OGQ#B4%?mh4@8#4Chfw&8c%a;8y|OKA07N>i?yj*d88boM(D zv8Xt@NX$%^8?y@6>7AM-t9;67G~Ug}OF7TLlk+Kywtxl`j0)yn9T+zUIt_#1n;%f( zz?C0I<5L!faPJz>;OLf*#DzVKmx4`K&PlA^=PDtj%T!`TZNi6XH)gsSbeFv8$dYVn zBK!&0WjcL5B5#Aq1ex1(&hM9(-2*oBHUqix^k~TlGioDlk^82A{tJ^f4+!e8yM630 zf=9v4))uzM5=rjE$XzLSb3+&bZ|3^;rAYW9Atg^$H`q#zbaAeUJfqPbVmY*{YB1B5 z2e-Kk){p&-=yhG6%IyUQl&Nq_PfFToUOPIFM_3}=^oZolqB~zJTYG}iE}WIGT$e7H=2r`Ux_u~0{g0l_;#H!K_Bip6aFKNrnIscLHzf4G@ZqS~V6A}yp*48%I@79KgO6FKSiBlv#mg*v;4}vgn-6#oVu;5V ze)u>3qgT7%BWezE(GQo=QJ5Eb-}RI3Ipw^21r)V=Br``Ms3UX)U;GRMiTlXugd#_M z9_K)y>T7l|6sXB6}Dyi1wO#h*B))oC3*QGpYa( zT{bYLMd}sxzIAguG9nh^Y-nS7$brX z?F{$l`FpH|8q2wZ8(eas`Hex9JoGRE?n7bZC5J@Kv#_y_3RL!QlDb!7*U1q7DcDe0 z4^8(9yZNWvc?0IVu@R@YB1JIirV}~?irJ^p;%pdgzU0^z4aN#sJx1JY+AYGzM3r*Q zi@Ta|-SZ%mfO`JC#_l%yVvPmYQ#+4C#d+NfPmDh3%-MF=(?FT?!A&wl(i*|mw$A;= z@W(=#{jMZI$@FiK4?sWO?MN#r!u%LNT}=G1tqYvZ|Jes^g&Ic!n0IkV15qT{)%|k z{kHqy)8FCZ=OQC{CaSAB^ROT*o!4!Xb$1m6ZGQ#(Xv!Jr6~QY$%u(NoN!||pgb%!( z<-IGG&|^`0X~-GMG9W?u=HUv($lDyT$V>#kZf=7&t$EOXev#|4c*t z0OvDDee`XHb|A&6j=#YAO)yqNLc!g?NQeQ*e-%#(cHCYpOkZx>#C-!S)&pD`r55%qp2=+gF80E?gbwJ=c6xp;S;=lt6pKdO)*44eCIZQQ7Fxqu35?JXPrt_=S z6lvGy+z#_2u8lL+5sLypxzxvMFitq)kUtWG7U8QAfLLl8Hnh81-MwmtI-WRZ?(y^k zq>HGLPX<)bXF>+_EV6_FU__3hY;^>|t8&T2WRE|;Ko>FIgXvP_RHoBQ&SxrEeOceT zD8;XkJ6O7{wj{h z9^L@e6zXgUMF1-g8Ix?Rd#QHtbJFjq?7Axhz0{87)-)nacdrRL=&!pqgU$s+FIktG z^_Cttt1VLpf#N(bi`TQNsK9+9%b)FnpD`WzuT>Llu*Q-LIOi3nsV%eb2G)q4b#&IH zJI=V7dKrW98Un9oUcN78ANtY0WS0R;|8y%BW7c`(btH0}~hCe3vS6;tz}=KJUB| zQ>a?#-T7;@H(RfK=++<5s3w(sKraOwl7){_mzhYHdd8;9U2mckucM0XDE(Qc{f=BYiAGml@DGvm5X{N=^K8 z5t{@YR$ImxqPtbd*pQ%O<*+tbBOtuhgUKy^TiGrsISe|*Y>)3%2}xbE z->*W@>TbkL5m#P0rNeCQ<`vynHx0vesSuH82NR83A(xV0* zOd?}ne#~DvvpRc9`ofN9L1TwT=z6c>*Zi%6vx3Tmx!Ae`Px0-seV>6tn0(YjC@mG~ z?4JhpJ6z1V;rvMfFPU3Y0*&TV&T0de`ep~0LNMl)`I3jd$AGgoooyLF!pbBv!(`W~ z4;JGR$^!3QBn%v#E5o@VjR0-y6+qvDn4@HW_}%s9yk2RSWW3V$%{j^0T$PqeZqlDZ zmRC4FfTp;6XU3OfdP*7YQ`Wo?-&8PkjK;e8%FTVl1UvI$lwpai;#2u@z=RdeeK7^{Bh3Gd{1Y3n5 z_0KwUgoJy|T#sq$4ru+( z^(%wLD&_%phE5gXGxR2B0-j%HB`VSXNYY zVsF`dalOA|Eve#=+(_~v&)@=jP#uf%)Qj6>K8&OOAxR?*m)6r4&9_(B*Sog?Tu){# zw=8AE>y7(KxLMq4VX=3{RG+fjlg4<0kn*gXVSD@0Ki2X9Ue@%BQ&1Rf!y1+~ipi7I zgSwL7Du0e0mb9;?CC0+UIl*+{d&ovhf|h+Kg?fw9dnaKh6t5;D_DQaupMKP>Sj`N} zf1Ukp5&$-Qx}aafLI@%@6NTt;o5-!|vN0B1y7$%5zlBl>W>qxuLEH1~r)UVzFN%G| z)K7i>1bDPzvb?Qb1D5Vc@IR(C#FvnTif?Q#dDo&In7xx=Jla#%>+E!Eb&uvB9wpfj zD#{O}V-p8xTv*rcUC9J0-22Mvo~F#B#g}Kz-Qyo{QL4Kn>!xgB`RMW* z_9Xsnl0ly9FPewK;U{XP!#hX7i7NGdmSm~=1YLRr;7{mZ+@s`gw~s07HIJZQ+Y#b# z-Ax(vMtjcr@>F8(&OZMeZEVK>`=$PSBA@TtqG39N0;50g*HsfVZ@0jTdsBKP_NvxOBSXlThR>&b23z+O5{ z*)}yaPYW)CaSG-)zUv5hNEfGe$)9SXSt5()^5ZXF8#y`K&+8KoA41;N150e#BJ#CL zPO<%Tx?a(hDT$j!BVnbc@DR}~{<%+z59f`?ZR)N5!C|cH@#=&WmCN8Ty`xLt0@aP1 zN5e(n>joaPFZ+l7;(J5u4jTtTE^M5k+A1AAI<;k?I?~DILk?ThGtb(M!%j#x!K+y> zINTb7KJ3ueAdZ~q>G5x<7T-S+z_TVdzBfpXyJb3_x8s-pqs%|#!*A`O=rfBa-dx&l zML2}W+^4}mCix2LC%FozBPvYIJw5;k50K6m#A-3!4<>2ARtZ~W!6ND(&j;VCQy z+(*559H_IclB!MM6u{rMMu{s-F_Ly6O0?G~oFS-__px?3FX|W@9co(dP|@vD;dz4= z(wHC#bO^c$I|#SCDWL75tX+`-^-JycK>bH37K=Fsveir3LL^$0RM=4^6$6WV6qR|` zmDn0L)rZ`hJ!7cPX~pqFvtY2~Uda)s%dI__=w=jlc)_mKRBj1fzaTp?9+ox4A|(rFBX zbSuM;H${aRd>7~f zDjS*&;d_y#=j#W`bvw}aFsoN4djgauqrRlu07iP{A7E@*fqlqcK}oQgQbYQu4wh_K zvLRmd+yk-j%miIwGgNW03RxV`vwiKV?2Q%%Zyua}06XV0!Xx$md~;b(3Avok?oM-? zT^w?D;no3cu$jQQeJOA%=E;%-&fHyVwpFLAFERGLkG@7BV^QwQG0g-9P9 z-T}<&4|Sh6mDknT6ID-4LXe~WNpNtCgZAhX`XiOJ#Tv^nn@|)-J2`iMCZ{rwll`+} zl29Y=1Rr9HGJ%~ZS%ZGgo{0kfLiIa+11aGcaYNLL(%q~vFZ5=N(S_}-YFAwQAm=mZ z3#ova|7pml+t6wh<_YRWb9MLE-Bk9$3M#tss1yz(chqT%3pov1ER&FM+ZCaR`UBkE zMwx5hF^)gFuk!Q)W<_Tw_fS}s%qQKXs8cOe@WrF+#*)DOb}q1wTlkX&lGMT?goKtw zDQQ$;ac|a(M~}xVAaQ{LK31DUS`s0NuWE(>wP&V@lF#lcKqSFn;b=MCor~&=^=Q!z zSQw1h!?P~x0DZVn?oZ~N67R)kiC0YHd+`VHy5dc<)vffTPSblU zt!2*9yLig=E`h-P#9E_IOzyeiP6C#IipLDdY zy&1IyOfQ3dTd~4()w-bx$0{)zStX_ZW;(%tF?$v z^=>(9aoHjkNkeLmA{8we3bdpQo054F4{p@89LTeXXD#T|H8lS+=H>L5tg_gp5f7d~ z>%kE3YkSCN+*JjWfEtn3H-E#26dIva08G>o;9k(= zOZA}&LGRI{osq(@s!bizKYFQo5yEt(_G5~iQc_4Mfj6P+#bk9_Xv2j4`Y_yXdT<22 zA{ko4EdK~Ka-iO1ln>y4ZFZf2AIHAL;6VoTmzi(``YB?nRI^Mo*rsX+4mHo$hYaQ z%5`(E>mov_^yt{dqe83OLAukofRd%y!i)BRh^yS9mn(=rG~G3I2Z}!iqR1$ONCWqL z_I>U{10p?&$YZvw{SujRhF_s^?D#Dj{3O>7)KAwJlqP2!z<>2wg!Jyq_Tjk#e=v0y zW1#RNM-~hn$L=y*xWSlO&Iu>ft!s8Aq>P)V1vIN8-o7S#5rO7T8IZPH>JCQ+AesRrNO>5}PyJUQhzxrl zF((1lF;&<)@hjy+g|zlHjUkOgvV{H@%8yPHeYoz9a(-NE{+&YhDEz&23M$76H`sC$ zPGxA?``FzOjOVU?8vhqPULN38rtigO?8VXwCM>fYN_$+Nc`o8o+=gC9Af0h=+B&K9GJE^-n>Sd5KA{Y!n_7$gVYpO2JErS|J!;Fi%{OH} zy)2n310d&)h+XFZJ2h&gHqK@Iv39^D-q;Rtu+XH~G*{XhBPS^NXc$mO)G2_M4h_Va zKM7zS*fFR{V|fWr{dRtA-3;`JJr3V`2d9;U6l=yQr+M7Y%-;DYG$>{l%P^7H_;hj& zAj!C|QEX^YRG5!laD)5aWIU+Z`S}$j%F}8sKpTzotJet}B&$FDW zA2Ukxaf%MqX-16TR{x#eno`my`az?ev;iKF&2f$w1VcI;?QRp~%-DtvprrkuF6Vo` zCnx({q5T~m)sR{%9|?{<^bNcXf;Gkqm=saD#WP00t9Bt^n-qbD(QsrD-?71s&f zRGi;8k~{2>_1a_xpVxdO(MRf*0Y8uE0=t4hRhq6ltz21NJRhUM7#esSc1>4dyB2m% z{smL9(O`p2tOmm8RKb?OFLR3C{ znWW!$J1+88H)@}E(ONy!aBSoN%{Fpv9x&qiFqcwvH>8-QvsR6~revl@?qe%*tb-at zmY)E0T4i$IO(sILw01Ump3qhAY2)fzxzWE$q&+ zvEfT9Mw}Hv+Z?!L{fzmP_~birb6WAufRtCS2@PO7V@+BhNW%JBmT@fH)xccz!;}UA zJ*uTz9;M1MJ4^R~syWB7TLcgzVe6!?5<0ezN~qTRMZUa#smkNtrHHV)@lDb1h$3|s z*YyGy@`cqO5|BBgO2)T@M>Ph%Q7&L?HXyU1A&5&u67C2Q?_dcjVzC>=^{BP#xOx{! zBx^T{vr@O$7v?rDIE0Pj@~3Q**A6KFtcJzHWphC%sX84dNAjFO>=Rr-e0rtdY-Oqh zr-yq}1K6hMeFr@l*IvlZ7UQO;*W?28ghnlYH#tlbkolRCB&X#HJO04%78;0;FMw!@`&&x}5ER9|}t*mV3z6032xK z;9JH{>91g1pPU=04b$B?&#`p)35C@6*h)Mn>;hI0P0hjW*J;I560#i!7oEMxeI(8mg)8Y^!JN7?D3)C|Bl0 zMu)gwyDmBrdgLg&&s-jzciZ4|b5m(vS7Gm`Z7r{&8UCYpa7?1G%uYJx(rt|q#FdB> zCL-@cZCct+n>qsd{AkC8v}_se?g^pu$tt(656zZQFdHne8cYoUgf!{H zTwAF-%jjAIT8LnU{PLiX#nBDfm@94{5=`klLNmE%-bT$UEn*Gp6S{ZTBG6Y5ERm#& z7)~#PsE0nocWjRP=8k|gk8&tBNB1ET_f*P(IxN&l!!BaVp!uQ6kJkk@Vc1Vxs^N^! z;zl#fG3dgS+_tB>z)cjEoP+Ux6167R2AqZwuJb}e){6XY;+Sz&;EP{uVJ8kZhLAfymiq6w(Onp#v;Mvl{6*IjshqwHeqv5la-&%F|K@PXkB_9Sa>j0x@2dy%# z?b9)p%rNo0XkC{x-r2)j{7GwokjUe*Wp-LbV+PLv?E{{$0t-HW#Baa%dU`#+?8W%r zi-G)gIX#1;aaf}`{hk_}+{hmUSiH;~R8ECZPS(^vO_8x50ldv0%wf$Pck*-JoqhZG z959F_I*mvLG)H`fy)WwPdMti_Lo1ydKC$v2-t)U=0NS{fdl^Gv^M;L~`$r7$J##V_3FY8NJj_%n(^ zqrjdh^|1Zpfw5{1+r$gpza|Rq-xjv}8uFbQGvfT+0uVC+G=Q6D;*HvI)K>3L4XyvA&dOr_7$m(iK_$D7&0 zR?WCWlUcJ)HX@SEDNv_e)l9w0q34u;KIvuk3>rDjN6o}2RG2jHc=wgl_p7ssPXBYB zdi7ek{*|Xm+Pj?7+f$1r-GfgyWdoH*F$I&srBVFLN9o%qpTe?htog8`@&Hb%4#CzI z`{zmw9sR3cw0u@sWounyeKBcT;9zP}4(EjscAa!GQiH454-U?ZW{hO%SSD`0YXn>N zAV8PxFcE{=v$vU3XAeu_+N6r>{|gtN9AfD5BfNO;BR!~n?JcA}SxG&m0|}ufVs~Cs z;*OdlFdp*u9}m-O*x%DD81x6m@YnP@+e*>>z;b9c z{N3h})>=`c%glW~@K+P7D&5*Z#IW3u)7KA^M>#OY!-Vav$p}ct-kl!7js-d8*Gb*( z(PaQ+>o-yI*HNJl{Va*UGZKu^m5|fc{0x$*AnJd50g&mn4w-YNatCehArfj-Ixh~c zy!~{-Vb_Q%d!nJ&pnsovF&+$7@Z1CwN}Weu_az}LmI&gB*_|MmEYW-O`94qG%1kld z0b+hCUaszgn_U*-SlVU+Bhi@*B1qVkU#Y`%IFiG-s*F?}`>hRm0=1{=`RHTs%@M|c z_sWd2acFJ957xF9fe{C8qRz|q>NY)yBa}rGzk-BPQ2XKn{M0Ned6_Uae_iJooAppS zQ3rZf+4fbt{Q`57TW9qiK9Q@}@e9p5kU4~c07Z;b*NcCzf!41y3N5PTBdZ2O1!U8mmc1vla zHUmFhx7IrMLNGSXKfkhmZ4y)EctYeH@Ecu|%%_~V|AO7w>z*?z?l8_mM`oJSq@|@* z`2zX=Z^Va{VGEEtB}!5JPF?^wZ+`#^>gK^EgDe3Wf{xxuWaUve^=VM=G55yHy6VMY zT3_vo9F>qAP@A$<_Df)^mRpvWnt6Hs6euS>014;V`+IwGjzn3XYY`^C;L($4(o;!B zfBY(JZ|@t~>kqqWY%CMObV;aD+_+76iM3I-4b2yp=La4Di&OCUM^w(AH5IYJnu96U zcM942HW6sr63=JG*{;0wd1AN!b{TewFlHt90vEu=6r0IZmO}zsy%{rSyV#vp(2LaS zem&K0IvD~UP5T^wJACyS{QV$$Ch4h-$c0oAH$aR3^H?5_GRIo(<8nwPU3??ZRIK^x zH-@+iB@<)%!sS1;0H=AJIGQua3st{O{!LBt%SQANt6_ychD)HWo5&gDY)C(0Jx#3Z zj*AkbiPCSGTjNm-?EA1_dXb@gLC)q>d$AgnvMf&M1=MVit@C?RxZ<*7-lD*rX5W)O zwNL%*Fa3)0zm|Rc^|q35My?zVaiX6=urn!C=K5jvPBLV_Hnxt+X2ZN4hHIN1xWI-Q z=G@^MSNB7(wV{E8ia|tj0(bx}GMw+K_iL%>Ku1{a#6+o~K!Mn_S$u1W-~36Cux+Vb zraOuM_1ur@X-jvra|iGVQyq%6jmK4jDUp)04gVwf1!EfR%Q)S{K|x;E0Rb5G$Pb78 zj?SmiKX&dlR^ z-M><^#K5Zl8W@OsQ)cVR;gt0|PZOqLUhx&8DYAXx_!Xid$m`N8CBbQc<)xjYl(zlW zrSU|7WkT7m@t$epGHTqG5>JH_gC&I)VFO2k>_L?HI)np0tLQK*rJ8_)mQym#<3e|m z78|ebqc{g_uos;u!)JCESkvFpgg?Uk`;d@-@{Wg`+?$|O=h=aM9haji1%2)2w5bEs?1_t$W^CQ#U$$Rv;x6ENP zzmznE|L9sz4h6W^Is|A*gRgCaFVeuwPR8}tOCGr#;&Jtw-IxdAidv<9K~CVn>V2mf z8$VL3(y1q1y!F>S^LC0twL5%PTLV>X)HPQ&nNxSKY;9~eo<>)dIwgD?O<>oi_~I@b zZMZsr(`KMQtG?Dct2=>SJ1phm^qrFJeGMc;l}ER9F<-BA>R7Gf2szS#{86-J39I#! zbdFjzK8|KZ%2He>6?*Y5A?>52oTjjaDVg()7243wV%ylXX7@t!Mr>H5*CR zDdnXcRlo}pJEV}w!=B;-`o2rNbnV;dW5r(&9HaQhCv`nqH!Wf9^)CLR@om3LG)mLz z4q&;qAS@22mUu2>!+B|<<5h*Ne*vP_&p_m!RY!V?led^Li?WSsvU?! zl7UUPcbdX%;VtZRJ4aYfVwN8yT0k#ZF`o* zMb+1Nx@+^}#?{NY9jEsUpAfSgq}Lf9T8CC9JMF?rdM%QfrxZ%_6gI#fPr7U zxYy%xWJW-W@>yOC>%8}eQ1;#rhx1RnlkY4L4Nt9K{S~<<*Slu`W?}dWuHR$&IuGId z?kve}?;g$&wsbO-)y-G8=oxu@BezqF+c~D~ndjg!-`@V{z5$w~BUnX{Ve~F3hOm7ZS zC?JYcbYA=t_7%3QeTfRzG@+b2T8Z`d=)RCEyxQF<#9}J%!mmljW~c7zTx`5%dcUct zM(24~T33Mn+phO2E=!Iv@>AV*@VeOHms-zzP~XHoeVy9EdqKPPL-{AKIFhd)S^5`3 zYH_4{Vp!v0lcyA{O5aytmF@i5CVPNHU2xO6MVb2^M2>V@uz~faV}K57QLHC z%Kh2ou;aW|8P>{8b;mD`s-{Yh<__nEy-|fTNNC(t#@8LSRiC`Dav}4k1{d!p)~guW zZQ)uyL64)_yOf=U8wTSM9f^>jAddPTP`;_iY>hdLLzOmtqV^@h*b;HHyW&}pE6Oe~ zg+j*qXGw>j08)DO)f!Oby$lng;t!DpmdvV0yPB#q`y26kQPWFa^vT2v@wOww?q{BN zZmJ=+;F;drdGHj>FCJoi}z(_iwNN8q;y6wc8D~QbYii$QFqRrJ)ekle$n5uLA}5>pX>zz&=-5FzX0) zf>umex_AUCKKF6XBQS+Q2)XYlvi~^f>M-6NDsz-`u}!HQh8jF&>yvtP-|CWXVrgnQ zt|I2D7nE0QB*KLBgwE?E^wH$CFSag95kD@t83zcDvS;aRR>r-B{SFzHsbI^bs>eyIbcH2CK+afXV1LL`4IdxVIK z9XeA-O)22&)!!FGyL)kIZ`x~F6N@d|j{MJ)ke&$WVZJ*Q(}Q$a^h^&hdf$IEmisPKWG`646n+lTEZMT(L*4$Q=GmVdqVWf6vl6L)Z$b53{0R2Fsan#3AcDkiiU zyYWBZtXj5S4++2QDa}BBhKSMVN;e{re0Zw8GA$$xT<`C<)dx3vT`QTD#8%l7P*ySB z5DSRe!nP@`rUoB zo+lkILHcqEh3gMXMXtwy2V5KwsIpKi`6NEepA-E*AHGQU0wt72m>JM`P8M&K8}&v; z|B8;L%yNRdMAs4V6nuQz{-&?!t&1?TyG-l$M${3)1i(-cg&Vff{L zpA1LHvP@Gks-!2-a zHmMHpEgeDx{VtAG^WzRs>_^GKOcrwgd?y&eNl$Os6&%U2;8$aHHIgYqO-2gHrL`SK zq=sz2O$+|MG8oY5%RMo<4u6^kiv;kdzX^hZU%a` z)9*wWYTShRWP#IQjEKvzY*ZKSM!{p$p&+L+!vFXTdS!t0jm*MeX2E$}g^LubG(pg4 z>JX?vq5zL#QNenv-yF_i`MEco-e+n118KAV%9riM%$l)-UZUw_jD6!|+DSrFdZR!Z3cXjCPrx^%r9)(j2-1 zdh&5Ex2e*CQjpQ{WE1S3CKtdPGQ9^<1t$bI@&41;YVnhMxboYp>E(Scd&2__>iHzl zmTC{&Vf;ubNt~UwSd;vfr?$2vyZTdB^$!K=`AjZ$xrz{b00iX&s$Bn+6-Mda4mr#(Or5>sf6Os0i#|b%EfyO(e{4-= zXI6e)92XQL6{dB+`SI%8C`0;=<&t&8DQS8@!1thar2laBH)#)L#+OIIpO1qE?v}#{ z^1s;E|6>GC{~TPwSajO8-yEQ~HysYAw5nYURpz}0vVR#3%Z00y`FjcW+_Fq*oR7t2 zMDt(L-dSy1nlEVeFE9YMXf~tvu=ywPJm99&ef}1+4jIW2s{hzYo|NFs(Q&kwvwBwI8X#F;e_}xGHFsF5XQ= z_3&KEelg{H_Ele)SVZJAt61)A1=LFT#SxL6VHpbPjB$T#tRH9?SGL{9US_%Z{w`{G z@^_qd)51$&Kf_3GjWn&Yx$ujVVet(>mld@qSg!DKqIkTzy0H$=>OY2S8_UDp6nfYr z`bK%teX;_@CvHbrvrX!XB*z-74?yI{eI=4(%SOMCXR`Hf-RQ^sl3LiEhZDJ^ikq&! zv0T%7-emX{Mr7(btd_p7{Zn2`Wr`0dBEmndf`79~q;y4Eu@10;PS|TT3DLt4fR6Q$M0KRt@Dr@}nj9dmoDrZng_k1=S=V zb;V;CzF7Xsc>eou3~gwam$0;rT03}%t3vE^4gkmu|2lPecNbu3ck#y{H^jIvT5XN* zw|y|R<{r)MbzUe`puJOVb#oO7W?Q;_0$M-K9}^POsxAt>h6sbMP-Ew6(LLUs70;Eb zF5A_Ai|L!CAY*H~K0wdiE5z2#?rv>;6JRulKesV`WQerO)eS%#89u8zUjbf`pMi*b z@obGLT+DGPB^JQ46h@M)B9pAxyvGPpo5ebL-7hChW}4gsT7C&wFE*&XK2!3#xom$) z>(YFE0GYY(2iYBUk`~^$tWRoLjaZ(&e6)58;q_Mx3}SsAr{2MEgb6 zH?h2^i!Qfj}@Z4UU4zu4l59+K<1s6qQAV!l*V?#TaufZVA7i{$zCMa?92 zBu#~P&{$VFrIr0yEUIPl0`y^U?C zAK>3UX-88?2_TwGlWr5or->dlwgw-Iyj$}yL+tS!>;JAFN^BdFs~W4aM>m_SEK8(| zaGr1Z=x`mXvmU4S1cLHN@3dDGF=6x5d_K?X^k8N|@O(}(pADb8qUZyhP^Np`e5bxCZVaG&qa_U{^;RTBiP zf=S|(wDW#zq>0UKnr%;Jf?L+n-VZT4(#A?6Un{tViAg2WMmx_{yLUgzSX%F_o_vG9 z!-YdQj|cZW%y?ye8BD~hIpI|uLMAx-qd$Qsz(Tu5nihI(u#iiukRp>N`vG(uOe_4k zZ$V96?83&ciKFD&b;Y|*tykg%y(IGp+Ba-83lx`uQ+Qv+5RmJ=B%$NY-( zfVBSqAFlw&Obj`r+0OLt8v%GzvjTF=bIIgW?(L}6dqtdI)G+*pdRMuwl=4Vr<|Ks|@ zL)ps|oNBh9CRA;``xe?c^03?ku;}~PG)VHR18UF;Rcz5jf%rP>lV8EEDk3jWHk((8 zNsE_ncSffwpIdN;QV0%|VM$XJX_$*>&LvOI30CC1=)XNyPKMy%xD}Ut8#cUZ(8BTc zL+tRT3AvO-QF-$-N9regmVa}U&^hp_t@SQVLP!DEQ`0*l;d1>(ks|%Jq(;6A$tuxh1@!kKUL;B37P74GYYe*5)Pqh%dw z9;U);b>KMr19Ebvfn2oy26K_1D2-B}qZeM|^W$YnW4&x~?W!G%Tz~EyQ@t20tI8gs z1MUeni=qyr0P-Vvwy0FRLL^g|}UGVw}@5HP!5xNlfL!12oqmDxf zJUIiyxIy>-R9tvD12u)abeo|MxMy;q+kNO~OFdNCkBXihx`5g;3OU*l_bGKJee2T{nK?0iHRTf8vC0U_eS6*&8hj(eyDIcX|z3@eF zD3jeOPdon?@$+!Kx&z&~o+}4{N2iv*s-tXID=u?AkMx?BZSVGNk1Prt;9d4tZu6wP ze8<;;80c#wf1RL(ac?&!)a^RP9xCLQmol2I&_}*8dlv+1F3&{W@T~uq?PMN8q*+t% zv>-!N)=~$&EF^ay0iw324T$|Y&<4E7s9X-14CU^d8j}q{^u>;aB%}F;!YqTem{A?N z@P+CpHJuVH?K9LEg{5tL4A3+`{`qW`6uvjgF9{(VePJnI!kb@TItOWKq8OGxdfa81 z%dud*T|R|8y7<1T+zEn9I8QTHdg|e_NTtHRpp0Y+_C9+tiySe@F}Ds5bJy&g`baX_ zbw#~hthUzlr4x#3)v=^==h1=^;E%hb08mVoY*$GdQix2!1EQr)XEH)vCA599?Ci}i zOORTb4d@1GJ*i|Jbd)H?pNT>o91HN6ogpq*Exa)tAJyWZduRNQ`vR(*3S$QyW^B%M z0g=73=<hUe#?Fe6EE|T3NK0xT`#E%zTWKiXDMI5HraE~zH zO}F(<_B5WvTb(T(29S2#sqG)cX~jCGGP}3nMeWVnTNSLNx2Qbc|6%%&l>qYkeI9~? zt=f;FR6%;V)2zLti?()3iwX_Oe!ZfN{m?Wg`nlG{}p zr1D3ror+Egf9Lyx{?Ju`h$GxhE##=g^ld1G=gsc`oex^mh6^jX6#NQIZorD#E`L$a z(3x_z3Z!U}QlMSl041YczWv8Vg1Jfw&jEYD-ztzFnehaikj)R>#ypcU9(zB~as(Qy zP}ih_^kO&`p7<&=eV1|w*_n7HQ^dLmRq=fGy%^&;qjoE0!k&k(EMh3BM9{dd8%tS& z&L6cHGbHgkN1^;Z|;giq8;BTN(yH3|ZGE)pq+8LF@Nz8HDzTirKP$Ho&}c_4 z?YMSUi8;KK$0DS)b*HqPnkIuDz4@C!k{4kM3biY$#itUWjEG^*c0~1}p>w-1006<7`u~kejv<~(|J<+vewo&{@ zB)$jpf>5ClH2~W(5?pA7ijM5lDi39?0_Ez*sFtxaNVmh;*CM%_MdDy$}r=R7Oqoi{kGX}DM6MAxa$F6^mdPs^l#E-khlUN>%$6p%oz!M_fs*NUJ+%k>Hf&Gy&ao2F={BlWVwV z_;taQLqW}yr4z%%P>0-m*GVi?X%ujispj9s6yNJZ&k*D~9{4UC;~NV-gln z#~01`q-B!C`Qqw`p?)k;U`SksQ3?Kf`)Tio(LlUpUEwf+r_DKNa^d$9U28UUe{x=d zs`IL;&Ta$eIr3!B`PDKXV-dBO;U*P}dy?sAxd123FC5C>iRMG+)axTK7Jbn0$D$|| zpJ`%K{#wX?pPO}vdqhNF6Z{B?pjETHJ(0rQXqDD;kFeEsl)3gAV>i0Sg3!9od^Fy& z2*nz~gi^C>IgU0T3AoMR5yK@ z`S7XV!w#m^rsrEm*U$Siwap)#o^57mRlevHp$J>N=<&85`;=BXeQ34+C2y?3j`B@z zovFBQ%55=?0Qlb$g~aSYC45b}^U3`V>=22#q+K8P8mS{?Vi{K^j{C#P+O6h~y51-E zrm{j}UAOXP-F|c`s^!{Y2&Sj?tzkBCO5vXji{WQ;fu4YZ}OOt{-bXA%jex%f%E@1O&ZndAO7`T^kY^Nuw+n|vl{C&EVIyBwx2WdJ;-9!6oJ%*Uci32 zT=`<8tQvaS|EM5H*GAFl^_~`7;3Qr>MCEOzv*det)DtYW87*QoZL6VH@dV^WA6kQbX${x4$*Mr^MOlz_sEJC7QCX=eIL{ z2!Q3aFmLaOz5Cikyg9Q#xd^>f)8Eg6UD}m|?R>f`eCUwthQ5vR~hW2moMjy|qSMY?~3y$KAG5Xs)=%3iCgz`7z z)V4FYyVlY{M8?tQ-Cf*e?!@oIkB*6JZZ;^Rty!8DZ@0Jt#wpJdZfc}yy`=V?jpq3( z8hlJPSNUMUpqBS>YAxPWA|PU%L)9%L&Wo~@+|5zuB5viTtfPXzwQ+cw8;HS~K|l8l z)=EsR)G7+S$!sC#20weuD!xG|-GDGVpW7MxW~s-6AVyFnm-G7M&``d;JvD9GAz-M? z0e;plgC5U>Q?5wQ;{qYzB4+yKXz`W_2&iy!SG6yc!78fes7|K~27Y=Ij$ z?BqqdpYDYuI4$qw90paWHq47VjAoEIAz_5*Z1IKrKylKY@Bl|;r{;4-weaZ>g01EVP3}V6AJeFn)|76H9P=<0T$fde^XgvqG+V+7z#J@)8n7)P|)A zK885N{>Y45l3Rp^1a)~}GG$#h7j!Z_W&_1Ja;#Qw_p|)%++F*Fi#y$8mJtJo0$8@p z-Ok$So$zKYOf*5|sJqnfcZX3I3Nvm>6D!M5H`*1AZRuWg>Ta)2qN3@*?k#f; zs=G_RdG~x?G^-@G+3S#Cc$KcX7yx$Pa7ST&94>!~$+bDu9g*Z``C03#Y*2Y%X_Gxw z9`kr@#5b?R%@lOAP1U@gZ)i4*$OaY)l z1S%C4sMQP>ZZZ3ATl-u$4YS}Gqae}p*hp+3A6E-n9N--fD+xRG+7$e*FWI5!vi^eB zYA&8c!w^zXeyeHo{;&BUCqlFt^is!x&#!(UzrKRI=e&cky!Ayr`VX>4Z_2!r8=kim z^>hK%P|t^%eeY{gpjCT#(lj)=7T}?wof}x+G9|@d`S!bSmpYf5SKUu5skS&?A?vr@tuk$bnieW_J-I)k5 z+G{ORfS!TZv8Z?F;NBmzBIZ9|ZNm4yeav;pm^#Nl56`m?S0%H>mt6DG`&7pDmYL(c1a>Em4?Rw@jD)GdPVcW-EoScMXzUNe8&CZ zE=F8T;xih7RMHL|jo|(4s?&rsJi8lL0OlS2`_qbj2e!t=#H8QFg#0e!j?p|SQk{lf z@F`s6E|75H(+3ICm@6W3j5hs`zl!_ff<5Z($GQ#oC?@st5ub&c@2~dqgt>nvMWl)Q zl}=e}R8WN^lr)?Gut5>QEVQx6)c1U#`KoSari zliSu8?;pt*@9iQE!VZlr=W}D;?a(!U#_R}*t0eLp+@c--_D0X3ao3#(I{QU$B6Yl{ z_5-AEK_?vy0Xgu9bZgksH7_LHprkAZsF?g>5`2$6qjE%U$6AQ{i_=Uw%Q_#JMP7I! z3>H)i`$&QSZcmFN_0;h=n5B_U@2{42*ztr_hTe~%8eyQ++mkcOs18qzdUNgH)MI~% z@+@7&Un6wPioODvPTs}0#iO%9zg0n`gCKEky7pUI_jXG!PaJI|Pd0;yvsy%RKB*+W z-0~SwqGO2;vu%PYZ8?zHme~11fW;3!%6P`*j)##08A3&C^B2#%@qD3l851e@SY3fM$-Wuq>5njAvz0AC z209fk5a6yO)R#^)GjLA9_;^?fO-wp=a# z!KVI?P8Y9f)ad}rYF|_$77Zk{^#TLP$AyrcM4@|@z3n(k<(rR@(b{_T$PqE`94B+2?<@VH z?w=54$ttQye{wk@EP<_=kC4oiC6?8<>Ltm>42Ln3-mAgHC^I|RS`x(4{$|!bLU=5f z`g^X72n<{LI&S0OS6}f_?aJs%@ZoVPoT_>>{Ylu|1LA^9j zfoyjRerewHpwA1))mLf<9K6#$G+N=cm=~JMPjktg_x>ieH1pIYMRSwE&OE zMtUDbkSk2L?Y^(KLEaR$b{Zu{0-!{1C8eC40@Z9`Ah!IH)VqdHgVO)_4-PB^z)gyN z#$8>tY+sajzv5M3eNvOlOP6agyR$;B&DX(E3n}9~YQd!Iwf#j{2gJ}Ay5UK_ zZNp(SsR&TgZ_YUOsqHDv0muv4R7~s2!O;46Z5GKCOx|fpvw#8fE=o17plDxuP%>s; z9=KqZ*9}=KMPc{Za<^m1pTtDkGd59Rg~u&dtYhAEhh_TXaTVm`_cmCqee9b1Bmv!{ z0el-dlNK#`$|AUttohc*@YTM4UyQEeEf#p$K*ty?g^~*#lOFJ^hB#vRFtP-`ATUR!oC)omv%!46r`A zSO8YMU+r9-SmmpF!b3Qx{+Gx;g>g*iQ>&Qdm zU@{Xv?8((F_J=)Ls{CL!X{a%)R<-1(&v&F;V9pG^g|N^` zWMa9Pg&6Y@xv=M{cI8oV0A2UuaVy>T0XnM}NuNfU*UGb&Snfs09gn09#jD|V zGTt#Pl=KNG#)tESt1-Ts=or4$m=CIZqwn{6(>-26s> zUhUIVIVQ5>^42B1{ecCB#Vh#d?B!NeruLfrUD}El0fc3;kZDvQS6lxsC2miO_7E1L zC2s}QRPU#jTD?n)9k!t9}I`{#Ix2l;ZO7cnDzxCh|#R)ut=6KT~;oQ?^dNZnt za|Di$;h;&Baikp)jpdS|j9yv#`AzORw>LDx2mn&h8|)4RE;9QZjhvU7!hHksi3-cU zm@$}5#B&-?c;2(!Okq&fZ6V#qOWG2aGRaF(o!M42KdJuBwf29J?MMJBW$F7Do~%sG zCUq#e=3D7w&_%p9thkCZ zA2Jzr(26*S@=)8YJ1RLdCkHxQ<jo6RtreY!&ooI z!DMYCErKwtTfpe?qZ8}&GGIUY1U)E-o#sLPi z#A#E%ami3IfH3Qo>>cs59R>r=q|2l6QM^s3Yn6@9nylAT?N)AO97X`9Ldpr4SqPsv z2DljSo>L5{X(`kWc~aza)MTb`PC`@AZ!g-q+pI!qN}g51GaO%ARk78tuA^ASqyb=xlYkAs|(3W;O{q;9als&3JR6IbSb@MUmWaY^&7%bi1KL zGKK}L#314Z^U2$GHWT8mnec_x$xQC4u4~AE6y%tfRL6hdPkz? z>&*ri4F(1x!bL`p(W}gsJiPY!q$__Xl!^5RR8^K#?X_O3ujL0mTNZSwuy}p3PZ=MG zq;M8?w_^3;;~2tI`*HbVM$+xH0>Btwv!7ZGr;3(5Wq;X0D&m^|@{6>lL(7zl{+-h2 z%F)c9%@Q7_i>Wj6e%)?8T>*<9=zh6}WR-DKfnKibwJ1r#HL@a{R=P<7z&Da+yl4vm z-Bo(T`*d;l&KKuDQm!VzLrc7{oNrXEewRq`1yAbm+phE?TMxSr{Ef%x&lN(K$U{}k`C-e#tWYCj!L63Vh%X4b$>B=#PWfUQQ-4a&m# ze(PI>Hqu32%0f;a=n~anF&Cc_k`!2rn9C3EL)x}O<#7^UH8=Y|n=Dv+bvmT!rF`z? z3Rcf`7nc2-oXEHk*vFx>)me`6+#nuQWW#UrN^Fm=p4#`PDP8s?ai6b|1II87OIX|V77%=J%aIvtCWKea;{;?)~0yKJuMba}2e|wPJ z+5?nI^E7I|RrDdFMqUbuWs(e1l?mCpXS-VE78-``MlWG%_?w&oU7;~5;bZIR`LbH| zbvKzr`u_vH88CW5(#hn1^#W*b_@g5V_ua$nk00!Bw7pxc%c*joCYgmSAQ z9a`|Z=qT~NIla|T4Avd^!fSHN?w!Zqzs0|D9l!xI?EIg|82|ZiVjlqF3XOz?UZuFd z`qFayF-+#O+Vik)1Nd9y@n3&Z60~PIm zO-e_a<^MeH?*_)Ma|`fyJ!p!G`Y)TyLI$HA~K< zFjoc;2s;H(OF%Dza#homMBev11(?j9`AYf$E&e5|%sez0y##kO%E;v0PQ%^}r7Di) ztEU<3RKT@Wmad$Q_c8=&DYV5>{`wWQl7ZR3eIxVBBobb$jKiYk2U0kj-s!Lnw{!-` zQw{>xNeR%KxP4w5N}J~kPnYr1Ize9PWDUh(Ukt=9Ta3C@eo6;Ct0g$MmiXp|!Wz4U z$JrM0hPAo~@aFZiocQOxm77ggd^{Mcr1#>? zx8SZ{cdw4ceI3qEsWUCa^rfnXqF40$Pm2eKRpE=6A-B1>Lw0mGb z(^F>BrrN8S!V=rM9~bi?uW4`y?RUWo41Ro5YlRV?@pB;2uPU_;Zk(Ck@thHQAUn_T zMIwI^fzPRi`m0Eoewn?h&3N`Nuj!*RvZR4|bm>o|H*DilYbOAkF80SIE%~)1c05mw z@9c2Hd8HxlG5!?}TmIN~@)z$~-gluFsgG^aSJ&W@N$xzG?(S%5Z4dN9LGkZmZim~g zVuF&zakAKxp{;T2dj(ZL`wj=vh$lmNC%Th$ zXm&FM-zVfMJmHf+)6O3?_{#HkI8$Wq`~KB>%S3}Y>fd6_zeOVj5#R*R_~AclwXc*M z7>tSN!t&UXc4s9b8>RPI=4)Q*PKrzRy|+F%YN)bBdV>h%EIY&A(Til$f=v<^$Bin{*cldKVTU$$d&Sl$atz?M{FSIdvylvlP3bTm3^8L z-9lielpkL__d#*KU>+0C>37NHnb@Y2D(>CcAh-*prR^ObU7!00VkGhX>I_K{n~Cyv z@_r;ix72vSS#7@7hIB=g>WvdsIDM@iE{ z9)>-msKt=1$!%Z-PqJ4%lPe2!3J-7E9kaJB{TTr0A1hiti%(?T(w!+;W0!V2 z-xO$a-I{C}Yh0yZJ5y6ONfjtYDDi>CK??>NKpTPyNN%EbQ-OtoD8Qmy$1|OBZZ~#t z=5+xay6EAM;1FGcVivu0$tv{sPwIip%drZf%|CkV?Frt++Qlxx`zlXi^YU&$gR{0> zE!E;Fqi7wc7@j8M3W@Aqqu$6nvJ04b8>H{J(C3FTfYdkhexNgfon??dF+z1KO+G=| z5pJC?dm`$MT!l(`?7~o-tJ!nlXbCgp)Q^W*M3lu?(c+V*%kC`Pf?6HewShR7qryfU zF^c7lUIp@(W#8IFTfy2Gai*}UjamhyvxWx4i6V!`m@ZeO9>oI!<9p>?QFZV7ohnbd zmvv&K1$kE&Omwv}kxSQaELoxmJ+v*Z{1o#9ZaaTKb3d(B1z=s zBWqk-39G~S0*DkNG4Y6<7ei4-jlTXFr=fVL@eXR8eR)Ya8S0R!+Q8Q7l@HsAIFgu= z&d@U;n#2-79lKiO)$e@kjZxjKCEq-YPY?8zIBuGcf}EU&)3lCydwRQGBpr|@yq0Ye zYW&SXLwh2Y2R+=24t4floN^QPDZhz*niBh0Yj!CeI3;~G_Ido)H#prBK6&=PIo`gZ z{Xvd9k>6Cib*5P{>*J0qOhNqZLWt12-L;g=O^OSzSBT3c&NJ^JPwz|3druK7iV-Ra zO?>^6dh6mn5=Ijw%S*F~QSY!Z(zWYkB~|-)ppEX&QW;4f>`3*Gx4cun!J9e0MtjFh z-Fm|<_cF~~uHRj|Y&Ze|_EjNs^!WgVoP{2&PcIGvg7GQPk96KRcaS7ozZI;CMxDiRj@d?%q=6cJSxa&Qjfo_jln;+ z9I9Vbt_|^o`C2~kSPfWyyf2Q%LT=S>OcGXZSclB-TO1!|B!DXyQm!5k4KIF>(=V8z zVBmSC6g~YqC-ivyX^XFkrnuexIO~yd4Bsu;?{glG2~F=t5tn0-CE^Uni6uG15}f(^ zgOq})6S3)-xQ#onck`KYyB-a@c^_`BPG${!nMUY7J-^EmZ2hD%Tf7i^BKG9msmtS- z|IX2ceumHP;phNk5AC{zz&nD-f+iG9U2`mec$by~q$Yt08m??Q@A|NAC!GXm zs8E)3@%%_{!C?s3vRCk-Vj15(fi$_4LKX@7ty+#4J+NO7&o--%x=W#Gtuh4)d9M;H zm+X&SXTaXv19HEHsi!Y9bd+2#s`+k2^ zR9lS2OgN^0Cqk2QO&(%02R&by9vcCjmdY2=cgzQ047S9{`Ff{_I(|0g=!g&XBzbw! z^z`xs68r%X{z~Is+m-1m2VCyC$uBtNw@luRmW|XvJymoW=(T$GLtD7^*?AiJo<3vT zz=U-#>B{io+@&d@reZSe3^~Pi6r?QER@R0x#Q-dV z^107*?7m#J(^lHS^5A4D%}V=i^pjdvnC6SyP1LW$JH?K1u!Fc-+lH@>85q1ZUg#U@ zW9iqgtE`3@V>gX2HooD3>pP`~OpmdS(Bi$_5@$2o%Se(B_x@TgEvWyex}PbFYI_(q zNmW|jymH_lvjM?ea+qLmPKAe1?()iFDW}=8nI`?TA1s<6zKMWNh7Nue)y~=W*5P{3 z7Z=VcoaOu_?~I_9a#p`WO}P9JGI#~s;=a0Ut2qVKGSmT-jsUDY`2w7t8zdbh^1=sF zr(=03JRbjD6NXz_y)RcpRJ|9X*dHbwx6a4qfeogM+8JW0U(D8gcAIT~7!*A-F`M}P(V}s?OJ=w3BFMKE1r$nzIlf&6%#M7-ZB4ElH0Y9S5TCY3QAR>{x*Sh0_`IAu-7-90 zEtW-pr~+=T*~;lw!JfNiSNpLxw-qV8A`wq+r4-9F$D3@JR#>e3)a;v3aUI32P=SWq z#XiF8=$oNHFo)!HDV&qt1e~Winh;8c zGl4UHRNzeErzl0d`T6sgt~0ZHF-}KyJv?fw?spMDLqdz094q zP1;0=BZ!?SHSb9OkOZCoV>5Tp3F~&~##fV#Y@|36IEG#Rwk~%Qw3GHjoTjlS;54rC;a% z*Pe|$ff66T3^12BQ@{VT9g;mSO~L;uO>W#EtNifQu|J8E)Yx&~#O9}|(2WcVB6RE3 z=N~$8uP>gS84alv=ab`Wgi1s4MGT03U9k3P?L}{#g+bXwJ#5<{`R=OT^k@_CRaIAX zmdA_5Y8nY7xH<;&TCSZn zl2Q(Hr<)m+ZyZ|>1Nik18tX=q_d3}@ck7=Rv~@}~MN|=segh@78iHX)26x3kr?BXd zWTGb)qpg?c3~Ymg@ec%3MBc>4XZ%_59B7b~DO76v`_-}MDw%*g zO<*ubfpc!9%UhGHieqLvASPKLO=b&L8Z^It6(2ECAbmf=VzN$3wAfBfb8RSD_@f5> z%e@d$)Tfwf_J*&!4?Gv|dL07TXa>EY*RzwW^EA2X(r0!cK;O<-yIr_CSMwoPz~ngK zd>qn#7;PmacRm+Kz#bwa?pC8AWlF;3B+&9E9EdV~5qESFfc}y(GP&J>M^$PDpPt?G z?-3^m7{}QE|5kc4&swJF` zVA5Tv1q(P#r@}sXCe)m4?(`I0#~_X$DiFq|?ND%6CZ~-fQxaXZ!x^LnTS4rLQNw z6)SO?>SGoB&7V(ckJIt2^^Kxl>yB`@2YM{+(GP#CaX`FsXv(W`h3g$NH4B($lW?cA z!^(FMgPu0U`^=JpacY7CB}v+Qx0;11G>UKmdzW)Y`;D@>TlSlj|vc~vGQ?QYDKN}95lE*Qv^)_R#p$+-O#nmfCx z3G`=WjSrzMG4j9axF%!+szA}OOuB6^LoI{nxbu0mch$!`T`FXhfUCnx!R7sE85+ww zSj38lalk~6{`S4tr^yJ7Wae4CzoT}+>W18_p?kHApqr}S$o#MeX6Yjx%}n>Ve&S^% z3_U-(b1Z{?zwlVhEfcR&|8{;-PSH=7)ET&oA3}jTcS?y+xY$%lA!a>s>Km1?`%%sxVviV zP$u&nsNOX+rSPmH^@79(ArYWUN0eWBvD+q6%iwZkxt@~7@BMwsdoXb!J=5zUypFu4 zQ_KC~#xg^is+-^!2K*QuUm~3|{)D2(|KCtlyrYY3 zv=cK`8*Eo21r|yqi!l3A#w+DUT=EXdiAKFc&Vk%xoBRAU63- z*+J*#9(4Xu)JXb-;p9>eo^CLav7*|f~7q5P@@s3`2DuutD#OEJyM({CsNLUkiOJ`7rn0q-k zS%%)nqlV2yiGPRyOOFDhsj(^&*jt%vCuh6ER^*(O3X5f*=#Xk07|%F5mM`jVAu8%I zB7e)#eLyKcW%)yZq4S(2@p(m&8-%lBeg#1(nrUx$QIL=~jkJCGk!3@`csqf5i7q1=)?3#4x9%U`y<(d>KvIccgpMvMH?TAU;4<^h2E235R11 zh{Rp|iRitqK=P>kJ&k+ivf(uXjgmdiTg{9+MmsJ>_nG3z%mmmr_=BJZnZhCalO1C$bK6 zZ@B}XoIWl)&1ukmlQ}&VsUuJ3xE8*KI4SCiex%!JnwY6#P2hMz5vki|b|MiHu}%kS z+JRh0Uk3{*RQS=LVe0C}Hn-%R6|=Pcb~mdpqu-7|K*-7+%uSW;mXF@TDc9qqb^(Lu z^@*9tHkp63$VVP{z5)2OYi?j+2GQ75)@-9f5{E^(my<&7eSAQCaOn z0`vInBOveyaMnZ9Ak)_?J!)AwmgO#UmeI6C`)k>)07e-o2Z7_ie_1t78~<7m4R z>@IU{Za2i%fRJE*{A2VhzlZxotWi|w9(F>(WIbD#@wnk?o-Ze;ISnDmH>PadlF98% z3k@7ZOF&(l`xiD+wNK#Rki-Ax5ZZ&E0nk_Qcv7pLh1|_fEm~d1Q{toeTX9t8!{o&d zM(|!DAD{lkqxI>7KtoUJjyei)wwnuVEAE}uev#Lf={^sKj&IvW&GhZay|Z>ciTXg+ z!PNQ1;2E=^$6J!$^nt$2$285z7*~BB3B3{K@K1fIG3EEw9F6uw=?>}$tCtBq3tfu^YzptWt-j?2S4#{&mYct&1AjmwQUE%EM z_(*|qo~y3%owAH^%IfvUaxxaO)}H=N)?quu)swjq0F%3QY3}zlil?N5TAm+_(cL=AhM8H zHZ^wh^rmMuf#P#S9mE}`o$Z2tgFgVPe(LB^>gsiDw^8@p<^uNQu*Z&rVyw%>YcI@M zYNzn7@6mZXfQ%JF@M&jPyLL<;bC<1~*%zxA2(wy;^w?(30(t|}L^S^7Lh+VU1a+nC!*gA|2h zs+(TT+s`jHZ7qm8Icd`0Pe@%vYfE$Aqx3WAo)_yMTt%UEqb zF3|TrmC^z}e5)|q@#S_Uq19?|JI)`;M8 zT73Wj`v{yR{5A-ooboK8cn@ev(O)H=IJ7d2cyW(EZO283-e2lvMPFWn*>Hn7rU_}v z@U$Qa`{t4EUUNlhrTQ7F^m4yVZ_|1`WQVLFu77cx)o!*fE!uJ6f_HtYMY~;#=<`1k z%ui39yT6d{A(pJ3*lE73%(}bb*&zU)7c6}(ZoJVc>Oivol%;F|WcMt+KC50Lq%g9k z$Dz>_a%bNKg4FS>XjN(;_Aoio7q*B&y0gnEUzmZ{{oTv3xKBrhNRb6>{v~QrZrl^q zs17ZM%%_vpk#i?m|X&;-R$Qbq$=vpxMkf0!;`?ey=z1ON`rXr(@!3&unc=#yTWg_NQP^UHptT7T|ff%5xNS=BDLMj{-4mwCK zt(=H?z9+_ZQK>5gKVzcKeL1i)rqRGJ9;f&t_xdBRuNuSK2S2Z4Qj1tBJJF4?Kk1&? zQ`2k)hRdD2qsk?xr!EhjbiPMV0CQJ^N{Hs41#F1eeM7p2<{Goi73!49Tet{rui#Nl zRki+nVfvMypv_b25F7CLO&UGda8JBP4b*C8GV=FvmnZ(6MAWtd=<{ z*2dU7JAKR6Me+P-%Uw<%CN8lf$e0M#x&5eouiT~H@ADvyr+YxG1;;@sqvrb-wTeTR zp>UCx8L$wUhH{1Y)iDO_DcG!TVt8)ST6dB@`NvPA25V)&-vtnFa$=UfYWK$p8*$I) z#lOjIL&7a8wncUXEF175ZOYBNGgJs}Vy~aa3wPCEK1KncWa40kB*70MaX(JZ-j`Dr zcQVC$3D4PUqfaJ3wUK-)BiqWRZLSCD*+o$TRsm);YGY?F!8fJVvcBh$RVs8X`R=dR zgF8&;=TzSzXV|VTxp$f6p-)q5!#!gh!1wY5zM)7S3*GM1Cqop{tnQ?h%HJ>EF0I;+ zC_g@L3R5n;`T%Mu-(K)u#9fc!WMDD_%`z_8bRu=<)%G+7t7UnCq>630c5XV#SN`oI z?M1JJc~n(P^vCc#702!_pz(s9nV<3=CEid0nDFqT$91yg`fkEVcW@(4A1?hKKk558 zqLUL)3oyU38%g@CtG=4Fp)v?u<17td^lNhtaQU}&&I)m|%M`Y}Az-*_9HTLM|Cg|n zx?49985HdX=GhVl$B#(qUn(t$&#~CoZ=Q^a0MuaVl`{yH2Hs#O1H1|SaG617zQkO;e2er5j z;}k{y3ZG=Fe$Pd)2YIl-Dm7v2%KiCO&*_UBh(ZRS@Ax28Ox9ouHw8Afb z7!(aMR2fKfBu!Ai*`z|?2$8ie_g*459~p{pUvTJ{uY>o#^G~90$?q;1#c92IIdkq? zvATTSdpn%=pE>#TtxuLN#0ZBFhL6uC=GMVevvh$dw+)848dC;w9Vnt!S$Wp32||KU z^{V61*%Jea&8w8to?QeDzw+JmbWE!|1p%wH-x5L2LEQ|(3~iWVisj*_ucp}|AVh*k z%fme^yWvJBm+A4I9G)x^6en1;xL~Z%zmn;|bLyo5LNuY1T{MYB-1{H2VBYpIA=EnA zitFjspWJ#O0c)^2Pa{Eeyv3nsmSCm_#NpF=?(@Y}zygk}JC|i^yL3ov=fewJWJK_2 zTyS0$ZaUo%ebVu=WEbICXxg}b;$n5wIOAdRe(^#>=49`qR*oSpi?1N`@FiqdfSaZC zH7n0U991)7efJ2?^H=%@*h4^@(faUQcui-<&W58hi_CE~apHDZt*&%E$cEy0if%bo zyokp$QHb6Nfm<@IK5wy=-P;~HbCOH<4zKH#A|aej)Y0g$?fVmrPe~Ac6H=4IS;(S@ z{aa^CbB&fUcyp8DCDBAjsFLxO7*LAAE8myFRM&OQL(C-QVS_u>KPeX1#Xnkzjk-w^!3NJF?)jy z{?iXL09H-HF%H|TolJ~VKr-yq5ls+{)=xesHZDjK-Km)s8d5Hv+ICN?+z{{m%HqLG zd7TYfW2pR={ER>8U7aQb^(RDY;fr```Mz1PUrg>~=jS&*2yuASyIj$w$^Y zKY5Q@ngLB1R!!)b1{x2E8Cvj!zqcl%DhI4aImonEY{$bh&#!|z=HXY#aN$BTy@4!o z&34frsr4#y`VBAo6d!=oNbsBpBSn7+ds&^fpYBKD6WxTo)+80)bmTSR_^bmh+<7Vq zm~X@pUS_7RFQb-3J9@$nU|DZ%G6hYRx>JwSS|@Y|sa%#mNo!Zg<8nro$x)mDZ{kw5 zjO)}QS_;@3yP9djEZJ>0QY;zK)JKL<-Rlpr>nApKd0flz^mQygX?lCC1^abj7BEM+ z9WepNJCS?z;YQj1`N@HSK~Lh0IR#ViKe_qMSi9nTz#3l0NekuTIBvvsB=b;uw5tmT zZEp2TA41K_F8can#IR?kBeK3!w^K!h;-bSZM(_3A>SsU^-WwXRzKPbRXuuVVizWnn z%nE~nNG zX`P>Y*SuQ&kJ zg8YzsyX&o*b2$Er>^_YtVR*HAex7*Tk?ohGO=V_7P{UdN@TLJe{Cxv1W?tya=Q0uF z&#P}!U0+;(-4?HVY2%)8_|Ds%LGQ#W#Tjjaam&l|MR97BmU9-Dx1Og@=5kmDUZ`#< z$xPSe+MHxB}O+DfD@etz7vHwdYeT-#eseL?sAxA#vx ztk~hBd#@iD_X;Gu6i0mMk{t5gA2PFfVD;YO!S|NkByq=$@@+d)?`sX%Oh_?3pfdDW zSQq%XV?AKM^*R4$mR5cNkw04giP#Vg7}oWjr4~0;6YiR(K>qQAYXSl~Y~^fPJbAS!yKO;oJtI)<_Kv4t zvD9XBL|Pfa;br`4O!Bzs-eXFQK0=VD?B(wENV#v$+34$lHR|`Hn z$YF0!ubs3a7SGdT?)w=IkAzx}sy~`*OT%U_U{Kyt^FhoB7`QJYPt@KpovXGS8dJaR?|JIehvEz8<-^U7uC|3z*VBYpgjRP4IgnrN?{qb>u&c(} zEPC!5lww_X9xv2oV(OD9SQ)KZx&)dZm9hj4VqVXuY;8dwbsQ3Oj|=!Veq68}8ZOEz zWGh7IB-!B`Yz2xO-il1m7oEwy9+AOhZ^(;Goo6^M3rG{PhiefTm64^H>u;Q0 zk{YMK-JJc5Y&fkND#mU@7INcu+C5P2U$pS{lUL;@;MMVv0Eta~*t(fRx#0O)m8o9{ z#KdrNx|zwbq4fO2g5^uCIuN=ZEmx>Lh06FHLkTbUzO?hpJ2QDu+={8NOl80Fa0HvU zri(vn{>dd156K0K1IWioNd@LkPxtilc++F;7cIX?zsp1IFW zDSwj7tuSladU*!pOwu;8uTzDLbqgmQt~x)NUk)AYV}`2GJQVPIRVQR(ldt$%+MTeO zCP=2z!Q>o`Sy(iEH?rw_QZ#***s$}15{Gg%E0Oc;sbV*mS)J3cG7=G>JXiDFTo>O$ zoQE}XJS7Qw?~UI|awGmIAL&;*=eO80I6Z}bT+t?^Pm zjvLK=jfW;Eu`q*P&9~K#&rFqnuZaCbwyK50{GO|*NoT?|iTI`x?Qd-#?pbQsvN|NL zhb)HnQME3uXp$SIZPtTg2vDpSBhQ8(%;BM5s`PxxiI%o9|8-G#{KZApF4yKdqB?On zH}o7ygz4%#J-+gJ^JX?X0ol<<&^7fPGrG%nNDh`i{1H=hodOpZFifJ%rVBD8y8mJj z2D7Qv8!HryB1yq9q`f!U>XRm(!2S<=?;Y1<+og+27ePP}5UCaf=^a97N>i%z zUK9c8ozNpi5D`%6T|k$9HD-p6{FeeslgjzjNk4Qh0Lb zepc4K*0rv+E^b(3?Q5%sBSMo=#hqz~tA6+WaE@x$%cM%yOZ8Tc7jgS5*yQ=DW$RR{ zS6`NTZ}`|E2V1aldWS_{sGq4JhCo>y)>)E3Gi4}>Fr0hnA-1LaZZRd-m)9wwb2yT5 zMkwyezw-_~HWT;fhy$ObVrp$t-0c23iWPE7%)}|5LE>Z}v!m8inifGT={MeVY^u6|(W$MvZ+g$| zj3v%$4)xfGp1qhJluz!tLH_^1l7EwxdH(6;htMKV^3-iSVba=g-dr z7iuYO-(kK}1xaH^HyhyGk5Bv-BRPnV?xKrU!>lMAxK>uhw?DKrw3@n3SR^X8(EOi6 z>))*16tMPS`uTuL3*)2Zm-|GvP@jf*6(a+5jwEIDNTrWyZ^v|7+v3K?px?)B`1 zh;|c~8fzL?4Tnz)x-9A5(VtEBKvI{G3EV6uu#Lf1g0f-n>qbdzy2u@maw#Te3UiwV zDRNcug6G@2OlhPz)ewGPD!(Qnk@+-j26RfLX%V{lYPdvb>d54Y0H9n(a1S zo-a1F;z@*#F>yXOy4t7TJ0bqZw7LIIqbWB+2s-AUWEM2%d{f8R$JUQa@*D@E*M|gN zFosIKK7+tQ5u`Gl+5bY%OF?E6CEyV)nfx-cDbd$*jG!YV5jPl|H&@+-7^q zT|3o?-t(^C?C)nx%Ad-rbi^N?GHBC>xEUXgtHCGRi~0q=n`);l^cKT1;(7>ukC1DS zVhF3d0v(K^)C5`p3H4T?FSBbf@RR%1OsG5(BKst3%XXBwxoV5YS_dDmgZE`s>$k%o z|L!6V-OaM5!Ab^rYkm#R+Dk?iLzt3&!N%JPZ(1K&ZTB@+%^M#J&?Z$|os_=nA6+H9 zUzs`b)NLktFs_z+)8{LT@qA&Uwed({py1h@q8DN-FQC#4omrnzR?@WE*05ui?J1h+ z3u19+c~*>#pA#v%z{(C9nE9J|>CK*&2c~mOCMAAbYQ$!_nCF8YJpv`ZJt}088DUf! z6n?(#@vJ*?z^5XY>*+Vq)x4~eb6{TLncUzm`l+}XYn=`kYv9(orbwwMa#Uw(`Rnxe zk3A-(?6X5VO~I!i1Gp&jDF|@d;_54mwU+*>;-1 z4~rO`szD1DoQq-hr=Zh)3Ggb@*eK<)dua2PZ_khNRv;aieQ8;n6QCtjzvx$zdC)W> zYv#dL1rnLa7?O9s#_Yx3#c$ys=miOc8t;&}%f_+1IGl7!YOD*-;+J(oWyLy}$Bb)^ z@s7iax%CsI9tnfm9NT6$)AZ*m_#e|1EvNXuotrrs7zc0YFHrw;m?=RBgqg}98H~wi z5@p}@cxP&Qzuu0MkHr^cIQh{-`AWSCGjy#uWvO_;?^9gpbe1tK6tpHgLcRW3IjQCV zWf@8@TY$8UiB&1@6O$*E{9G;zU zk~r#L$CyTLtQ3<*@j9(~7x$)bS>L0p&-(S&(0(_7>-aFXy@MdX`mSsb5og&juf3 ziawh375kkG=Sw6nQJ+gA3Q2hdKmGMv-BP?Pc8qs93s6DzJ0F^^DSIfQZamzqw$BR3 zxlr?{zswDrC7tAoM!H;rGJ@T8i19Y zOYJVlja$WDe0_Xn#~o9$4%PJEivQt^k8<&ut=x(AXn7XP4Z6}CEtdbi$#bNz63~=` zxU%cRHb}Cla32_^2cDAj=&wH3=U$SQCY3tC=Hgm9{dBT)43?n=IGm5Jxh+U`)zD{a zx2L#ZxXynq&D7-n!Xg zatHSwvn#Q6eZR3!)S*NHhVl4izEPEiPPhNjvJ3_A@lj;?g|FlPBs6qi-bP%J6;&_e z-tn=`OP{e|#p%sPKF(%{zV`WY#2g@pOEqu;pWWbXU?Vc!r+Z!TTjZZECjwk9DxcHo z`k&_yp0inn$sEtit0&e@2E@^_gf?m-AMq2?rTqv5CPZ8uE6bLN0V&zq>+`*?)p1m& z|IS71M)1|ScTJG=#EDJXr8ch00XCbty-%j_Lli@#JtP+Q?}^vHXJ7}qazC01h#r{y z(}w@!yQtWPc-iJ>x2^d8ak>9=Bmyh`pUeI4)b@Y>@jrX_&zZvi7|MUfK>o*2{;Pfb z?`Qr$c73xfBwLF()aV*tG9-R zr=yYUaT!g_ZCaEzUkgcg3j9a&vUMSO-(EkD+i;-gH_H*Wh2}1CXbdl~2IX`p$4I`s zvD&BD6?NM^##;IR@wfQ%*&z?uLLgMl8kwgFiKWIcB1gi$jD5G>IAQKtFWcU?uR#7g z;-`-Apwj@j!L%&l-d1b)#>R|B8xL@GT z0eVEc!gjBiVb;zoSs~In&%`DAvS}sJWY8MwBA*#B@VV;vUu_KC`{rW z)f{1MKy$RsK&Cu;w4mS=XP!$UWb+YY(#>#{Ek8G0U1u}W;IUSyv7LIb9Cxxv{($9Q zSdQG&hrmW((5QZS<3mifo&X#)Oxw7Hr}9~E zO&WUBg`#Ni@SE;g!i`;_QO2Ld3YHdppQ$(E%!S)``ry%?L6;$lV&O@_TT{n-}|8&e(7{|L0+G}L?p7tihDMLE5Re6 z3iBGC5IdM93#&?zMlMwcpqs?~6WO6xLcya2nmVocn&2Xz!;}KMRTG``C8$Gz>ANJm zxujwJ2m$QMsHT|ChvnZ_rOUqp-~Ig$NA4G)0+!<1?ATOPHG>CTlS>PZPUXwP$jotO zIjo_%z!GEC=HXMUsQ$U$}^QB{tv&GilmwcRL;Y9-JimNvHet%f zZ~8ce5b%(389tkC_H7SnDD?UKG?<8D$bc=RjzMo@3^JOYGWytILU~|RyubB23W3oI zyubx<>6Cr(=va^^(a!ron{Kr=Ryz0Id+WR(baXQ45-cINW@g!onH_5#PDu4D`I2JxR%udsO!J+5UvbZOjhu`sOeP| zc_Y_$+-xo9Chpq;Dm}3*J@m)4px^Z6dsULF$t`!r{jD35dkXlEx-1U&y`)hQ-_@tv zx#84Jm^Oa1&Q9l6UIk26kjxX0g!P+<^%^cu&FNnQ)nxiurMs3BdrGuX3fOL%=AfPV zZ29V!OUiyqhJZ-=tJ&kcUf9r@#g_LVbboy3J>njR`&1++UWvHLX3d@8Ca=C#!eMZB zDB|LdnoVPl^T7}g}R6buPy1GH{kEq zPKerEmdsXX$jt52N`HR9oxp30g_T8-9?KD8!f>t%SPet?gXA&bJRvV-?1NoL-S?R@WY^c`Cj1xv;BU z!g#Tke2ZZD;Jz~e27B-1e20xK1GC8HZ!RFc4WZ{trO5SPl(HkgX*h6>dE`fPnsBL$ z_)KQ!v%1&{=jCh+#VFwWiX?(wIAXGuSqGjPvG_jxI;FTac%g)py@(Gw-{nwoH=*{E z*q4vzIHHN?F%r4~OrYB{<@fCp)m}IBCcbPlsOxEHS^U{3N72u@k;l70YUod)<6N}#FE6yKQLfvb93^lcOqIcLCx^`7QA$}Buk-#?!GtfUL|s;o%#0|6_tW1I{Fyy5 z%mxIKwWFkZ8d)#kRdSacb_HCU8x@j#E~uHzN2cvdeSsFB%bn&;4L^~-2ZL^GU`|Gp zy1<;xUjDZ8r*@=>Jo7GjthS^Ll z#--n7mL*P78!lcQvFP!&-jh=10Tj0a(M6V*=kexw$i8eDXzoZC9~W2LVthECY5g<* z!yUaxyAZSw3o!R9jkC^{ZD;m~-RyWDe<0;ZjMt*#OK$ccGjf=wAqSD8m=+>H@dOxoT)NCf4mMbcHyM+S_67cW$98`LX`B8q4| zi_yqDf5-BnSHY3A1-6FUUZgBS4+^rzg&rGtn)PM!ra3!I#=@MwZ3e?? zCDI{`5`oQL+`*$RZ5L&8PMoZlsAzNCZ^YS*L4yqRve_9rl%J;_7Y>lP%clFc_~H$x zt>CSJl1{Nw=K>2Yhw}tGE`)?XAgD64ED(mJfA8?L_32%xHhlLj&{ui2${5Js3=t|m zb(7d^rcblAq(yT2u&13Vy%bbkaXnb!t|P>4S-wG{(v(JuWvHTxW{wFKUM;< zz^Px<*Gmy``8>${4x9J}92X$1R#7KvTHhU`wL4&eo~cr0en2130O2d+Q{_QoZ`j7i z+E`14$=A$!LO^`eD?^;pe%vRa5#k?5&7t9#>3wKBn5io5kMoN0C2`QC*qV!AbM@XC z$;-=+Pt_i&+-fg?}a~pBvgB|~^*(sTm*XDO8VU*D<;LJq6 zsEYXtk25;n4`7zSEg$J(pg~0|Lc_%`aTL|ks8xD*T-qx1*JQ!mH9yEMHGES7^!f$# zX$x-UX{nVPDM}~5Me#y!)+$2Gvugx0e(l?vP0j@=+Zc~V`>U2|Q-HZiayS~A>36Wz zfwG&9k1TVKC8K(cti>#sv~T&e*2W3jdZ!9_SF?D&vgmLN7$a!nrC6H{hm5}0vNBbr zO+L9+d@q!EX!IRfN&&9bEbF`l-qPy@WP3&GO&S%8UH)G+VJv=f8gNt~DTWl8z=gw) zW&ri$m_zY&fnl~@&54}mA0KHltua}Vgc7CJIp&glujexb|4J1li)NG_@fdnl_;NQo zG+1zpg7rKVlHv4lZs5+dlR+?<*_>>^gLuS(JTbc6Fd(Kbdo{MBPycUou=!OEAYh8M zZGJWI$a?VHf$LBWk1PI)R5N=>dFYcyt`!7XU@>$$fJ`vj?}<0szw?CS+jSVx4JB-? z?L{=Ba7b@F-^vV1%x%lG+`CzUqF=IDza_E2$BM`x1EGnIvGn(zoA%l)dbwm%3qc4< z=F;QVi=m?rOPROQYv3%v>qP;mkDBFk9{ z=N+zvpc(MF2_N=jMa<$5X=k_9)S!x){^9wW8Ovd@-{zT}WgIi9n6~h)S$;gHWRi;i zGkN;;0AhNK*SN-O;7+ErM`J9d(G9B}nC;YyT)v86q4SxR*nX4`4JUSngirN`@%9c`j z#h$2%cNq#xVbT)0LTH->>vJJ4TF z-#M`2eAnkEPkyjP7O9luRxEM%bn?{>hF5uGl3ozGT*qK|jaECCYL?i?UJ-ML)yhqm zil>TZ*Or=ee~?UlK z&|EhbpQuB0hbYFOm?eNO{oIYIlJ;+N^7@lM=)w|%W?IYpm<$O;(c;8hHOd)Ghrw%2 z=v9ZNh;6&ON*3SEahM+H{$~;{pN<)u-7oz1qrU@|x}sJcs!$T)zC)EO7K^HWQ54Jg zM48`Q+N2eL{LIy?s>|I}*;wr;pdzRDz=_z&eOe^?rmXjxj2h~GU~R`o_`}}YB_J~7 z0nPV$r|#gu_bkck;sKLv-*vMCCZ!y;el-8tFoegvgMd}D<)vv)Kfl^=t0_(ovBh)z zuC1_8t=fFRSt(Ur+*OYCyFPoqWBY+8x{{We|JT~4I^!+1G0ap|(kj2<@&|&S&rwnh zHBgHX@&0q~^)baWqy2Vc4(F};6b^kGodrMK7M&BmH2w*;hsHM~bW-w-+?*ktqB*g6 zDlI=WIUa0rj|205DGFYDCR*3KAm~tOv+Wm9Sa&@Zk3c!*TuxxYUb0`RoBXx6IRTi8 z1)r1W&X2K5G#gckcIZ9rhni)Y$dEcT{VMXRVNbI};zew&GwQ1%vDAJvqqtq%@@l{A z%vwcH<;MCR_h=Pvqc#6XP^V6%W+TRnFg=N^hZB+(Y`xXa)QD~ zky1pj@<@)_r4xBfb{gq<7)6{$hMH^O7p_PNov*?!Ax0Tg+;z@K=gf(-^4Fh4H*TyKZ~d&ICVgU7x88(RDEh|Bl6$IrzE2d0U6?j>3#f2t zauD;c_v%lY-=>A}=7*o9Log_LT8lE9oY6BatrX9*pZZFUl&j!dm`!hH$tGgPxBF!5 znTS3heG%Uz-}s62umw3O1WXV>odoyZjCYKdfZe80)QOONhGk!X< z#_)O$1RUyRQL&z@e_^%YYvqFKq;#>y@0M3Jf6M zi%Kra$v6O|4adLF3oi)>$NFEnu@H`(&KjJNr?>)B5ihazcV!D)`DbX3V&aSE%F*L$ zDOHkC6^R(ynhmKoN#7069?G*TKt-3iO))CT-7HJcqt@dl7z_zqZOr7gMHSl*IP_@F z5=WaM^a|R5S9*0DHvO)GL(LEgc6JD_PFPr5v3+gEVZp?sR^xK42cPV(2x_G1?lG8C zYp=_2p+`O069)|ZlDs3aK*?vy0t`)m?*$8wwt(yqc}FGjM6cK`Q$75 zk~XznUY^<4?X;K%{kXX{^?096pZwZn+Ej@ zwFlk*Ov#@KYa8C<>IEm=8ux|QeZ-xQkz23b^^W&hP=NqKhGvLB=6Q2(knW=hv;NPI zq)5taCjo$mh(o2xQPcso@tvAFV-tuB*Kb5&t5*BH%CatT2t&j!)0iA#Pi%&4=gYdA zz-$YzE@Arv36AE5TteHVwnqcTrlLk!l3rYNank-rAT7?|QCb$eC+VgdT5CXKVCK~i zxOa!rNB6Rhxe2GNN%OMrqB~|mN-=jg_i~BJOb61g9`{gYDt_gg5gT-dep0r6glcF> zr47VQ0a}FB_hR?yc|6Okdep#W#dv}S(DU8GLxM~Pgp=z#+e{y&UVlwwXyid%88sieIqRhO6M@e|()vwC=g+o-td z5uR*~Op`}i7F5Q2oR92TUn7nY=#qSV(}0PSzVK-4#T8aYa37c_L*B{+!~H1cr8u1R zUhI5sF#S0eE_3zsxvck=3thlw@?cap?5M$@Aybk!JV~g4x%j4hD$1z*S3{n$hqV4h zC-}=O>l(aikMPk}{GFXf{gs5fgLT9}TN836rcL!HX;gim!S}63@xh8H7^LcMa!1_| zMy9ER%Rzn+=eob9@bC{<_%~R%xHMiP+T+eE;(Jm{v0ar^xujfocDjyka(!im*Up!! zv0JAV=i9pAw7-6y@%T)EmI}_IjVwf$EFS5|~F8FE5%)9(q;E8?vX>Cv zR7?FLa$kirW__K!A6Ciox8G+1n(#V*Gu1dSlKu%*+JD}DcoJ@Zp-oS(<5TmF#-9p* zvEFszDbFt9(??Bwvwro97W+@O@bIa6+3FYJw}l=6XXJfWG0pb`Jjw3n1H74~MkqR- z07P>DEv~cKIf|DL-i=4hl)Qj7ZRrJPCZJ>cip;|Ag%>ddbu;Hc5`McF_*_rhY9@54 zA?CL5uBEy+Ad-xu&BVXe+&+0*{9b+PqV!xpi-h2tUjnV3<8)~%Eruw)!TDQ96(rQj z1n%=3RP81y#oj9nE8X5UzDEY(Tdtp-m+IA$kJ{hT@28|N-7`)@U`@Aznch3>G@tN6 zuhYcLM+-TCo3_4Qh%xtyv{CYzH@`6-IkOil=8l$+i1j=VGwGrF{F^jHoKW4r{H+Oo4Ls_ zkSrPzw=+FaHu__Q()vTj#lZY#AjOq8UxLVLXJDu$dveEAy(F zM)^-7_k8J%@Icg@iHgdxXW7dF(ffi6VjQM0XQz#cysD)n$X!vp{QfN1n*8Z;F0MD; z)o%^xgdG;K*&jjpV5_o1O6KD)&MmkJpHdP4!|!b>Rd0&n^VkmbKl=E=E{4CQ_nfzgAvnfD=S7%=af7{=@{E_18M0k6Z6*GW z;bGT^pETq>`_|Ur1t2D@Gb>qhQi%dmpJIlkTfZX)M$M6v355pc7G6>MOi7P> zmzbgwYv-f!xU7_8=Pd^hTk`q})sZoSZk!=UG0fnijbg#4wKPUd;QZ?@QR-B6_PS7Z zeeye=p?G*(Pj5BfnxLZw$)G|Ex^AqjrxcLU5Z@~8Y?%=T4HpD57trRsV%Xf z9{WQOukkk&iT&4j#A#2Aa*fpu)WU>drvn*kwWS;*D$sL4v*A%o@_9^q^Y)ScCapin zIw&oiBFemg1!06eyjwm`i`q8|e&uw7MA}&r zC1z!U?+Ly7JQ>9U6}&ckuFtgidFKLk(=>BWkQhk@;R7Fds&FiT&@6p`Oa1tfp2Vg1 zkbF=D0D@Q9Z*Y{%(lN;#*TjHZQ3J&We50G?O+sHOZb+Mj#jwg|133{ONm_+@(DN_i z3VxtI6)1(YtD+ySpeZQSz~{;-A4m1Qkp*iU?M>SXy+T$O#cx>pUfg#(1q6q(Jfk@u-G#B6rIVfNu*mAzw{x6c91aSSpI+7~1dpAkpKtJ!$PQ5+-=?G;%Ss-BIh! z(y8ddJOSk=ULULO%k|G4uexrLw)+F;qi)+v>;$LUlLu54q>-`l^EdAm0BXG=wq-~C z#>DT$-b|Q`aym_m1oUFW(#0rlk=!2wj_B=n86A#T18Z2JZE9LGDt#iLe4>Hw`}}N` zKfQF^OysXg27x!%)iZ@=K~^`ikTO#mHyOn}ZZ5k?`E4bKKdYMJ!(lp?SPgrh6K^+P6v#jfCZ$<9ZFHtykU7H{e z@ml?wA-6%(!}^TzH8euO7^`Hw+8>xkOu@2RXKv|=?61pz6LmZN()H&l(CFz-w;>*G z#uCtyc0r0-ScboTc=j)BM*J_EnWJ?O8>aVf%coY%%)vi#KgH5cjUpjC*_7hWc^vLi zO4h0}^q+q@wfoPKaHF#$ay(}Q9vJPee4h^BnJckRNU|Gxn`{&!A?w>Ka#!FDr~keF zw>*~tf@_mAC)-DJkMlOl!~HqAXjb|eq{*f-qd#os#s8k%>eXG#c&wwXfzE_&kmG-rfS|kf0s&`RshrRii_qY zuR6gR!^R<(oL8VW((oq5P{jVQMP{pi%CKl=FdZ*wPuJ1V2jk#pqM zXR}o=x9xl5(N83U771y&Ogq&mvdubv#HdlqeC@;jj%yowuO*mBt=5h#d(aMaCra?O zGX~|W-CaA$teR?f`HB%D7wU_HQH8vTILdW7b{bB+vnQJZH(F);blSPgej6CH)-D$oD+X2D z6n7lLytD`D*3%$XoC*9?Jp)iOC036gi9vl5j5SRTg5Vz^5n4sEo#qWck z0Ks^n_Cy5iS7v6ND^Vb1nNyEb=cBt6%jCzAdTq-niv0QL^4dtfbQ7S)VePy8a?9Qp zPKc{{#Bo+vS8JI8;M>LrgHnGe5lgr)8});!+oWrgx0_*wqPDDJ6Y1|H_RzHhe!SWo zKWdnfVcvjUE$qo@Z)?mW(nN9&w~dig*_bxjr;1TM*B2C{WRyn>)W$CRDPE;xYm|zl z9|%zeU(c9DCSawVPSuR-fmf!R%0ZA)6xnyu(&P2?m#fRI~uhH)I zvtr$T!H8?Ok9!)AKpty7WlNK{odQTve@Y*KFqQ`}Dw+I?PyVxha!;s(WE1{flHm91 zSeBA`X~?0;fo^t4cg8v9siB{8X}ipJX$oKL8;8Zz6=r9~2u9++9NN*=tTHWESF!+1 zw*olYC4VK*Gg&?+<%zRw1c2P?6;*nEk=&89H?n*5e%+$)=S^z5r6!c?-uW7C`Bdrn zvzv0Q?4=@cQ)_-jL^8k(f7i`Bi`7If&GavsJDGThf^h^mJ`qG@Y^O&pvPqF_wJWvs zEZUw?CE+oA3SmOm>}J{1?r4VhRO)&G%Y5QrG2uYj=IZoJunjk0Wo|?x#AdM!sB{(& z|I-6IMXx*ktRL3>{S`|5~}>GAC~P2@xA!HNtXmoO(MWVjGtTu6B@DPZQ~6@0mJke|JpP+_4huIhFB4 z!WQO4n%(v!H~+(cjp@Wrw2gY9)8;y^z6rO!BGl)Y#-;ogz&w=rZggE4B)iOA{|b;%WerL~oFOTqVUC}IzVTo|2YK6(G;>yAZdd&eE(i!SsBA75{z zqMLM@#^M2`{s*kx4bt9f;|aE}Hg4bt)Y}kAr%ySVw-{tyidd~WaXR%dD zL}~6@p`K+Gjt;BH^$ger6xRPL_I!NV5eY4OLwdTh4_AcMSQnOE6k|lw- zXrzm5tbP*R7i-Vr4M|x|aSQzSBxlQh&Dz<&XVC0m6J~qhwO-s6UvvF^d482`k{{`p zeCQU^Nt4sblnp-vnC!TzBAfZ3j@yn5mAUP@N+zb9PN|1gGUV}iTifGF!=ax2L_Qo` z;4b?)ip}`v)RBb%ltMa%AGSmfuJ!fS!BW)4+zZmd+Hg|gyM$8*k5Vl;saUW zC?AH1XHsi)rPCvOwA`P`w*}q;aYV&|LRq@;LBwNBxB8sxgs=@Xe}1;cUo#bqVQy8h zQo31C$)D&ARBSQh0DW+y1DWs^9<*4@GOy%q^3el0w||hNPmY)D-oT;9Ln$-1lPm(& z{f`3COiBDl_w$}*nzE7r!J;Lb)2EAQ>gRE*bAgb_ni*99a(}Fzx}Ik%5gk{cQGiDf zfxWjY{zBsAtrXezGr!ifK;ST6h2uT?*T51m?oDbDILx`KR?i+|Xr>B1cWuv)$*)Gf z*QD7Y=q&37m9iYN!hbV(H%ku2Z6B1$;*)6_xM6^N5kX7A|apTXH@^7-*Uj<)B zd7l)&XytiQ9q~Q@=`F$Myx zB?|3KQ&q#P&~!v7`7}@)HSKW}dtWC2A~(p@XP$0!YVu%SAi24v4DxtwAl!8Il@z+|f+~z;my{D4Olk(s;a`92`uj(Qi=FhAS?T6d4LwbqeS{Hfs zyULV?qzaBdTb#1T=$Z?PQ>Fq>Z-yCc4}~A>OqqFvijrct%jyU8H{rR)vxNz1*56R_SN2o=-}DP3*mV?~Nf1%KSevbacks z3|rUf0niIues_1K3O;1f%Op%ZOwJ&QU%*UKbv{o{QhbY(a;sc{xL2Hq$x^O!#&_SU zW(7D&U4bM>2P-!4VM+S%oXlFn&d*`+MNFrM%w7lO!K;QOQw(_Ky}EZA=85pV{P}I% zVE0|LrmUGELUlkZVL!9w45TG>)<5dEZJYic=J&RKyY|jnNMK#?li-3(55W+k+dSa* z5HXajdRmR=`P;ceCswq3m_M<0);uNOCoxVLA{QR^krLmJ@8}IRl4T*a!i7`>@agK! z!bBz`nhZ-E^x)UauoW%ek;ocJv>|zIH3?d__b(lfkL1 z=C2G$<|-#5tMJ@h@^s%VRnR5!aP%)Go$*mWOIz?#{Ts5F+UFh==CGGE)8#K|SgAW( zA@r@=*PjS5lX@(ys%iMdeCYjl8K)}JRWa2NGkC-ipYo>Z0qau~++L|9kY&NCo$=9E zZZ40O)`=>GJN&jcg3S>F9_^l=B030;4ycLxt9`cpQ#hRw*}^SN5{+9v?<(NCY=knh z{yS3aK?e)+Ez!USAo#*x*ipFu=o;)jaLpygm8$|KZcae!X^DgLL@lg#tTGhNM^{fi z2oSKI}2<;R7Dv_Z(GgNzv5O4#sIk*`FA+Xtm?~7-yIMADl;`|g z8QoG9>x9HfI59fRI+WY~+Ao_L@3JWtBKsU^YQQ^hYXV_jxpf4?J8mBjfa>w5#<1 z`@>A(#eHgOw@l_vq``N6?-YbMk;`!!$RS2`s2dd4mk#)&oxB9opCVlOvgZPh2P zV9nzgfWmh|B`aNn~Dbw5T8LxpusY&UKb=X}WS6Oksn{3QKv37bElp>MCF z+@#iO80&^Pz>s2wAiJWVfutoY^Yn_H2TG5*zU{@olT))57a!<;4G~uhP2|1$!kze= zrbjz+5e{ApIz`*gg6tu^k}PH0mfj;XMlK+H5)GsS4e}+xPk5dtGrj0Jf(#KqKTBok zaeGq#-K>f@`i^M0W>?g6j~J;x^n-T9rK5&SacKWWazDP>O4sI4Vp zML-9-jJ>npm>m4R)+Va5dLI7dQ-@rt@_6dX)>iW&1{q}@n8yWF0RScow0NqE-ym?@ z+B$?FH|V9bmxRBPRa*E>hC6-L<=e5b9ZpL?inK`G{5esSwIuHmOnmKi*L3Ck z!^__B!I~FD7_m~5q2{_dw@5dQZ}uc`##cx3l!~B&mqz%HE(UOwe^`h1MZu3&)dEbg z;dQPX$v`CGR);yx@4Jt;SFzKPFwONGARp{@JdxX$wRA0cVya0W^r7+l6gDgHlBi`$ zwEJTnFkO(Mc~dEtA=a#oB2-7%m`s?xS=Fb$%v{lUmc;e|Y9#KhAs-R_QFQ+fz9|6K*UV33 z>(Kj*J8vxShKc?ra09sTt~R+(=O^p@ZPVMIuDp!Gr+Rbm?%@t{&YFUe^lhYjvSgdM znI!zcvJ`WI;TaRYY_|E)@mio?IkbpBLR6tsxeMv}0V7gX88wRQrC zPF~D~+FF4*#$#GEcIpKm6sn=M^XUN^sa~<{SI|DnJ3$qOTf8^+F-suDW-xQpZ}2@8 z4qam`T9{o)@k6-jL4qHI=QU`Ba68o%U zwfKm!qku~F1$m?A<`Ybj==fA%tp{UX<0hZXK?9DOc(|2?r5c#*{(%uhtqJ)@F;$W|@fPmnThoXgRh^XC+TpeaG(gB=3QL)vq2ZA#bR44Mq z)Gq(r&AQsxbYemA$gkc#1n6b0B((-Gb9VA{KUZH%XArZ2@dW?xcB%sWhWD@y~E%$s(dZ@=x*JZebv{oyNNm}58E$S!RraVXa= zx1r<7(>wj_n$sugAANoJ@1;Gi1_qf6UtTnxiXHAT7}62by-|Z~jM;67TYjp|d_~El zFvY3846=tMy1#TF{e0cThLk#dX`K}Rg5hgTE=bfVTOd8Hd(@Bo&0ZSw=LdB2TQ(hQ;xm-M$D2v zv05lR18u9S_5A8~ybn3qvu1a^%FQ&OR*C2QtO)}G0ozU%`oVN*Ro}e^qBu7zW$hGg zcIf~T=nXhc2%TEzQi_72^^NcSxvgwcBcwLyk=fjLD9f(Nd7T}K;FN|WJ~RFN(dP)? zqw(Yi{neL7dGsMSH`Xq`ljl9BX|TTgYG)ZrX?+6ggBIc8MM{&*vxyKY$Jbcd+r7JO ze-j;%Bh+q@(@maSz@lKxr0d5Ac?yJ!Bo;llA5 ztj(**3qbh@BDIq?X?jhC!_1R^P3Q0~@iS~M{oV(FKEXQ3o8^1@7%pT>DQ8mi zD$;mkNsA-Bp=0F^)0GzdBD30#G?e7Lrow#TffnWB=4^PdR8PX=ec7=ecgY8TeOd@o z`QDo0tRHbiv_WH2nrn7rRhmU(vfjrW4yl4wciLVC%<2r1{pV8ZKh)+Q1_1RQlH-p- zQhpy%zAmC>x>c?(QO&O_e%^zjqeYt#C0B-Nf|`jukM)WTKM2`8W3^Q2L=MRP$l`Pr zUdAZ>Ca@WYY@_bJCsSXQhn{Vx8db3IJ;M;5^Ko4EpTBR`y1ibq6ESWuv! zWN2+nmc*03#5NcYRVNR_wF(-WHNkqGY_M+~`vJ#8^xwl5RQTc?0A?HNQS*?|NrYdl7Lu%A zp6z6h745jIz37-hqlInJ1zDOS*&VV&kpI(&S3kfBrng?p=U~hE(RTJ;Ifg|wS1C3s z;eD#8d3NS(Q}5P};p^ovDmLnNuuz^QzyO%8RA8&P$td*6{JZWwDj4a{*k&Ykr`0Eg zf4JBG_#pU-3D`{efMpISsMOX0S^zb z*J~_{6JKtYGiXLBFAUHZrCI=v^4Ej~z&(&+eBP+D(sw#(fGEh^DchfR$hG7Z44* zx}QlhBz_)nhheJooxkHe7mrC*5~V;-wpwpZbl((*Q=4DE9vt?3{%8(bkgE*Md05A; zwXLND6dc1+jGrg`Z5;eps~X$FEnrCt(lpOM|LZsY6B~w54rtxE#uiEcFTO4t;XhYs zpZNfceGMPDg7(Az^4h;@IK`s^7J2R2qT=6Ca{jM}1vLv`8Z;>3&$mhc)^q%$HU0HA zxof~8DXx9-`TI=u|FYKLFy%kZOh;-D75}e``RA?pk4sQp0~Sf}_M!RTz2U#^BJi*O zbKCxl-TnXJwk0vf{*b*om%lpqNC2v?o=fcJ+B}4pxrAkferG+bYv{pvh$#f0p3n4~ ziF3=lFM#uy#V}`ypX2(C8o}dcyS`+RXT%S%E)!KynsK1{H?NhHdK73!+H-j%zr*{l zev@)ie}023UmwMD#-6*zmzlSy6k#+JAv11Gwv`%Je&K+A3MEjyA)N)L!DR;VS+uIj zvVyZ=d2Qm%z%tSb+MZ>k3t6!^O`Seo8B73FDIKVEs~N>za#xBcMk3?d)?K~qb+Z>0 zv$BP42epc0qz@WeXTdMJ$&ZL=G0}@4CU0>S7MEquGNY!*RbHQ1pgl~z;}p{Lxr%zX zTHe(eRmN!6>=|jVTda_$dOP}9IzX}hDlK#j^6QO?=Vs|xi-$um6YeWNOnmG_CRAxy ze2yi@6vR?+$eVUvY&UoMTO(B@kb+(FbNZtgCM+y}r7d6|T9tXhI0H2xc=JdtLy@HQ zLZgoN&XADT%qQrB$9Qx5ZF?=$0#v&`9k$t;6 zcscK$+C6*hba5ugX)ceS8SJKQktXI`Sdbje+O(t>A!wRaw#~^kO5i8kdKsN6km0>` z8p6`IC*Xx1x<8L}JoBWFPLmp4X)`ai*WGIPdC*s6QRG#Sc{+{z70P`JvCF3GjS!5y zLYwdgUJ)1{qi!@_99~7P+a`pySnL!$dCiaYh@H{g^{ z=uK!euZyWWvlMff&3DQ;w$)^yg$bJo3SU*Hs%LTzmjWDygTX*w+YV(DyXle>zZdQO zzaN}bJHmwVptvne8*E6kCuZ~ZY}O@;5%u&2@p;Y0WSl^Gtl85y{!5GJ7ANKlw$zbM zrrVkicWyZ{DFbB`gR*n^qJ(!h!a zt9q@v_rKVC@2;l)FKt)>DFOin=}lA+>Ai$10wP^{2azDsdk2x;RgfMKklqQs_ek#q z2qg60ODLf{@i))R-1D6^Z{WU%zp}EjNX|L?v-frFa_y%$zK_2-2xop;{rb-cN05M_ z=LDx6Zv&MEM3G=U8FF!aAlWK>Vt6+gw zMIln_aF^incINJivG|{QY2kA-ds@Jy>lf*`eaVq&Aqmqw7op$ZY3?Z)~h|1 zHK_1xP?IqfQ(1Wu(Ec4=Y!%ja7ua}4U$oLljTQK^Da z)hPXr9my_~2 zbvqymVXZOL*|hOy9R`J15mNq^cPnl>J~G{s2K)pQCen4&V#3@Cap{|DZnAJC#~-Q_ z0hjk5IHK!=XYOUcRW}hT&ebrJXSAO`^e1?|D{rv;uLplW>TN^)T;|&a)?`n;(}EnW}y^H5BfMc~N6gYM?c) zU)2i##=Nmd-tWq;D>3Ly#}z`wu&r>U(#tU^%mxobukzZ;uj!Sv&W~*}f#>jpt4T21 z!6;l^4vsrCr%--ps9k!8eysts`?@84t1u*Rqq0UTURD=WsL9i|gIO-Tdulem*aH9= zk7V4$yw7dqUo5B?-oeZ}?z!xpe|dJi@`#^!Btt@>aKAC$UfYIMp?`L*FPf6WUvrHU zGwPYKj#g<~my}8}qcUl5mLnmT5$zl_uYXH`c$s5G-jv`mU&hIuY<*T2a62Nkl^IYr zPX9DnH$;DNHPY~54K9Q`tjY&&SLanF=hf-m=!sYo@u%pRe3TIJThU=+nz6G%xHEKO zI#BHg{0B~Ytokv{=i3VRE6fDy!-iq#3z5rhfgZCWpq6Ow)mC7=`i`TKlWDVOgu2we z&U%jm+jU`Jf|r{FMwYo}YachQZR92Jct zTY#GMFa!UK)^*GjZQ%J)NEBtztqIYFANf1|Cf`dwuvEa)0;FzM;{Eu-7Dm<90BxnQ z#dG01)yfsr*&YZTZ}Hzj$m^#V`t{oD@lgU`Kb_6wl)6YWCvu02gLGp!1v4fOQ8NwN za!Ho9s`GIX*>rYxr`Rp=ojucjmx@bGc z`vrG4qgqVIvJP5d*2>?-MJ+YWGhcm6${37~%E>!VEN5{o%Tr1V#YM&P5|Cg9zi(0+ z&!hJBHhh5hr-s{?RyNZd+H#oHD1XDEHv-Rsq>S41b5E*Kv9T4W=Xnz&9L0P?3uU zxw}0*b-o<0#Qo+d9Zvl+CoJyzPyWmJB0^6Fbn6w%reJer+1V|2m@C-3^{qL*aK@vi z({26fGLsbNj|2VrMurJ{AqSlKBwCltVqd*e4cBfkkvtczCCa}bqkW4erp{3#W*h~B zhK=7gO})R5pT-VdnQXNr^Bekj7f)jZx#}VYuy)UuIMI8n^CIqsWQ+lEV3ESEEIQ`W za^VtsrFxkZ8CF&f0rLJ#m~x_t1u$6*YwXkZpqK07qB$#4PeCvM!CElQ&fS31m?K5+ zqAW^sfRxbLu&l5H_c;!JLsYB(%%N{`->C(pyoSGD%dOJtRPc)X=yTYU!M(NhTCK6f zBY9*W#=jw8nCuoNOC6wE+F&3{OxMX2SU*2UoMv5SfDIm?xOPEDZOpI7go-_cOt>B= zcL;d(8qKwAuuHrpU{GhQ^!G-vI)A!DwW)7c(BmZ1oCFK5_+ZG9KNy06znzq>0CA~~=UnLpQgQSpmNnNCzXb0q*-^c!ue)ZJb8R54kDnbl^-eBK54JSVe|<$>ZJZ`;i@tGQKoDlkpJTePR!L zZnyu*z>G<{MRVp+2I^J(rzjs+`@IX|^@WkZ!K;-WVCYRMTv`g|11cpYrDzE-BtMEwcfwMB?Rclk~S z*8VhVYwkB45r%J`Gmn$mwycR?%rRUp75KwG44hH9@xm?Q3JzLzEm|Cg+<7vuZd1{! zgC5lTM|z22g^EE2m2)IUL7&22aOjjTtWNS*i&{>aY5E@U)=58bfW+YE?h&1$pI8eo z)NP~9hQFSldPrqLf3aT27M}GFR3ZYe)A}=NNEkFh%_?xoh1?u3S--pF1HS4VPs&3d zij46*MU|0)tCJsB8K=${`@R8ZVpq23lB{$YRw`#rtg@l#0%#4A<{2(3Tu@qg#yJ*^ zcQUm4)8xeHr6nzLo)BRugXxrUY9&<9JG7TEJP*FQJ#}7cj(g2!^R_FGGjnsJ&yKHk z%lYF-kw5&&@g-;M!JSwC_MF-RP)sWcBgk)MMq zP5r#XYFd2YVxO(*9ggY=b6yBiCn`ctDZ!qi4 z+t%|dkK@~g~=3$iQXZ&t5u{Znsz6^W1F{qYZr9aENA7_mO)QIpPg4c`zu7! z_vHk3>|zAj=480iAtE~%8UY)OU)O|8848r<1MfstPb8a3#;7|p76R#V_eTKq+Oi62 z*$eM>QEN1?;zD4aBYpP8s0LXLjNlzUkJZ32k%^J73Bj?Ojn35VAqiNtTrhV3QyBhH z`s0?R3U?k7k#^o0Co6nRv;0?=Rysn2Cn8zpY9oZOvN_uVBwemrrm9SFUBA)=2;P{e zktOUe!Eq76Ki_PYfi}Ue4uO(zm7HF+34iq2=BDdm_Ose1gJ;u&nJCvi(@VY=`1BEh zV*m22prq=9Ihd8c+zdkz8O{8xOoaooCX@eO;7FPw5mV&@?*x85?Zci&UwQkfgLMtx z@T~})H=!7mk?i)HzLK67;Uo2ar42ymd9At2O9uXTR*cv*p{IM>k3HLN*9{d8hp*LH zbQLPtJo$ZyUdrfK(0dA%nGV$0e%-?B)I2ELGs0pRdDu1nQ13&aafKTWxa9^n&b>dy zIG@MHJ^691{-~gpQ7qMrrrR6jP+gw-oLKx*DHqn)7I;|YoCXilbe}cqHJKfr1qPk| zetc=^A*Owyog!I`D)mIWASGV6TxV3-3b>+nZF3*shP8wNlDt`wMq>30bIXl1b2obID$M zA!dj=OpCVrb(G8aOh;3`X&ivdX{dyxb&lLW|CfC#l7{Y#L2wgv_=n(%6r(u&$W*wF zbeYN9NA$$-<&1|E6nX|l`h@+0f!JkUg_%9T04(>{ch-chnPFZNsU8HWG@~^2K>bU~G!ZYZi&YVfERz3UHh%m!%oBV>g5wJy)s#A^K|5;yJI;Cl}5 z%lQ`B{nJTaAg@ij9i;Uvc<@Y<&tIoN>u7Y&0fi{Hv&`TZbVxuLwSNhkY^nmrd#gSd<+7Ow+UgNeci#+fvgnHv0xm?7wguw^{x^mw=@V*{X zsBZAW&laJKXvE$j)HdbHw~~?x{(?UcB+D~;ADdyu4p~@ISm1m=%<8KR71bB3z}hst z(l3TT-*G$h41w|Je(iCTh-H==T`IJ+0SY>_;svA(&FXfjyK$4hg=k0K%I1^;n`g@! zYWJO@yoGb@0%h^MSYLGhGWlgcrIq-YR5&q0`s(rqOQR`lTVNyi)$Y~eR9^By!hH## zhP+_VnCr_V!@7fL!HeP3E;N;q`{l7|wfzdk4LE2+)M--MPgEn@5M0ak*DpHM`n9%N z%#=3}CqOP+z8;zx>Hoyz>EM4Yx|9;~hjNgot4G1O0|*Mfq6qA|_b5R^@8czE^xOb{ zBk*fE8~Za)wRYnXcm7cn>*N8+i9X((&atp{*E+ob$)tY9gAwHKJaxuTEmPmUu4TQ# zLU4#2yX)j2q2*f#+bjJ=58up?Bv=QCH0K;Ke6XqwDp>JU%c1W+AA2o0L2;l|PdP<& ztIMvtXeOo_Uhn7`psRsA(oYbK2r5$kMch89>7Ip;z^7G)Q}3R$FGK6x!t8I{^m~90 z*M#e6>Yd){^RzRW>*)suJt(Q=q}lSjK6%`nKbX8xQFUSi;F9-=b@0_A75&8_!Kr(I zP4qU1cBYu$QXQ$cI|_ijVDxP=zsJ-%g(|{yEwgM~k=*4i(tj2uL8V1ZY}F9w+OGJ} zA{QJhIyK+>c_8ya%qt!Q{p3*)s$WAo5zJ?j5A#aQzc6t;YhRMkRngz}VzR*PjNYhQ zZ_ddiLH)2c1u`E&1x%Cb~f;OX=MO;1fZJXdJiq%;EZ;uyEDSvDLiTmw5uT5+6EtpB~i zM*8>P%A|ER0f%JC&Q!Iw6}ZTqlg0*8AhIcF7QCPEn)PA#i)1*hlx5!jW+tz7@a@vo zMkZx}M${%_OX<%(4WW_q2EKA(+rf|Y`3r%e@uaDvURZdVet8X&`{LbkHywVBo@IE0 z*_?z3o<}0~$IA4XoRiz}uv_zLQ860xNa0bz<|W3ZMP-?Dg9LBRu387w*ip7iPEslX z5%hZSRQj+4Spsg2jlJ*Mol38P!MN|&S9P{j*^`-qgCb;;O+wOgm3u0szpP-yQ^)D0 z?~HB~hOjvIQd5F%mhw0+1rudhKcve_OM){wp1X6mf zjNLpLNOVp~>e=1ADzBTkEYIuCaL!fQ+~gtsFx05h7@Y!^sSgJj4$^chgIsVkQUK*Q z_5Japw0CZT{Kss9zdBr9ezFZ{Gun>k=f!Gczb-3{3y(!SlVZrNF4Tm+Qd(Dj#q*IJ zmq%9IeVm9M#$M!-46MKO4=1C~al{B?@Q{!^U2x4)*U~m-4;boh)@R6MOE4X(EwjdT z(SOCWQ^%&71*uax39S^XB7Fpbp3s!w)+X=V5o=J8>fu{2lmrf313|Zf4bv>MkX5^< zfOCS{SZS@Isfrh0UriM4S4Fra|8(EIAmBWt;JSFO`klK2I9WUVFB;%-76AVf2lbUR#Ymyi81|oO-j>9*>uj zzQ{fQC3-_uldes#imJsRm)UgO5DVc>QWeC(z`BCVmLJ?PeC zF4+XBipDzGTI@bBcusOa?fNmt?u-5nIol6lp?r!w`VY1TNR4`JvX9}jG)dx;w1ODO zF%)sF(Q^w`(q~sHaMxi z&3U}F#&geMwtv$*AJ5i=Y;J~|P!07(2SP831amB{v^%rwOpQF|YoTFg3maEE`}G9s^iuGE6P;r!julgtfM5cD%R`+Ed9%KJ{{{CmML{76xc0w>2;yNj|JCaZ^vs^w&Yg zUF?Uvw3QlhFKX#_;Z1tPj|YTX61OE$aOx{2-8GgZj`Zi_69m|S^tq5$IwZIzjX51eDGsIkKYz~uRtG{ z@Z7;HW1XBHiK)v%ggi~HGfK~TRmWy|tin#>K(1d@b@=02=m?jUqVI0= zLvSIQfpX6NfN;k3Wpg1005e;#x9h9m?PPF@4nLP_-GY<(Rpq&;sfa^lr|mxC<2|YM zXdXxoY2M(|nO~ZCJHl|ldcf1Nt|6@;F6H~$nWq;R=+#LxpG^OC__H2g1z3~`ho7cp zIt$X3{6SC2viB#v}Li8DsnN#YFjW0Ys*J71+!3AJ^9KZ7Y8KbT8*L>cz4kC zhNqhv?gXu|rh%1^nFO(j06-2SB0X*tj1^KA%i(%?L=6)&Lgph_fudc&1hLMe99ws5 zWct2h8m{8emh$~R0c(cOhE{w6cw~-PJZZHZv}4$IsYQbC^*u>ml+)$I65-~iU+ZSs zY(gCt3X?MKZ%obFU-6$`rl3C?)=!X|H+MgQoJ^8yuH*n9qwz<#=+k2WNV0-4qH#s= z@S&5%y4zirMWvG72#w{0(SB!=^l3}#EL>ZkG>~7;n<%~g-a|)+{K1YDFm2LFkrlqh zG&XlfT7$U`RU&yMeQ>in z)5Q9k{l!d|c{`>5-HZ~>X4L&#LP6v_*NPCxcoCi&WCblm5{c`@wJkKa?s=+5OGKp& zrKH*6Ar=06!(ZUV-(DtPGSz!WU8}sx*~cX&LVcl(voq$<^zY5AGeeE;kjlmW6nhHHi*!X2224MAaARn)h`tz&7< zLkX0VAM_ePs3y&Z89Ln{94_?GOLaIVvyy?A7O0TEo6ML<`lZjMUj4ZL$jeUjrWNNv z-pytd5<=Mhus_J9!U#Lj4!^r+D$zNgZJE#NrOW&uG%XC2n6n_A5=@v6aG6IX0;4lE zlvsbQ%(;Ir02;&VM)NZF>hIL;KF245Pq#oZ+Xd%l47!&mQ>e?wse|Wa%yM}A{a&%Q zai5DhnzaH`02Ir%9sMPE?qbTkaHaN@3!oe0uOuMvK)OETbq(A4jfzI*Pzu@x`|@^u z8q=?*>m?`7VhAx_Y~0ytvB>MNK@R2*QQxX`z%w5GIXWs?h_qDN&NX+G9Cv$B=IU&) z%UK7!MTD3j!$G11c?;b)E8ZE`gNv{ozs((VP*=#fjF{l9oa5_y1%R$O40Kj(L2Ex+nylRLTa9cT$R_*Uo$cy(LcZ3}a%kMMdWaBLg2T?q zt^-=pN|x3(kKK*=n;V*!eR)+VtPGZ7}Zk#fsS!6b}E)QKhA>wYIi z?f8)^>+@T2T71J-o`A;&s4;QpJ)pc@H@mxN4U{WwZgA-~XbrmYwCm~=~z%(T>O5cX7~ z>*kk}4do8nTfO>^{E$u8k~?oh&fT;RU)VsGQJLi_=9lR(p|a!dk#6=9zhUyW3dG5% zM!a}t1a)kt#PCte^%P+;mr?dK`}x$;$KE0IyvDi>Nc{R}&~Nv|1%|?qa66|CHwkYjrq`qCXZ>@Vg zOcN9yO2K=T;be4+DZwIcoTeCC+saF45uOpDn9(vf`lAny_=sW*q1}eQlHin8dK(SILfngKSjd~jk3jr{ z`;^)0zY@F-7#vO)%@>P6jkcDzYCwchTjDE{n{R^pv{$9Lpu1lmH4mMgv@h`rk}|86 z(IsFs1$D-*)q>5SP787pVl`jb)<*HNPagy*J`L zaSmW}K`Uxe4X!p}hO%0k>xX-^GdwwEDo*%1qW)ZD{E@x#IT$;FFok04iF}!LPS)tY z%Ef3esH9khVXVD1sm*H8vUgW{`qAYj)S4Iet**r|y+M z!a<8eXRl_&Y@I948&oeJdoycz_!5scIxaB_+A`Q}S4X5817v9IY}INggz2Uq$fa2T zCz`7pv_iK?255XMHSRRE)aU3}d(Pl~K)%yO{hUl$ z)cSd314CtbZ9lomzl9UoOMgK)pJ~ve9!y^;P>4;uY5tGjZ=zpD?MM=9D6m9#b?4KPwubE8#rgO?Y z$y8UbgUD^Fj!V?In!^&q}~)11mT%=rgBu31O7rpP`i8G2WirIMwf7go&genT{Gegr3KTUeuOk*T2q+ z8@zcR+vXkwVe;DnGi3>avfyGjw>TC)rZ_JUj5yJlz}z$DUnHl|1Qrw+n$ZF`^APEt$A%zivg5! zm^>2Gg19J(e(O_x`{0%8rNxLFF8E*xqs#SUgD&*F0#A_T7XvA3^C9Yjyb4TJDy!Bl zkmsX+D~@(g<)rs8gbHp<6nPhe12>zucX-ytTiwHYXo*Whew6O@WQ zepOf_!|2Ntq&uDNLNzJ$VVC^sHZ12VosA-l+4I?tVF}#rHM(p4v;uf_DTSknSA+OZ zPGch^@Y~fdiK2jYA|Z^F-cMX#$c2GTQkR$AU%i@Ubi+Q2KnD;*dFVdgXTMLnW*b%& zm8)hZbMHW~J|Jh|l-2W@N|t{>_gTp$d~SdS7X~qVqaAQYae3%0zDNvf#?>RH7X0ph zaMMqn{Ku(#3gAtE_}i9sH8i*N(ziz2w7`d0`#hLX?huf-M|6h%d9q(@99VIYRwd@F zz1BhZaiHzH2wjrf>XZ%GgF_$F>o`$%KztF>C9&Ot;7VNdJI%M4EQH?M!&7?x)2l zo2)d*hlihhhj-164Pc|!V@7|A9J%9BbONfHjfa~r=8kPSjLsixI!R)LUUoJGKaV}? zI5Q~PY!{+eHtebfD*M5A!czIq_}Gr_cVczR^8{G{k{|iMbMrJ$6a#(E$6T2UA^z|D zk@Tm*bua;8mFsEE^?e?=1lLB-Fwu(Tc)FEtJFVS6<=Xn-84@5A4n8mPM)dYeF zX7*h17cA1r9_J?SDO=$QqPD5)Y%$6r>Sz{EXPbXU)Hw$jxjjvRB&;pH)Gp7MOu~MW ze_i=kw%eAz<4dp#V?Z|EP6CHXw|2&NnG8!3{@?LXqH!-wX(nxiBe|A~G<0x;yg6v7 zSoQjD7D(tQ8d*ZDJBs6xgrooQ>03vAUndXDUsy$j=QmP(Qv9-xh0DqcX-5hVQ!~mW zGz_-nI()j90q2)!P+Tl!ZCIEvo_6p87iqfPy5_aR4@3qG#XmMpR}qGBT(!2ISa}L- zh7OHn7ZU&@aS)|1fv;BbQIgs6#MRrH^ma$<@BQ)&I*?H z5R!A1KYOd7U~MeF2;PGT8QPzLz1C;QM&43rT?z2`6b-n?Kz9mErz7bR$FRvQB}dfA z3ZQs4ute_h<3C%U04dF_>c`9{P}n0mFV6CP+uFOLXH5Hz2U*W4ybLIIubuS|_fiO& z?H&bfyY7uozh+61FYD7*}2r)X3~-p^jhRZVG6Rk*CldrsRKQsqRQS^Out>k$!(qC2|L@Dv z=U>{JehMgjXhYob>{E$+*o?g|=aTypf6g|?ZLViV>peV5e5x=teN=UdJLi|OJXV^C zhF_tMN890X@lx|GHkqK^mESTi9<0D`=hLzJ?Gpk!yQIk7k5iR9`utc94LdFfxuuBw zj4`{a_z#zsiEo3)-}3_x`*$>#f+_7;5?2@vtuebmyw*~ys}^Xz52nCrekw*=%5SJu z5)?NjdZ}`tU?Qm(MH`F(+opX@E~-FDr_Aeull*Y7=+SF^;La@0N!7EgzIEF2v~-DL zW40Ua2EHd%n|_!@=#)i&R1qHA4O+DCJj)axsSJL`yRd}}lQ+Iu>ibsz?Cn?T!t)8B z3J-`iWL06>H6gQ_{SDAOLHEUP4HK|OMniV$h`kJur3Ui)=trfZY>z`47^TP>o}1`# zwUtM37ZY#qF@GaX<+=m57VypZcF!_XHdb*&MyC&G&14Z#P60 zCDLE=YMxVfuG5y%f>IiME*5j*x>V{A^d7Rr)5Y#h_iL!$c0Q!GmY?O>|CWf3le`ot z5Rb@IcBjk&dzGS471*e^^WhCFiE83!>D_+r3P}X!RagfOw`k<*wPi@o0~3Q)-i2z- zeCq-^Ma(L&_wVBhGmuVcD)IIHlvCAG)PfmA;jf>kl&~c7tvz+RKxgji7mkxV-RiUE z9v!;xEWpS-2aF;A=?QceCB*w9xnYu00oQ$2OT#{lMCaFK<6aZQD<4YgrDgIUJ1QYt zD=E5n0`R-dm!&%d%_=-4*G;f8c!(${?ZuOCf|`GQM6dgY9WX^LCuH)2d5Iol^ApO7 zw*x_QJ}Ei`&6xcRx2r~v{;m z1Aa>1E{d?c&s2^=y3e~qM)@Baw%sj-@m!e7ehtQmj^4M`S>yz6qxU>Eo%+gjZY1{) zZ^#%L>#nnWz)s>mgV=5I_m|q1E9EwNThq=6TG|cWrEYdl9L=ILuiQZiaay$>tG}V* z)ufd6OAm?|hgMHOSIUa*4+2vA;5S^2;^O3#ulheZ9lX-uPt*!!_rvSuA4P_*@KRgf zs0SG3J=yzy&6WoHc-}e#Fr|XZ@%7WvmO;F;?bh-${Z$2=6$e72O;?X+;?=?+XyUgS zJ8+@MHM#9w%W90&?5Dlj6&>^M4>r^?VCacm4-)Rcy5-M=ikyg(vvl^|CMW>K-!RAg zI7VAl9QBQ7DnT~xTTx{<$v=DzClkikJRntg!mZ3ya&blPDtT*SoRxu&Aa_vP7+~;x zbC7xhz4lpPbo(P|o|w`I6W_;pl*Fju+F8ZPE|PinN67~FP<~!bW;-fOHl17b^|JC! z36as6gJ%qkV4#;dPlZA%aqj-CiGL~uDuSv-6o8;MnCIioy z0)ZqmlXy2FFO~qor{-i%k447F=-PcbLTh+dZ6i;N%cJ zEg;jR@LrD@mjg0nvs2Y ze&_G1&O+~9L$*Wvyy*0RqqU_Dg;3RuZCb&s(i2X*Dm^l&0)3-evVCgdStCfdP1m~2 zV`%c<5B$2}6DadU3YG)stFbqA!mX-*?d~p>txJ+&7R`rsP-F@1_ZG6fnZ{|FgYI5t zz}}dDH&u^!W8Zs<-mA1Q&^eQAT73KUpc2Z!ZtVFC*ks#gsb{-n(x{@0aTMHkcyRSJ z*&WHfm%=MkEpYyTpG?=kMN&$|hLux}&D`B?Fw?o6s3!Vv({Aisby`)V18_DM-aPSb8`jcCc(l$Ok#n zxEs2f7{YI+Ed&;ynwom;#nc1Q(YB`($Zg?*j115=+k}-DmP1Pd&9Q78!`yj%lcuT& z^ZrMAVTr3h;)h$=joyR2y>ELpyl#BKNoC+ocaD(DtPq0%&KA#X&Nm}fE7hQ5uN*&? zC?!NqN^_~}?#+m#;+@Edn#9DIhM{eh$5sOX>n-H=-d#e!AcPL7w%G6^CBz>f%5Tv_j=K?ER|H-( zHOf&jt?UlzCI>!Vc>fC8{}I1`J;K6VX_jClykp^e0V*N=LVlu$KCn#5LPOP~nd42M2KDir4 zF3;nc^aUIA&u=|auAieHzhN(zWYoq|lE|h8uaxuS>r3tOi0XoJD$ zVj@Q$4(ic{bHvTxBc1jp1&Q&!^W}) z>d!{UUHIR(aGQtp7n)8odDxU!zW>J4M~Jw7?ug{S)hIA%W$@VxX7rpoQ=BcqWG_&E zu*teDS^Lu1p3U+4thotK{jS?4nqmjqHF@z-+eITHppWQiJj(i4*O2l%jeJCVhLBh4 zX_q?a*j??75AV*bb*MDC_g2FS{}{VwT?PevTgLRGz6_Kpp0sYqKFeWJ?sexWHaWu5 zIzeNWsG81*79oE20Uu|j@Y(uXm5&Q#Rc(4)TJ3Fmc-8#!zu0f6&|QpAPJ&>Q<+SCm!dmba0O1E^Vka6fR~{@;=A%HEJLMZ}wz; z(1lxhvTJ)b0!lZo>epCj=vwZ+zMRacz0WOno?v7%RDDDLGjsk6AIXl>enA^^x$;*?Za9isv#3NeaK_OxQ_^^BT%I0Syxh6C(6;Xn%y8HG{>ObXG3xSXFlr(Z?wC?r92^}79;wBsq?cM zWKs`!o1{A=c^b6cusxcaX!Lyc46Ga0uln8kw8>7!SZ5{b@&F3iZc`RN3hpb8y|Q0N z^oOqBcQyLPsQYJlp;AKUP95pnIfj>;nlvb(`;Ty4%8Uuu+RyJ+Z{ODE-*3?xjoFre z4FDHQxRLV-h~3vXgXf0o#Q)Y*5%qF5G1b1KaVs~bc(476zlS`Y72oFq3foJd3_4*C@^7?1;3zy+V&g|xM z3Y8@J4+Kj;9r%i$>_0)UR{GJG4t!F)GgUYiu)uW28l=9umPeQOPuxWmA)#B-f}-OE zDGr-|VsO@8;k`@pwp6ev+%WYOJW^G7cVY3FgUC>;phuiC!^PR$*g=_Q%c`zwWKtQS z1HR}S{DwAFp& znd@|2oVP#V@1dEI#;9jC*C;ig?QvOS$ zi{4koFKl+KkH6Un?@GmriWoOD1!hg^D`AHQ_jy^pAUTO>631Gt5q~w6Vf>_qZNV2U z5y9$3gI9DJYx55ijd>)!xIo2iD~sW@n_8Gy2z(+ODD%r2OZm5O43|>@B-_;Ya_S9n z8^hZmxy<-DOJP01lKRm0w`yM+Awm(mBDg+9&+=FzUgh0*A;u$JX99g#pnw#KGI}zm za<{>PXku&T(eF>o+m7756FQfpI7s8LX0oGQcK<7hCZ)2+A@w8?uJ$Ct+>l>ln42?2 zy^M_`7IcqnYwG2JSNr#fM(<1J`y-F=_Ne2|le1Iv>qGD1$a`Z=g7@OzT7Av*Sh5tf zl1$^IKXY6}0Pmj0ey?)aE@Ag775wcsI1uYYm*6b~H3bo%_Zuv>p6R_0oBzj8_{59( zR!-bVkH6r+Kv8-F+aKL_(D^)GIUzl5?p@mvAx3D6{yy022I0fV!Z6I|;^EFB9FD4F zy?fI7m#C`SE}?d9Q!h0{G87(Eo7Fe?RcQPvL*IkN-DK=zq=TAO7fnJKMD*xA2{yho*YxDnJU;1yj;{R1w`RfGV_vTF4yn3uL7?THV zJHdMKTlK}^5o=~}9}8F@nzxqW4q^;gRm9n@(S`3-D0!k9HadwW)ELX!Q2nZrk2<8yR#|vn zMbjfZSB5Iq#b|>uWk;di+8tGMWJ?$dtYBZ&Jxq4-0_6rX@#s42jN%S4X;~-KHRa{u zqM{!%TxtabdaqR@n#4?wA7XJnm7cXHkQWd9Pb5GUg3okl&s2%UJq{`iB|a}&MKDQ)>jzUF@>%;%kY6$9eJX;qkpMbCY$N5dGs)n`CE zoD;~u-18Y@p!eQX>P`t`D3RAwWx=V0rAh?`s!L|IXeJ&k{OE@kVf^THGF4!X+PjSy zn8_Qx)Nv|7dxn7JzKWx~t}Aun1y8$VpR%@Cn(vGRAN0%9r+tJUzX89z?jTUFy;~$0 z-1zsifZ=%6&oD+!|Kl1@jo4qzwq-B2Z$HGBwkzWusfTJNf9J<|6|r6OW{Y;iAdBC3 zyxGA~41@l7baWxLLGcRb3oZ&&5>~n{68FEJ+h{ceU}|ukiG~i82)~1y_Zqhw5iLOi zEFV=eAtS{38SbFti|{T0z|f&aF6JknZA&})JKh0jP0$4jxc{(eSIE_4f}kainAVPG zo3MFAEZ;$7ZIb`Jzg4=sv!2D7_5`wPWxZ;koakowAk)(irEeA7a4TljWF{b596?9T zVg+AX>Ub;vOF4s~FMIxYbJC`q)U%GxV?cbeY{WRg0p>pA53kr_p{*Ho5HC_Q)1m^3 zAFXPd=ct6DDs}YyXGhBqW(66w^V+mmzA%DS7cr?|hfPuBYsLDNSfD-Cb@#I@8`y_Z zOE5#nck%XrS?u>~$SwI8ItQkfuqP!awEpbrf6`_;3?m1~kL{gsH&adP*(8s?BfIaj zTkE3taVexTm`sHVb6iRf=f6Fb0n!b=?b)xG7oDF0ztCAn;`~h~K=h9G+Jk!z)xw|p`2I;q}@(imUkwQd*%^6l!QqD zs#noM8Kte>PtjAlMl=*gjs@;SjKd^S*WXryQb5}pnlNwRysX*Br|K+faU4$cCPxJ@ zlc$FaCX>bJ;t&s7+h(uju~zghwL9E0e>-DYmJKT>vSJrm-u9jDta>Ips(Qlt@9lkv zn;6O-$lZFuEPhjN)%lw6|D)=?J{eO>Q$D_oF@ zoV7QE9b$BQHbLigek`nCT?ZASWF}hWn7;x4@5=k}^Dz%uI9{}2pcx0H1}kI+VSe_i z(ln>Dxg6eSYW%BzJ{{aF95n9E{Q5*P6(?W+@WRW1-6E}H*c*|iGsj*-DN+eX&=Wow z*Wlm(MkM?L-&(w^&^VWUi!p3I|7eaSek>fp1j-le#`Of$ zif_dejW_$upPc(qXWl2@S5w3OoS}WVdM``h&)3Wk-3UU9?r+B!XW`MXci(kVxmLoD z>cW3~kb3%`3e280+PU%+7|;KI^&$Vczt=ktpy0P#*y)b@E-23XlZ0w-Hn-3!WRKnn z2xN2>px*0>*vlfqCi+z09rXWQN2T99v=!AF1!&3`OTCt*r@1}mTCk?ku)E096+}oP z&Ns8y8gI4FB}6RcqsqQjLH2n!-Tg+dnhog8M`?i}BMqYzx6e=1PpTlwtCF1IRI*FS z_s)}5y_)71z*zbql;MZuVfgE7jkNZ&j(g7XecagwhOg4?8vl zKFJd0JEP5=OIT;EfTjFzp1U}DgvE{n;P&odMF}6ngmC6CulHgek5lJ}`)h1kE7;%V z=LOg3_+sENMtgsce;PwZMrr7ZJSNT8Gw^FoG;YO|K;!keErhIhf{Au2W^k>U2BSm> z?Zdn_Lh)Kbr{UD^_Fj5XmdFqO3G^k>i3uxytUN%Lx)r0_v5^CkZPbNr^z83RV;Lh8 z_qi(>^reDyg+57LmoEt9PP|^H1jAZt1h2{YQqk$emSP^4tPUGvDA(7 z)y6iLT3qk=Q%>nCR);uP3bJ@p5$}M(?lxF?7GiJYY=um>izbk@8lYfe*{vF*Zbj$z zm1t9Q{U*&!>k(n+M7~TQS*xSR+x~V#%y8F2hcm&iS!^gU`gZ53uvAD=%blsmS=-zd zaqYeJ9JkRq{P(^LkL98Sm1du;P2`d`z{H+-eub|?5>T0S8+sqL>rS4T(1%mX2Rp9J zZbl+SYpg>?R|RO#W^=fO@bB(ip6$E9_JSdvwSOz*T9K#olAS(A zjpmP!h^e(y0m@P~%>W=k%4~7g=y%DC^dkZnB zJ;$K-m1B{`N^OfLBp6+UQA>>+mMs1}oy|MHE%snsX7#rrV-4Tv&g$^_24ZNT(o>CX z&r8n501d(ybCJvCuH;lI#tqj+dBV^w0h(FhL=m`E)gf+}(2Bc+{T3=^Ux}N3r4oY$ zkv~!QJ#bgp8hFQRo~x-!=0MRz^Y$uI<0{lGjDoP%b1f-G9FGYgG~YnvqmZmyj=iHv z+DUqqORm1UdK~&yCXrS)4wM_>Ru>~>WhGIFZ5U7SWZn8s$nZ?ac7_J~v&YT;9NZr*O^|DrJXUXMNL9KKRP-PXWBP_XT}eTJ$) zv1`B>+BHt&3S@}(hJQ%Z|Beg-`Qv^>En=gqz}McV#*x+AVhep6GvNed)C@b4`m~s& z(c1BgE7bvVKmql+^;$e|UlVW3GUS4BvkLWVm~jtwZC#F5O*&L@y;5$RTf4PEom?Su z1T|AA*I^>ZfxZ+buCR3UK~zmZW5@VP4TtO+C)&^lKP$at9u}*%Jfxn0sIug<-4EqG1)X4=HeT+Cf>XiP%3uij;Z!*X(7RoN@? zqc4>GW{$X(InvfrK`4W%j4iymIh)qRL{9vD7~q|(!fKXz@gnNb;;`fP0%4Tvm?9mk z!IktSBbbPhCmZ&s&a2MP%Vc=khvgUz?JUABty3C$v?l|2gA=sd*}ZqanN!=>GOB!1 z+YF>S8En2x))54L)&zek?Jn}Sc9L`TYvl9~_X%4&n$PAkfyVme_M?2tv^~L|xaUho zDkhIy+$TmRs3|N@e@kuT*8b&97 zH!^_5TA4}x?&jOob<4%omMmp&UW`iIY=sIPuLrPy^F{?y?QDNz&8GD!g2#NK4nb4|soC*Y*EOs6 zTYH7BvE_Wt1L{PL*f$gvM3_@rq^D-VguTg{c-PQ^%~CqWc*(vNsf1fi>l;#?3ibBO zvT7=5xP!OJSWXPf!{v2~fwCNDjcS`V0P?^Uaq-*S#^GbDO|(BfcOpBf&`HZ#o)9h* zmaQ%i;4n&Mp{ARUFE)b{*6Nwut=&@0lpm?4QKViqiq~jjw#dWV_I3-71m`V};1O8g zw!L_%T2PAilR=dXCN2wB>lkLWW=Z=gBZ&%ngbMO3 zuk;KGRci*6HRl`VaO7#u103&tWRsC^G@=Su46qn3uiPj`7B^FSwLzFoogh{Q!fz1Y z^&T}}@UoH-J(GR=mjslQ3CY^?Q<9X;R4${FUoe+}^MotSj^(Nf@%EbrOFoa5(3Ik+ zrY7ls3nZLHJ9TC1Qq(QGL`C7W^;yP(kIlZG)p+LKZcxROi38dlBy%*5> z%Z?!!CoY=ge) z=~lFlIMZXsjiI~E?lmEVcl;h6!dG`HTa(GAs#gzA0J*dREiqEQPCeF-tIW%Jf0I|P1TkNbxh!OZQ?DrmqNua%O{=@YQC6 zF*tr|LM!fY)-Os)z=Rf4&=0#-d<(Cs#mV92XDWOToSdgWsr24xZUlot&?0xM$9PGS ze}r%A{o||Fu6m6RlWhIk=Td{@CF=HnNt7T}sX-Hd5@mH8%Ymw}nB(-$H~!3}fA|3} zFhX{6q}g{b`(Q)17{xZnY*o@rqlUxs0xde_M@4(=(7OmujkVM=gYk=&BDARnvveKZ z|H(!+jV><{M(r?uSvKzT!a68LV_%F8@?goW+get3uDllX^OI=kr*F~4{-rypfHfD2 zPqM%3>la5kOkgN#b%UVrR8jR*!;RG{+UqWS6kt^}i8b;0n`vz+X&SM|Cmqhua{kh` zMBj(sGQ8VppF2~L<_(m+W={C2}d9nDgt=NFUG%LKzR17e2}IjX#v z0?-%)({v6bo%PaT>=_5rdZQroY`=OVc*EyuFGh=!h5ls^#{3*?SSb;X)mZo$2qKWJ z?c}vYuHyUtJ-#M-u!_wknxI6p#3%hCd=ne)oE&)Ifc^At$h3Zg@xI-5T)TDxNoi3h zt$H$9l|841PCm!4IL@|2R4{5aMyk#&uvRj@Ig_g6VECfyWt91DwWD8k|3fXOxdFd%Wpg@it&*!h14f}S~44nPfl}pIE+g&ty z6>{|y)3;`RVftsQ;1iOnxwQQ~uU1uJDEd?B>}r!El;{~zL2icU*@;~28rqnh*1LS# ze~^UjX}XLUo_2}q8<$l6!KY17ga%b}?_ zi4h0a8aF-vCiTU4vvGXlzIDw}uoE%EWNRnnlU|BwL?f-fyhCt*_Hz_~hi`okX33ny zwthAMp!Tddn0+K%b&{;FC0AGP|EurH$y9!l07BDpvPk>ok8o;-!ib4wlgjlZI9NqXVT8xp!~JvewIl^iVP&~tMPf)U-j?#t z+(c6S*r!KX1Z#0SK@evhvOvX5Uhs$;Z+G+2%)34Jb*21C<}M9f%K;2+7nci*MoM$+ z7&;0Pzko41da9N~b&2y92zN$m-_D&31Slx7Y}?Vj!^V;&F!(dimAIS;XLu_#?0lZ2 zKOxX#nsJ7gyJAyg>~p9lIkWfqrD@AOo#Wo6XICQN-W?uVA%TKOWOk}bXmFx0JIiAx z@TZHOQ~)m;x~ucDU%}tY71v@JqJpbqsj<$xf0h$R<_P+7T&=Ss&El5xSubb0K1M!h8g1)! zPj>RLO7NZ}rk_Alza8s0;P`%&8BS5xRNh%qd!jCr8a={x&pwk!t_S56gBXUzf(HOx zVD_VgDmlGrc800UcDG!))LW&~EO#aUK-cpj@NbjX;-*0Z{-Z?&!}WdVWO^QtAt}D9dHhE|M?koj z5I(e*^nVP5#BJ{viF))rh`U_#)>vzk-mp3tb(N{RRrd%HxbYX4=sTMESFD9e8(kT+0?PpRuc3s_LfQx860XEbq>3Bs$nT<{b$a8GB)9d zLqedLT;Y@B_9yw{35jVqXV)tIhRqOxu%_=@sTD#o+@gMWu1s+Z&7zd=gZP>YQjmT{I&7~aLv>x@% zJ~_d|j&H%XdVhBcK?udtR!+X9bHrMEu$X|x*4H_0w)9%y=?Dq`zzrMskXf}qal>X| zQL=kT0VZ($Y26x2e5kSt@6;GE1)cc&JDVjAZo77|IWFP4bw{OQ;H$Zm96`a&txF9j zJjM$v9r-`Hg|gg^k}P@8`(7!j2;re2TTPq&8YbCEJ89nK20gVTgQK|Ni|GR`2|Um- zVQ;vN`X`$vWZ7?wIH7oG~xhHxuuif-2=cF zrd(CX3RW7yzx8YWTfV$o+}c6qF9vm2)ajIWiSkY0jI|lj!&Fnf z%Spon3uDQq@Y;Ei^rL*(z$CUoIrdyE4e?yIgv7gVX5N9n-y`f*@Xr3(5!QlS)==SOTr0Pafxrc)S6K~#PH?%7Gu?(Xph^m2O{-!qg>%I%IzGbZ2 zmJ{maeN(mW+2WyGfnA;e{GbqZ#xa9%F+fF`?9=$LX58QlV0S0ZXWvLKCzSUp<#hcO zotCqDm(eoU`B5rUZaH(vH-*K7SpEyVEtY`vCbF*VepL_SHEvpQKAbdmpApAOh*GG0 zxpO^#0P#K3D&1=#4Xt3X3}k1PT5mbDwEvy#p!v9|n)Pcr0-mE;Nm*+=mMB@Y22{O# z#Z_;$5Ff%2K5V`MJ4AcLwl4zRRkb@`|NE*zE(9e~yC>h#1lN3XA7;Ukjl!`i5cnd; zom6T;@v2WH{k_Xa8w_%j4F1e*mYB-^J@3R#>6$Ij=Gb|coxO`su3oV_;xQ{txQ`)) zBCDh! zQ|)Ig-lPi8O@7Iif4il2+x-a112eF_|63>Sgte2h#RtDVWm2Twtr^wRzHr_(a4StX z*naP3bfxEH_p3BvFZcoxDnicoR-uW+e6NI64>Y$ffwF|NlRjU+*o|8jzxjPi z)WE84d3j}UL)w6NwBC|e6KqgSE%0P{{P!SJ)*>vaDXR!vayKb;W&ot__pita<%#!# z<&~v%D&o=vmC1Vd)pN3)qq-aZC?-Eof}Vft)_*g0k59A- z+{pju9}BZoS*aqGaxbjD;;vkci~obd4*Q~jI}G>AS<89-#NwiJLmEZL@M%13P2IEz zMDHK}C`4LPeWF!jVFSzAS#73h7fBSI*pez6F(UK)U}1@q&S9-Fo_n)%#w-$A&2n<) z`J1iGsmBt9?e?<1J!$ZqlaB(R6vvw4BY93EB%<=#)e<`EpSJ=}sE@;w#LBhp{PMWU z-nk2tdwmOMi|%8fX1+<5fD7UO`kE0zK^wyw;7VfB5-5t_Sk+dDUZpylJC zgtSp7UnY+om{c%9ZqYlHch`6*N|`@x6TJ~NtgGUc{TmqLoaCB=Ohw?GJyVt4$3j%h zV!{?|DS3(d+$ z^X;o{mbN`HB;niYTkF@%F))UU%1ur`GI_Ytqc1E+U*BA@a6Qj!AiC&Z%Z#;ViW5A; zDNkXjS=#=1XSl$U29e$gNjVN~Vgz~VxhOr~i$7K@TD5D3+^w9SNH#d`l=U86H=hWt zXF?t*LMn?kt&n|}b+7CsB>HfbZKd6w5!lQGI2OoO&D=2@?s3w!G@J8qQT zhzwvklasvc3XXm~W9yb%OrAs`J8YFlLg9JW)UcjbdG-4W?Q#I$^ReZCi@qzv*5|978hu(_ekgUWc(=BV;1Cgq7o@3vrSt&@KVn zhf0)I`oQenHN8VoZH+zSax7|Gl>36lffWW2L5nt&DpKPfw9!fG=sw|ol* zvCrA%L2Wa+=wJ^cheBJ|0^7s|5#yYLVlJ-j5O;oGMZ-2>;V5VnZ*XG^IQ@;;X+OF} z&CZ_fjxH@D9!zdV1g>r3c7Ye?OQ~d~Uj2pMV&9#0nF-68E!2K!;sRd`06Q>nTN}5i zt+2(k8dn0j1%+mzKPM^wxqkTnsVaY)7~GS)b^GX%;oF$ly38lm&uWHUfxi=R$TuzU zXAk&@^us}XF*mi(o%9HwUx<*bV<{FPcG7x3ztqGp-gHRcdKzp>wf5>mv8hLI6otPL z&xCpTlK{3Fe_Ad~Tb)Jk~k>{k4BTi}CHqhn~K*DOKb?Y7)4Zp#_dc z?{%fWV+2L5A#9_(2(L;#ADAa(VMs{5=`d6QXYNVXA0?;k0e{^#j&X=6_+dYpSO$AU z$+vGfpIMqj=?6w>G!J6qh5L6lN}jZo(4(v)-(7npM98l#@>#-3>0y^1tI0I-s!K4X z$j-W367aCcc+f@-ucdn3J$IA8@3ET4?n4MMj_xpRmC|`R>?*@L;a}oW1CjTuVx!r0 z=~JtOcbB_?q#Zsjg0Su*p~`J4SF`ylJ7jHT{kk*7aJ?sMqo&!Oybylb>C@M6H{E<; z$g6mANLjn~Hq}loNm*jvGZW~y1}Sri-|c)NU9f%e?r{=jDI{Sn+1cN6dA~jRRLRGV zp9<^!BbFk&#lBM$v{M?%4iqa<25!c{l z?6Rh_s?r2l-X3Vblba3g)o`dm+{g33)Ct95o_2$i{|iFO)k6=jp;cG=aOO!v)D_Cf_MnQYIEFcYhUp zV_neg02~=;aaO!Idm$pY`i+)xd9%bi?4gK{riIDxDML2k--h<0!ZouUyIkMx;uF~P zC$m;bLq6=~o|oI1&5c=f80JN^Wex32jZW_5rbvM8o|5SEr6I4^&_u2+VY{F=_}7Iw zM+E0-d8mbNYlI%L?4q=|;tPc~adDl<2$5g1WH-v)y$i-sWp2g!d2xrC|8+TqIq~$~ zE4fWU_r`uGfIA;Z6x5gR{iCXp`deb>$*G0mOJ1(BsS&v z6(^g**A=%1=5mg8IsIJZtTWT5bwagLP!LlgYThx|7#GN?WHH3kph+6RK^ zbQs>`^NUXVc>1Bul6zfJu4%$0vW7T#!Q&U@;iB88b*+JScl1PqLD1hGL&gupW`}#r z;6grKajV8sfW3jTi4yv*mC*gZ3(h=_rFEjwH^0mxl#^S58e@`YdUS)XcAAhpA3TnF z6e;p07`QUUrge{4GbR%e1$@%>fw_Uq4wFT99ARP;5`0z_?96QTJ;!_$BYM41p-Zu} zIAL?I7UWcz=6LUK4QY3pa41DoiLB! zVVL%$2fNgLIRZ=T()LbrmOJX%ZLS+7f1bR_Bghwko^Nl>uF$T!)RK48rDYPGn*}y~ z8)*iVnB-8@*Fu1|zwbivE6Pz4{xqz)Lwuzm(zz+z202+kj}gO{zxekbqSn^B%eeip z^&N95Nmh`)e;FuLa#V3NrFZ$O&wg*5PR&2pynR*|JNYVSak$iNXTO`qw(|Dc0!Zr@ z!aQUC|BPR2L@1ZNxh78=5qGtuYV08GgXl$Z3C?pweco5f?2AoBR|741Z$@axYO>Vg zN*0>E#I-fMg9vJw5My$+PuR2w+Aib;Wk1nckdXBaThcHdO;9BwarKs&1I6@&@UH(N zAb2mH^9bV1z-_Ojs^X>n#h{V%1e7EZV9x`KMOG$seB!DU6GG z3kh8OI;==985}M{ZN=L;pKj=rZ1j3H*A|vFD1uM!V~jKoJgZ&A?G~VqJamsCr#{yh zYzoo*j&W?e_>%Z;;7=$O{bH*>y18%Aj^7JwD&b4pvlO~}PJcLki@^({O{2*^ z-CUzmrVMO0+JsDMJVl$kOGGB!R<1;qLut}m%QccnXjrW=gS!g{ytr{LAI2ua0gpdD zQ^Z#n8;{Jy2Ooo1su>9Rcdx+q4&1oe!(%IU7LsrcS<1m-XN;emKZ_o%Y7BZ5^cos} ztHR5kvI#LoxGhy|H;Csijq%PwxpmYAFV~&*f6jQasHSM*z_cl)CYx-RkF2OhLw)~q zz-_OXt28@_#{ma_j*`#C&Wkw3;}jLf9-AJ)WWYZvdXh|05Fle*`0=5MnF7^W*?5_0ZARs0D)uggjBnuTV~r4z9w zUEI3TMVe5_*xcIKOZ&7lBy{h$qcs1{$GF&IzaZEr&WRH1B7Do(;wC5= z4q&L#7I?IX$#K!3ky;?i*kxD|L-w~`c4`5Gm97M=f78pQ`F?e^Hu?m@bE^MrB;E*U zn(0O(Dj+n3-4u^)`sirh?%U6R@+xu4d9A><{o?!yHXTxp~;!^y4HAaeuTd##<$5ks558&qb$RN*E9$X zW~Z<(D;|H?c9nRI+57eI(-D;JYBfI)ZDz3+8NFW?a;WDjQx@dJd!LaPJi{zleU~MA zDoHqaVofmI9_$|<7l&iQrj~>DotUMC36Mwie4#O>12tl=F(%}zVO99~{1Q5Bd{|GU zo7Klmx|ypJCmxXJQQZmN>uR9IAT=ED>Rf-%<2L zNUZH<5-F=8PSv|x&0qK1i>B`E{=%5se)C)&ps6Aa6Q)Y3aaQXbp!dFI@=in7Ucb2X zV&*0LN{zeR~ zxU&^=x1&4$?w8N?QV@WUG}Tu7*F$FA~z%fw|-2y@Zv!D{guPUV53@}LG0BU_!t7? z_?e9biaG@Qt2Hw}cA?Ep?%pnGTDV9p`G{{71YXPv@p@TDdwpEz{ITINvA|a4Z6EBFeIVVmfz>Ft%9Ue9V%1iES1_#Xudj_dfH=3@e)7A}w4j0qY!Xl2C*0CVYm%&; zd|vf(_-w?;JNj=O!HCjY;uV-x+dY7w=1#>KJVAf8X?^ndt+l$@_iASo+oWL^_8o@z zH5+}QG3cdM8`qdbP3Pcm;SVr=j^-Nae$AFUxX!G{$bB|JwBmB>^}2Z%B#p5gHUcq$~KRl zs@pE^8M4+kS!{X#?b|ut|IA~=Abw637R!o>^a{^yD z-VJ{Fp*YujY9_b>CiRv*S~( zJnY>eLoMDJH@PI0@@(%GPmvN)7oboiJ=bbqj>`zyT1J8)_o=4 zRi-b>m~@7Fx%~M1u#hTywMp+l1Ugd31eR!9i5JyYM$=&|E}V913Y2kWlm`{0?%z>b zC4%-1>NrXFi?qD0K`F?F_1_$50iI*zuy`iP*`F{s-~?Y4*`9PA-GVf2caKww^3(2U z@(lS!>9Kx$P~k*NXMpO^Nc&N>>EWS%!|Ye>?uWg`fB5XJYHaQCsHxMgaauvo=Sl1} z!3pZEYBD4UHPj+2SN;W(wil~W_aNhx@mdp{JYvafwu52j4~Ov+&u04))ToX4EEkMp zkp5Z}Pn**3t&6KMSoF7+j)^BZeaTGMDXST`U zLg2#vRBjTnuaR~8BdHv-HtrH@Lo+(de#-kEoP2OnC+It$Us55e*ZVm;I)Wu8$_EnZ ze@!&g#=6*>GX{VH$j$-(E15lK0tM=wAy6`JgxSu0CX2f0Ysg+qw$Q z_)6ij(>?cjApa*NUmru+uqvsO0<&X0ZoogsHM00W!?uy&0XkB7suD+ST&5R}L&t)L zV{l zq7{`%!+Y_|tHMdnDqAb?pwF7ps8=Cs?8(g;n=*2Xf$bw6%$v#vBe;QW(GS6hG$W z+dkD#MtH3c6~U=n#h~2F_~r@JVH==&aNahj=Gf5WhQ4Yb#g=1;cAXEraSq{5BNv7Q|^P;k~=%)rIgxX0KJy=mg>u zMSU_}#>ICxqpgo7 zPu-v1%q1hfP~t2=_BORG+IXk!R6(7!JnA>zeo&4MdXYkR0YU9jGh@qme}Z+XoUkfN zGAii0?v)qaYUn}iB0KwQrBP#qk96W~g-RNVt?0(-zYQ)}bfW%a0-(C}C>8_zvcjB$ zjvEXJ0BF|RP8aXr?2X-kjFcjiiFd7QjJdeZ%OQMH0Q4|qTMugM5zB}w$`o4#P2d!Ua8%n%R|UZP5<$*+|F;;Q z`TK6$#qq#Hrt9(fC~QOL3>A7GxVBIQDG`wo4D)sX{74kD zM>1ee`Q*;F3FMhpz^^M21@oicZ)Fk$5ggO zwj)XV#rm7Oc&&*^Cie<}38^Bj4sc2*okLr14p(FwFW)HtxcaPlJ~;>L7q(cpVbr4Qz+F%nyftndt6 zZybVr`Aqne=b3<{O9VV?Z0WfLNJw(u`&ZzOEsKqR^fKrWqq$ z*=e03-CVhieQM$9(;;{@>Ro!JxOJ^G`6gt_c9yflgIBA{6UaQFIO#s=VG>>_q@1k# z2qYlqSn2+?*Uc~<#pN-e91SDEym*o8!-{EJ4A4c8bbMrY{I2c$Qj-a*6Ig(ln$I9$ zZL6km*rDl}l*YHY&D=W5ycjIMx-v9N>n%pAB16(i`MDm5pwFoe4_ICT49A4bKm(ZCNIgRS=$eEB?wR5+6)-++=(}}R~xzY zJm4j>2$KU(G+@V;yc}guvyYJkfxFt3C$Q*R2siq9B0BpOb@QJKA+KwIA$g0d5Q3|f zNaY%W>zOa<4l&E&sOFX8(_s+Sdhupx!$$)}&PXNWExr}T?At)6tsz&xYMKs1pI#ix zK1Dc#5(1TSi=X0!F=kih zwBIp1?dJmf7ez_t1W6tTBF&Ht{+HbgI(#bf0?2_kf_mKInA>o(D;dh=Se6~j$gJ> )mpPgObK&y1N>?f5thm z=3rMC#rJJV>;-o*=d1h&9|RfXtnTD4?{tiwjoHQ**8;UyAFW8=mbKR|uANT0TzYe& z^g{yK?Ir#1!$X=gqI)l}&taVtU4F%dX-OR)HXQ0adhq6rK-KXB)ZDNSfLb07i^ zKx{vA;Gf^Rj)J=df;m5i_<4D7>CsO^v&U^{h*$gVqRzu7&{c9voPOWa>y-($CtX*l zL54W%gb9!9(JBJxKCHm2I{4kNPjx}{Y4(nPf9k%t3kGR+#U6VabK$g^{HgT6)HW7p zy|BNSM2*Hsoh*7wJMYtQ`kI&63)}55shBgG+wH;>jw=Gvd@YZVHlBhWTvQED?h~pM zvaBr#3T;@<^e<5ry9Rgf{T3ma`aVa~D)4go%wl+EWSG@ojux>uFT0PiZdG#AfkZS} zcUD>RXO?K&0$-a_J#8g@_1$ND0?UKQxqT9c9RLGUz6zcEpDci-Od|kT6%P0_Y5pkP z8dd~~TxEBGccrClXKf3edKNV7tFJN-q^K*JyVM~)20GPhIMJ!XTIuf1k&YhP`Zi_s z$-Q)rO8XGtq|CXmFx7TD$<4o+nA>St9XRd%m9zfn>{#zclznBg8Ha&@tZ3t}g@FVm)PtSbzWyODAGzKrbd zui%xwSTeDUh99wSHXek}Q%X#m{Px_)0GZ7_H$4Pv^^GKDTdyg0@L%l2_hyasp zetN^>B_yr)9rM5Ia?Wg;`$a|jZ}TU)kA{WuLaKJZAv4od0YLSOWQ!$~rEgeh zJ?{xK@d&IpaU*_y^09cQeO3Ff# zRx6rBY>n(`5cm1~zG{Y?ENB2=U~cguc)~$hFons`p3<(=;gex~%2xaZS)tE01>s5@ zL&M3gOL&*bn|CMd1OvC%WaUazGFl51p)~Bu+WOWI{(8`XIi?J^(4AKGg(npMaYkC^ zGgDnCc(YH1Zxas$Z@^y$ib{JA)w$!v;@%R^9A|xW%#>{GxAT6dD`G3d0-Ce-^)5{*1*^U8nLXP~9wyrF9M@RiDnZYnsSU)3Bx)sN zdUszH%?kL<2=rHUFMH?h-WyRRy?FaIKn}$2DY;_27cTROPhUI}@3-(jf;f(08KXP? z<`6q8Uj22oDXx`wAbsmgg(SoI8&l)rpD zvvrqee*C-q;B!3?o6dzAR5|lRWrh7OMr>mw#KMqk^M?Tf2P%%4>uZ6A_=wTNneh76 zI|l8n7h7kb^5JSl=_A5=5-^*wSDQxOZz3BJGcb$}{wg8LNR@bNfvww;5zwESrL{WL ziJF#Y^JG2yj=GdXoqU6nidaYo>9c`+a#nb0pwqP^PW?ifVw=RXD{D4_pv)=j0mEK? zoh*UbSjAXQ2^?`?<@53`ORuyKDf)Z%F03bf$(3_X8s8$1Azi*HScF)F_y5|go?kNS zDfv_@sBRCQ53U6<#XzS39ec0(AfpjibeszL8ms!(t-g}O_H};Ujr6BjpF5r@(KKF+ zeg_NLKE=Dkyy}=F_K>0!0eK2J6aTBmd$i3jwaV1}yMAU#18AT<*|=p0CI1}hxC}OU z?>E)(zVPn#yXC*e0Zn*~)<(mu*r>11CF*Lik9&*JCFhfDp3`rcGuNg$g4!nzD@df| zCyalA&k`eCqi?^rEr&B;<2TEi$7NkoZ4^2$dzY-YCA{_Sp7(29JHI^zH>pZm$9XH4vZEQOi^*1g1%#FoUPzIzo`X(q?PTO*#pdFGXwa1Oxf^(GORxa%{ZBNP@YB7edc&(fmbqH`gFrB%}E88-gWBsZn zE%Wxb8KdNPF!<@x3E}!&?gxe)GX+4<7R0NUd*Nz)IG-K>>WW&*^&8mV-|8B1?$;FY zZ>0gd0!$?WG;wSsT*9w@iM2j@m-0A_?Qb|B*4w&yigt%ns#JMbH9`^=sCgK^NSt5Z zEUDT~b0MYBNHEIOnK@$IgGiH zvb&9UAK?JT;5Y5LS90xgBnPWu9#%z8YyO~`dkm)!lv0xEPEV|!wSE%a8aH`kPRhg)jU{m3}}*3X(vZ8K5f2FJHB))G#>H=m1$ z8@bkdTi)%<8loPMqV1SPFEZgQZ}yL<`$v9DdpcL?Vq`&!1=1p`>O0Q%XRJwaEVy{< zV6yqo#;dbhTs__4G-{RTyGw{v%iLp&O>LNX6i8tBkFViqza)9?BJTV9R}?f2QJ#CR zO!tvX;3*3u1E658|Ez!KCCgbL@V`Wy+)KPfrDQ{p(VSlsRe&m`xS7G^VM}N4?75HF zdO0OatOs2$gwJ#pJg``@n}u#i9Lc`;Fj`ppd({FjdiJz;Y1Mt%=&t5buu=Z$25#Cd zYW2`CmKWiMIVZq#4-fpnn5#h)N9McCAYzkL88%&un9lTNOvqeF^E+G^v{d+!!QMaDX5w8CN*@8`8yrlh+CcrlmA z8&W2vy2>>!5t%^-VIXkz2ZI*#d0=Y)`CvKhfzj|`>CvL6PWZ}4zC*i|%!HA!5b}q6 zX#LF4!LEXln`Ye|JP|AOlwIB53DWBd2?g^iauy?1C1)eN%bTw{?!_ZeOx|8T8+-T8 zrJ%L}*vMy4B~PT55%=KZ)hgGc{b6h$Ir=@`Y=iV zxVlUK2MS^`I#eso^TMq1I7N2AUDh|!n=`Y&e8hQT5?@QA3J~3KLxZ%z|5g4}&1}@o z+Ao|_dXkXfv8#R=Kht;=h!^q1W-sog`kGg@Kv&0G@_+FmIk#S`{rLCG^}lLf`3t{{ z(Fd3}<)eqYpS1f7q^5=VZRd5D{@e*@tpbUsrgx%(D~*=>dV!7lMqm2h_ZE*ULW2@R zCJqaz=VropuMx0j{+4cA5@wk%ifK1}93i#trWx}niy)_<-9+VL=e{P3BTf)TO?dHc znB{DCA_5MAsiT4g%V6zY%e)B_=aI-U%n!- zH>Qkbtf<~ILPG4g(CbHM&aY_ z(XJUj(CWJ?QFGTN+B{44aS@_VUB9 zx=2DDwS3iR@^k?IbAg}p@T1k6QRahoZi|(L~BV>b4cPs_mF=ltPyOgCL7SrultBm zWf3A*Zi@{RKAs3gSraH=dBYWa;TmRr?`~r^X4}DC6CXHC z?hcqm6|E%~yYY9HX;2&V%qTknE;&i!G$r4+%6?|44Z%P1mJ~Pj7^6IIV2>7A{bbwV z%e+B*y@(fBkRP?dPBT_V7Pe`av-@;OA&JO++hRtpkjd~FYc>LOOD=SD*DgN?ch5-@ z0TWy_s<7d~(%i%yg70<{!y=enM{I(Q8eU8XmdX|^pvwM__Y6!v|Q3C^fR6C zbFpi-B7V|G=B2MOurF>BS9dy^3wYKueYSZXl7~;MileKi7b40^WhNG!c|>Nf%dYNd z_U#YLkE^Ve%D<{9rziQln9krfUN-H@u>H#Zt8&8(53R2IJ^%j3%FBSW86kugnF zrM*{6BZQi^dPtlX4`~oqIk?RGE21^l9*LKg%p5^{F8G3%PJ7w%7bKFH$;!JnbttOa zwfv#wl$m3ZV~-+U)4URW+XQ~U>R*V+gwI~}taQIqlayB zPT^W7yyc&!Z#nhy^XL|2EA{gv^86-hCJ&oQs5X7y{(;xHOQRB6r-$&`20_a&`J(#r9h~~6T@6lHbn-gqeq#DketPY&|6j01 zH%p`i)N}M(hc=_bL{~#C?4$ovveHSV)@oCuo%IXHWix?MkUneE%-t9Mp#|G7-|$H- z_FhL{8iA6-P+R+R@vN8X0yptkHVIB&O#F5c+U=1O!x6R0LE6q((u(LQzo=kfKP96ltL)2q>s1MU*PNDumEN zAcZ0-HFOdPBmwD!5LyTj0`K;e=bWdU-}!%iuj_otm4waSlbKmFYwmTgdx5}j^yUIT zF1g*)Ds-{78U8K+_2sw~s#V`ufW18LS;jExU29UDoCGLR)R;{IP6{BmIhs2g>aTh~ z5hq=Ym8|?X>WE_`fWwVf4)6Mn)B5(pHlT(@RxAJ8WBcS}=OnLbg*$57e2|Hj+3hl# z1^nsV%Dh8`?ET?8Knah_XU5#B=P^l(1eP|BA@YZ&(FqUDFd&C2m4*}7O)snsND!F) zHxKyNTQ7MalPRSox#PF*FXsXL6^d%(K40h_2kqd?W<2*_X6(SVyqB*)cgtgC95hG|pHF8u>)f@HO832G{%*GFGxc6cMq`ko!vS4$9}+{*s}!eCTxzGCDH)fWBOd*1A?)tcwmA55JAhZyHNM7kO&9%ND;V1YaGqqT*)8XS;G*q!>c0lE3Q(Kt$`% z)?c^TGP!N%@9zE|bqafT9+@j@+bi|whyL3_{M%cbZGhCQCQrK1@!!VtKdS-)KkA+g zAfTeA6Zn5#^jq74_g&r+%--?bF#5L#{zDcFxYci0{p;=R6u^W4(T%OYnb2Qc^sj%_ z-2zO1=`CtkwExe${_`I$djm7NK=B*@Up@Mt7yULw|LgewJZAsv`2S>G|No!y@z3ES zw$?9YnVTB81u7^5@q(*c==D0zA*~wH=a|^^)jYzm=auoy|C7zF82^nKG%)=@$y;W9d>V5z}t4>+xzhWi>br=|6i7o+O!wIq7(8WPHa(H!4JF(Gg3o? zpDpesH(2V5tMBud!hiS5S81=W#+3S zqSohogG_1OB(NoYU)JtYHN1o+7Q_`@<_U%4s?xrh% z!RO5Z*guHdn{(?B^wYDMgQj{zKnz2W^e_Ncxf*9t|1&~C_}4CW4v;rf70=pWuxXon zfod%Ix-UBQq%^%4WbdBMeBr=4Q)`%neAQCvT+dKm{jaoYqFDezc;$@lL*IR+ zcj^`>bOeokG_Bt(gUZaAybU_s5Om;;@U>mTRs+_rcAPk&e!~0|tZ)q!g{ z_xHQRU2pDsT+mpZQ@&5`K&Sm?yW2u5Y`n^$O%uwjOq=gY97`O-IkbJ^sL}jV^Gs6w z>%z=YPft=XK6`QpDrzk4Ycy06Hb(2AIll8h>z0C?o{Z!gER4msJmIbhE&#+v|6tbE zySe(Js;r(j9;5s8zQGC>#}z*05-hgcVZ0f~5$4LETt&r9u*YT@=jKN|K79B7opDe< zaMC726PI3ZMGdOjH~uMbTC^0#@d=WWN%P9YU0M_k(m-#n0%JM`WoAUw;N34nYH;0)XHiXlO6`KV42ZHqjb|vY%{IVexD0HKzk3M|6h1N+d1-5&e=r*C$e`nY z^4IN4Zd`u~v6h;6Jj5Mr70EW9aLpJA@LIY3*^*w`p#eRXRg3k;?z!O5CR^t$6;yon z=7&@J^fyIxffRL5h2q1q3WQXlRiCAXaX5?_|3)V5Kv`=#e^_JV9orDQ31YsykW97# zzrg-OH#94>Z%?O(-fZbQ;l|&6M_T0J!&@duE1|Pi^Sust`@WnOKHK=?73`tk>PP^3 zmfcuBHS7)pYS(5z>@V&2RZ&GdQ}_hOVvRQNXy0V~Vq{xa>9w*?16wV$w*)^G=o>y7 z46RV@zr1J1FNwjfEC=%;P1n{_RYU7Zb#aSyb(Kq7bbr?y|LmMbhpjx!S3-NYQTMfY zKK7xjgEOMRBKl)$cO-rXh6uGoz4A%31?`O;Cli5z4Uj!nJm)O8l&9Xdcp#|RT#xyTwu zU{H`jQpH?S=jM@1k)U|__ps%rthiet_>=8Bf}sna3PL=xXQBMTP>(dfYJphJ`m?1S zN1kut`S+J%hh2Mp^*+T9?!g)Kx%e<=70^6xE^h#VyK?g@_M2ADHWJjeK}|Q3Q81@4njow){ZM>Dd}g z9V5}fB+khQEPBTF_p{O zud5|z2|o3+xWZuxg-QlfsUF5q5|pMHUlGKaBFyDPtnR9hW)3eSoM(rZS@ z_Q@#RF|~lBGw1Y!PITRP>qyHI|R8rz6zi@m(><{Iy99u92W94Gi=qr zIM?-|=gxrUrN6 zgmz+C(H;e1XiB)dv-6vo!75|rr+xX@56vHqVnvP978-21v9ZDIH9_$wP4C$_m|L{V zKt;<@J5Y>t)G39Ddw}MkF?Ikt6WtEx9J`dQ1Co`|`=B7DPj4h5Y8p+gzIOaHQ_zr=Nl73ERhX^Biv-gA zNcamQi4(17pFs|7F0?3=J~bnkNPhIl{-8O!cXeta&F=dG(ypORqAS~}#fIC{b3n8M z3ZQ=GiViiGaPUceG0D_Y+xj9=5=kl!*orSrhZ)BP_u0`;D zpGlfn=8!y>#!8Xvi&jMS2^o|8BbyxBe4W3U#Omq^!XIcRw4D20_h(4j?v&i0))(qfh4EX z`)4T_--)SvLF8)T8qRD$=O*EH$&rxY0Y6QEEuTo)?+N|&wK>l zB%5QFV5>8W3fZl7{4GF$MGl}3=!U`2ZvCa#s?Dpl;{rD6EWg%Hy@+L}Pf1BX-Avgr zjm68!ZW`?d3mg1WZ;W+M5G0^?T%G z@=|=vBYUDZ0jy^<hZ@ruDJ;t_US~DeN@%-*4 zMGwbZM#ML(Gt$yVqSOs<#+`j6a?qPI#07EvEV14oZ`-+VBer%mF!o}6BII~^sY(XB z2T=3VBO*A4fDWeVrX_L;kE*uerQx6rO%f$fo%@=tYF+&kvO3=&m4jo$JQkW{wok~) z0Oj)dmXa({{wP4g`poIO<*(M6D$Fmiz zO0O#WliW@qU+!mk)JrTqgbj0%cHT91EgxAlEF40Yu83H!V$hy|^1%nX)U zz4#Wi+)h7D6nnASoke&0{^towZFh3j2AJ8x#K@$ARp=vT9 zr=gB!pRL(9H48hUEg@X3E_PvBT|J(YzF}Zj&AYT%1tGa@Y%G=84iwE54X0JCk1BLW zb&68GhpbHfDRQt)-qi_a(7FeIoLUFgNK#x4t{6k%c-U9Vl&x9jGxauoIs%BL$m#`< zB~zxs7qlR8XT$yUz5HlDb=zv-?h>fbAM64Luwq!rRzn&?F*t0uTIlw>p;ASUG9fA>;$gr8fh z=fd#*;mwH)YhSh+liJWN2JhT;844zg!#I{ z{0_BKL+>U*i67h58d6V$@-2M~Gga9^D3hsh88~^rT~(o{b@eAycy0JYT)-j%J(g>c zw7nofE7ze$Lyug+Jk%Pcl&O^l!(Dvvgug|#GkaCWs?QP)H<#N4@~Lm-4Y&y`JLrZ9 zN*&dal5{;e)-P(j-qGG#9CeGAxpPGK7Kn*!hx-+s-j7SQ$iiq`czRtMB(+UeRk}MD zEKR}Cgj!QHoI-XVUQZ8PXqug?{v=X~TO3~>O4_x{Tw2VkoY~O?`drvwH+eem)7dOZ ziCO+r_PLz7=ZS7LWH}j)xgnj5>T2sxtNIvsdD*mmWi?tb{>I9kLWFGG1&28O6gAC@ z?Pq(~icjV)gV&W8TMM+FWE!jtfLs&o*1nL2i^C;92((2}G(#*+8kqEFHTp^GcPstF+|Rkih^NH>3TFSPKiJFN z<3ufgeOl7Lcg}x(r6FgtmS203@%o3#Lm4(DApU^-!@e}cQHy)*gUkM|bbhhINrwT~ zOqx70EvJr`bBpEKR^cFXv1r(wSwogNjDh==G)8ds=1Y=jdWzDfvu~Flnw)*4L^MMWck#y5t*B5a3|)ou5a__ z!QgjA*gMn!CfA12YaJx|cEV(cI}PX*MVK%`xJ!#g+woyb*yi1{G~s^K#(V>z>x&B~ zfs_U3V`9n*9^R$D8G?Tyt!1Kx6Y68^<;acn#c}>_>%g@~7=AnWCKtL&dySP+>ke77 zJpb%#f{ldY`xK6c39yDNSmuM$CTSnLW4z7Y@Fl9j5U?_^$I%0eF|QC)&Ov>9bstic)L_X@-v~_K*(3 z+!_9eR}@TI3imRGOuRiav5j5@hEMy;nWKO(WU|OpZO#wkUZUiTJxnRP>#|ms`T1@K z4O!@5;_?V~!%=Ioe`?v@Hd4Wlt+@*4uG#c&pGXmTp|D+Uj#zsgV%iTL`7p3^ z`-NtK=lZ_G^6idm7;)ES;I!8Lcl}SMFO*d-E z*)Z*=NXt~Mx_a6^%_Az=&OOD!`pFwlM=iQc4qByf?&gCX1f`cezo3j2#ePy0X1vQ5 zq&J^m0X5oYHS{OHRxO&~Of?DRWSZ)zDAs?fLf)iTOi7Kqqchm=JF_GOd*`4!=j0ZN z1I!mDC|)=8D*5&m$j=Qc_RSITBJ}tx>szcOx_HCxaP95**V7aC7Nb!4rqW%4J1C|0 zKmk%pxS75xo3-kY%a_IrH7j&Nq~wFPANSaFBUOC#n(aGUE=t0USM}OzUA8spVn*}x zu~cF=3g9%$0ItFLdRODQ1#{2A>jTtMbH@z@5Be!KYWfZK#UIBP$a zl2gChdUPUIN>MXC$gL`OxOXGF;>jJv+Nym_{;k6B*i@Knu9RbPh>JNYy2#Q8d?Z)N zOKF=oR_MGP0Rni-tqIE?KFHs1Zf&0H#m_EXiOMfh^s*K62XjZi-M*#NbimizzvV{M1cf)N&hk=zY4G#bQ=z8<;*LRtJih zhYz+J4sj>af|~EYE4pQ>a}BbYo15FHfoyVT^J{YS<^yKwr(@HSzFgd}AteyWQ6CMF z?_Ie@T7fG0O6$oa+{{jy`xgps6`KF7(}rAyNJpm4T|K7vF(Jk_C%uM zy)k%vxf|(6>c@^1HVql?*@i!Isz6DYf0-#QCOIG{Hb@>lS;d~6y`rj^862Ut`w^+F z*N%L4jb`PU@A@k9=zNGDas_0&gsVZKseCn}UI8%Jxt|>#W9ycKliL%<*S{AGr%>8o z&9(}DUgw?L=!A@93mGjX4RMdH6eio5k1XgSeTqc~lU{uOe7dW=i2y%U1oEgvd4e~W zLT`I6+p90rKR78qsSA0TN%WCZLKWT~-c~bdW+g;N`*S*nQKX3*i*YX8XNytrJr_rN z?B4?sSX;$`pk4-UW_yrN*;)oq+Y4n`_=o{IYQBr=#S>-Un6XRv~2Y*+Kt|kvDNS68HeUK2W54X4tk46?U6czf>_!sPbiR{H%Q9k zh5}W;2DFpQ%YnT$PE^JIgY}|S@^NJly>e?xk#*(V>`C_k7EZ+xrrmA~MV>6y{KV}2 zNy_z9@-2bvpmPF;tEP66&h#X$*)|yLP)3ll`Ib1$ni|Ik&R^w>hqBo5+dw;~aHl*G zl>DvZ>Jf1X_d`t)9}5jWCD(B!&o5@p-oR*3p_yhhobwJJ``&5$t{~0D*5-}+L*$F) zoZBzNFY3YI-a0_1o|n?@WOg_ipX671NU60TDTSsgB4CHD|QkSLaP~ zy}G`=mD+zRynXlY!Lna4z!3P@uao~V&ogsvH-Gu0-N9&OC3AG$Y_*3sXBou6`?F&k zWDL9dH-^Ht*~5K!m=6fv;ECS77rA#Jj<6aLNiso_9yKlHhbk5bfC;swDw>iju$04Z z-b!d{Uo7ve6Ld1j*f}D($W!Ray}`hiZ`>y(T$@S~-y?iUn&C=6^$T8W?H$)bVwDN! z|8AE9z^2rxRr1rxz6(-XQK}!X)?`DLEa7!aLDpI8)~nEp1P(S8)LJmgKR;rGnuTU? zI7YXRlG3eG~w8&Q?G4s8O$0T@$l(AV=qz%!PTl`)Amtr z6h=F*)azuojiJ_UJ0Jx*4Q(03y@TQ48S(<=J$LApVbro@^CZ}&IC3pi{6peui4{f6 zo6LCb10LG5?AL84R_Nq8F+K>xd*nO}bV})PE^*ci-?gruk?~IJ$RluiIx3jow=RVW z@#+XaF(|g_Q&v@?eGcLBW#Du`<576ccZ`Hf##fjE`C*yzz*9b0$Jn&46AHsp;|}4W zA?9qfvfII3yB)OD4GdLtGR-UphjOkhTI2+vi|RZiN-WSFQ^@L-OMTUb^mLe$>E{X# zyec3voBqY-t;hrRVhBWWQu6`xyQrXf!Q#p;ti`pCQotG(T9@*2dV=|Vtw|44mejB6=t*4k!q0{!hlFFiQn?*HV#rPUqQcF9=*#hRzYfT44wzJI zV)3%r;!5X*#GN=xDRz=qjgC@bkRMT>@(%oMZQYqJg2&slSeP!cX$3z^Yp>}+LgCLE zGND#)J9XNuJ0M*L`3&Jg%P@}b&?LZY&Jas_io8EtFKV3_J5{5!haqiYapM^TG0pDq zq9d3?fKIz4pYZYb7GQ`e=b z&UNcJ70cW$Sv0S*e?p3l>CQ}1qmCnkp0kU?;9y6(s*mY@bNdatXDFG|BR-^fy-J{G<)K$Oe5 zHBs)~f7U&y5Ut}WH$uEx!kKgJ&4uMB>Z>euvY-s|2~#3g?H&h6%br@!aZ&Mu@TacT z=kF|?8S&J~K+_W9Z(FAV1&Uy&4kT!KmwvjV%HY4Q>B?xBpAGPzbR11T>i4m%Z7F`j zen`S^%QNB8YeN#UFS?~X>3)P)11LNr{K(J49*%bqYTi1ihc)Xn_37<)Xuok8Nhw*S zFKI%BNTp3wln>VaaUuDoSkJ>d>@PN&=n7xMZPb45+sV4BqaOo4ncKfJ4P9LqPykL% z?#WY7i!Htl^fF?Pi3;7(#Y@pqap|@4wY0{lbytEqS07SF7~i_Bj3#0g*d)NjCfukw zJq#d&yK=L12g9u|9!X~Op6Z=zmOSZS7eO3S301B_8Zf4Z!fb#L%7WxEXV*7;NNX7C zp*&UtA?0RijUL$}t@myxtNSUqFe@z<4BxD`&Y9DC*)1^vh$)`Q`{ZL%-1HXX*%4C0 z!<3N$Gg#+SO(O6PQ30d&$ZG=7%BmZSw+r}P4W6QLiA1pptI%T`x3e4Jr8GDv@!Cd# z1)W;m0NF|!)RMB-adjtS?t4~c-I56^A$pM76GY3WXg7@Bf+RgUR>?nKTNx{~CAqw@ zJ2OY*G!+&b1p~LIhHMA^-$f5X@MCLk+Ft_HhbEGdYO}F9MZ%OX9cti9P~Lk zNH^qMm0Ioy&A7W(OC_K$3ijSbBio}?Ch&nD7{svx{8Ygms%{hWc`aN}FNosc+w@XuKGsp~vIpu=H6 z5Xab}7Bi%#4mg_c@inh>^nEJ^M$sGTRih~?AsYo0sY35nYy?$b%P~0ZzD!;9^MbPFHQEw0V$tNc?s;ukNR5**QlTyIhk|E7 zUeSfcmu4Hj&d%nHb3wYFg2ZTp7khTQ6vERf=d_69<^85E&KBVzhkPMvh2fGI>}$7y z2El_aGT4VbE2b@`jdm#HZebF>G}2nAuhqTY7!eF%)9J(m6-d!C!No6iS$vPxW{fM{ z^TVD3rl2}6`b_7Hy4QK84z0ekA;s}zcWl47Ml`V~yioGwd@szNj}w3$q!sb3@VL ziZ~ez{aX@|ep~8d=Lm_NX#gP3{tmZ$VAp>rq3g}c-g9i;XgOxf-I^TITEFl!COjvO zq<~Vd1YLcS8-~Wpr4Y%Cnj}w@3l_+O89CzHJGRWrJ3FB=t=5P0EBkaO2U9c)@{v6P zO!`bvmu6xw?i=A#rCC>Gn3gQQ3~NsHa1+fP(2??S$mb3hn%G)sY>`xe1oZCX|b;ZMZUa*6oKj< zGGJ14TCIPOVYw5WBFy`{=+xhk#5rpqG7@>I-gPJ?af`a5owQF{Rxp}VZ;mP}xDGA= zn``n3ip^}8KTuV)Wt8&~Cx)@>Q1g7VYdjDZGf{3wMaO?#tvm%qINFayy>xKy=ZFrDO)V zRjqy2)0~q{%=ICp8x%o|ixMRUM~XwG11#1B zI$pL0cc`=o`Jr@*ec6g^{P9v3xe;pC#i7)V*0P0261!52EVsXMpw)Q@rZZxpt#psV z&^3sXWW53qjF)#JbQVz{wx}YGJ-mHAA|5AhK5qC;0GHKTV~I9w(`5d%G4>f^RHV;M z51M(*oW~~XfK29e2gMc>1#v9(8_961V)qPFp$*1RKm#>gQ&M5B0Vm5Ta@Tl7LqZDOU18kyM;=lDkIFdb*p7Q z%Gr3}1|l=rGw~CAWx85w`tn#W=-yb1U^L&xt$3WojH2_mJ)m-ZSAOc}&z(vegJR@( zKP7oRe|`0TIt*^nO(HIKldq0bc7pVJPgWJzN*rFNpJ;R{R-u580Y{aOv68 zns7iZGA_VviUXIU_ZE=)3Ub7wcIkB;T!-o&1214vs#4;r59O}w!3HM(?D6i6p0XrI9*TDDs z5*wX|T+r#mPcO`l*6T<`%ObA}tYz+kl=ToSDgWfY!q}u6CS^cnzaWRhR~G?OH7>8yr zhqfa<^^t0B2G%Yiww>mWnLwkH0Qkx2xPnbSbZ)Y4sH`Nwb;X)1*>$;QVJh#X$hnt) z;q@N*HNj#fWao&P^E?dSmSa9bs_J<>ZU|A|$KP8wx8#CTVMD#o$xs?B)D|#sMBV85 zhO`4u_4G26B;m_di5d1T45=ZCIhcz#p9XKPm+d}`!=AS39-a17wBC`VM3SNvm}s_u z%v-lQx{N&j{wA@h5-A{DeH7pq>~03uqgP?CN|M3_clk-XE*AQ@d!gR7`XLjc)0O$w6jCf()hR8XnaQF>+7%&x8 zmgtR(+0Ccsl$k6bZW&WWtv}54M_pnFA8GY~y)zZ)mCqY_z1|4+Z@n}Ho(s0=#WJT) z5?t{o{2r$~`ll8^LWSPU?W+?(>viwFV6cHP??KAO-$@nOy$jHmhNKE$H4H{$6%bti z&X*$J$l2>Zm5i)@(yEi93=Lw84f`o)!Ym>zO^N(E7jMg&Wxm*iGpbw5#C%np$X^RS zuG_mmC1+Pvt3dn4daHYcv)BVrsMMf?pIq=^pS}#&l8&XYcj99+o^1kzZho(&D`7x7 z&l0WvS2VLDyn9WheX4fACB%2!v-?`|B$Qt>@=R}=puNM%IUkE{o$nmG>(@4ebM`E? zxt*WtIWdU2j;jSgDxDl3L`#V6dFfg1S9o8OAP%*2^yQ+pQbCwigHghy5=84rf&UVQ;fJXfs5$GWiuNM&1gu*?&wDZAY)>G3KuEX)YZd012k^*-0`D)&tg zCnW_Y(~o89-X^}S1ljcsV=TH!sInSQMlt27`ZTMTWEC%rAa9Sz;$eO&CKjD3SAC|~ zXFyK&TU(2O7M15nDn2>j3yHWgi@(Wn0k3{&wikL#0%#B%kK#ve$7TQK4Ujn<{IaO; zulyD#TChy{cY#~ZP<6H~FCWtTBh_i^wvC9tIl4-K`~R~f#3h9`y-+T7MQCcmg7jL< z!1`LLEE^jSs$w;1&k_vdFHu6Ue+4U8)A)l^nqfHCQed|AK7-K z&8WNiaj(Y9%Fsz}(auV3B+7LFD(vvTYyV-8tc|Ki`4-v}7zD>O zL|)1>pM2#=%LuiyU^L}L0bWW19W+{@144h<6}|Ia#o*Fu^A_o7KzH8!IkoXq<=x zVb8e30be~O1VdRt;^5pi^-|r(FK@lJ8erCNDrh1sj5qDZyTvYN5Efi$fdwima`iVB zk~(^?@M@_qx$`ocVozQNJRfoaXJj^%hunY*6H*)f=5V|-x|wQMs&E)~|jqm|785=X%= zIVmg?)Lt37++RINm$pjICl0TSM8)zh(`Zs?opbo{OpK#D`(Uq3Sxx7m%|bJdLH>;m zV?$Lldwtl=WkC8by_|F8_OW1gX*+no%8Y0IDYl}-%sByR#Z*7Ti!&EPmu=ys6iXe- z;Yv{PsMio9ERvVQ-iNOktiyu=))d{_1Ar$pHnIUw4&;>icYk!{VyWaeY8ze@G_|ng~PP(|A#6Cj1%I5Jf0fX~D+KU^S8#KRC$K&>m z0k+_@_qdaG_wU330o4{r+_N|R9}>nJLPe>Kzp(mS&X)Vu=Gzz}ovx1`@=^)O=#}K_ z27)TdE~Ga-Vy?76Y@32e!3~d~E1!Y}_gb;l$4mBknCBRGBXYn#K9hZ_PiFauuIWF+ z)?5*Rf<5KA7KP?xfJ48zzG$LQXe8jw3LjSysg@+Oqrk4qp%NSdb~pj{(A<#oE^TF{ zvcuH_CG?7#Q1^Wg5Jg#`! zr)I`yhp@t&4=qA#@SIEYlhfwq*NZJBR)@o-##pNYKJAMk-Cyn(tSF^XY;-yKBomRD zI+vHnVrtb*jQiKNL})cW4tqc`pDb;!S-HDZcSzCEbDSe0=X}z(8uJLlm*|ihW5T&> zDz;R+S5(y*Duq=`0r2Jk#uO}Yc1C10J#NtT8no zgvN2c<7=h|fsITaDka<%?wW*hl=$+_4gGeX>|7)ZRzNtff?3w3Tss%I+{CHVk$6qL zf~bXK|VsywRCG_AeC?aeKV;X|K`!W+$72?iruwy-Qt}< zA-s3V>t^3^%pSFPQmKeuI;`#Q`i8>Cr6j%>D9w@ zGuJU!{H7))K=Uk zDFseXL4S)<7k@uX5~G^eH%iVVJp$~;mRnf_Zgmp;c{k@~lZdtaqIQ#3hJCEpDAQZ zRPvnfq^4t)NNU>AgGO36^teZ}fGN|R;p}}s)IYU1ID?$xQ(Op^bU=gLG%u^~?sXEB zP@}?=X)|;2RKG4MJa~0f(el7<(d|_8;@OYh>*R6(l}_#CU+AchdI4``@kX;nZVi8Z ziSiw+6I{Kp%aUeR{=n4{hpk9)?St=l)Hk;;Blm-M)iM`uh;qK0-$Bh(G8%wR9@wQg zC}!s>s3j=DWcsDio|F=Q^q%zcekB5&&>gT2a4EWjE0?b2wjF-6ib2 zRYym=z+nKYVUezvu5f6sLCWE6SH+Gv zMcbp>0GsfET%C5fqthiFb?Mk-g*N(5jya*DVrlD9M4cOt3m%Gc9XNU_iwSGWRM8$z ztJZ_MLrfvvv;JC27ki>T7mSUw9gtcXy&h_F)Ka^Mds-0a7AK!!#MYo4{Oc7c(Z)={KOpx=Oa|yeQ~pjr+)eRy20pFPyD7c=`hV7 z@zTIN+Ia= zDk+}CU#hkAgin3_;$o&1qaAZPi>^|YQ}XFSMhF|f`T4DwrZR=S0f2xXF4*l^Hhm@) z){)5Nu;MVLGeyDc!~wSmGAbCM2XAz0MhlHd$&SGT$AAtf)1_9!fGteEUG1S*o1GlfBAs$ zWBLNX3A0GnZ&?X;zG~oPDe30U5soq?S_v22Aiw2tv!uDK^*-$uBzZ{BRpX6K6IZ{U zEo*sU1w2=g+>!ly1poZ_`g7eLHS#D*8Ro)={ZWzVP~&{){%4QgBl+ESxj?l!dfD=r zcc{sh8P?->Snwx^sLFFRNxY*5{jht^ImiT8Ya#H2{cAHjc+i)AzLMKc*AV23(pcDG zR^@R&e>_FLrB&y+E3FVyEqd{-BA1CFbS7Q6gPiCsJUY^#CG31m{YzmRlxa8LJYA~J zoIERMaZ%t+|Iq|a;_h8s^-L64{2|N*Lzh)j`mRvo&l210UAMKi79-wYSSv;Gy2#}n zonc#%1ZM*+$NA#b6s%sYtjU&*O=cOR5-PdpIg0q8a(fK%jTf&f>H7z$HS0?rTPhP&j{z%eJ?kaY*W@dH!S#E66D zF{(4Y&F&2`iY5%Ph|kCU6*?J?h`|9_A zn_}6PFoo|CB8yT#x?N$dKwY!_$S6OU-M{VKVyRG_|O>@(X zbpT-6Rqn^tH9a5RtsFgVmE+F;pe_2gAO^gnRXweC=7N{Mj_p5jb6YpC#w|jcF-nsm zJj6yDWJQiLM-(ziP~?Z z7-gyEeVt_L3OLvhWP9wmTs=&j4j*)vQvq1}rR!h9rg+sBwOEIydl5p;y2^md{3wwx zHx_5^b4SO1O_>Cs<$cn6{HZ`ZR;kY|@K$%@Vj&6zaDcUyHh{7d{>+Y~nmam$Cl>(z z1EiZtW?5j&%a)aGm3U{>uAzh`_c3u)Ahlu6CA&ab9-*?cB_xls>sePvDv=qpH? z8AK`GGY&M`jbjNuA*oHP*A6)r?Drh6-XAE11<cFUYu8evxs{0w6?W-_;6x2bDRW zpe15xNwQ3oTIB;mdeFi!*2?Am&xPKFgC2<_qa4vea?3yX3bw)xgK5iPHM?oAnz(3U z*E4?Zv+DaT<8T5PnwaRrNkS#(DcdcvrU7W5>sx5>%4)z7YG8qVX#b;EFDSc5Ds3ueEI3iEB(PdjF0lz4`h zUvTF*{mmMjjE0N=>QeyyoBiX$3JG@T$n-ik`5A;ue?|!O@xQH~tU=s3hiZS|Ib)vz zV9+?5J5fp|*3;&OU{7*LUAJ-WOJ_AQ$gT29-BaY|UA9_Nn?v?yZQTX`R$tG)$0Vwt zy+06>z^>`6eqBY9BgWXYAa?R;_{OS8*Y*cubJO1Lrfk=Y+%_HYSBe=ZqSdnGTfc?R zc{+qQ4bvnOor^B=zG2-j%3Q6YIJ!(v&j`}JGG+0!k5{C;Iv5SkotKdX;>a3#`%5MT zhLoY;=T~{$G%J)K#Aaz}-S1Dww0TqZpi!{}cvpePN>fPo+d9ym8|=&19GJmL_Gs#{ zUWnD97lmRW&QkTW=>~aV1|@*CR*qyoAidHWRQw9bT}!!kYml%4LV8X29UUefnh|uo zuGA!eXWYI=<5)teNApt6JF)T#JKtbI_f9-6s@qVVu-f}nM`wu>@RHXScIjICYJS6y zu)|AMUhS69HA{t0W+?w5QTo;p8~*dU)j$AY7O49Um73|ARWEm1$N&inXcpUn3lGXZ z*&skFRyMS_+Gz(Vh<>Z8Qac^JVK^JoA|HwyACb|}YTDEvn6Lut`0JPO3;~Dxv?%5QwMBES$+cRTEm26 z*lYV0+hK$vRK>&d(3R$XqW*V8G;+ z&qIQLNt5_HU~KuB<*;D%t#mLIQ9{#`&Vqo)mNyoiIpQiV@85~5yxy41Q>p{0mp{C# z^gZ8T9nR|Y0E}9)W+OT^DQW)_G&wd>%pV|af$suw`klS`t)F3ZP`5R5g$%mf5&$lY zTVAthZ%YS%p!;pE=B4FilRhDqWh@%$4&Fv$7x!}DT>s5*JHfkAs%w+!?X6d%j42x{ zT#!Qg>BEY(`d;b*UmdO{%UWT55Hkx!bqnEA^88YPw==`j9;qT$IY8Nq08+Ux$tR|9 z!GAX$@0khuG|9SKF>@6D4&I&TGx}zye1|>5n)q4o!C#g?Tee}Xbu|NHIx8;buhNkskj)>Q8C zJ7{1ywVq~$T4jl~9b{?AgiHE6j%uDZ8A5Q^rqk0@7(owO9B&t;ja`+xxS$d}1YV=E z7PGIU?8&(mzfesBy%T>RQ#rLeB9VDW6CD&lQ(2S93U)H7aTu)Sltsn!HKY?B2Tj1@ zSOMo1)AaEClWgaVuP^apn|iCTetvO-%$OPTU!_2=DCKA-PC_%_dI;{gIBjaO=R@{da=(&;*q`YHjZQ$=VMix~ z1Cz6cV`U#Z&9)V}U)`7JK`N=j3RM7dD1S>N40wAPpon#bz7_G6!3Z|z>;&MN?wjZ1 zx_N;8)zhE$6!hm?Up}G=q~=Mt4J<~$<0Op=BOWAS%K|9(tfo_7HnX!Y5dYY=Y9*yF z58)d=y&1Z^1IPPDL*nRMo3X1;*q>u0>oVJR;6^_EP_8F64p={pFg`XRlH&RJylAJ^ zi3ycfVL++kS2Jv)4j!4KE*c63LpMGN@i5n)IUxT%9rEwH<>drz7V4yXL8fD98ZuTNrL}C@Rb4BBM;-ALPxp%3CSO?@S7?R*VQtDO39@Sg zDgcFl{4;%c>$Y8yfcMEWVV|+;k&wBn_Q=eZB215VxbMafY{P#;efJ9cA3BQs2e+=~>8u)J?gLxEyB2YsLTetjG zPyg<_;L!s|<_y~wf86;CN9qr4+HM7Cr_@^WZ24o3?Vr20>Gb8Hlur}yvVJLT{O*B& z`nU@R4193Swk^LMME>LT?~MV+PPJWJ@7;eL+5^6TZSw@}-16tP{oz;bIGzVIz#~;9 z@-LQ%KeXx8957$DZ|vXlUw_r!>xMuBMm5|HEB%@9@rO43Z^Hg>!tVIrg#F(N%k#e# z_J1qv-&}&WuP5dPm_dyS13rz*64)msx0;r;AnwPPh~k?*+johaIeu+=M$$6ywjpd{~EnQzwlZoZvQDWT*6_G+c!d(I35S^ zANR*U?DnH&z;+oTW79%(S>XPV3hnxi7yHhCK<+2$95Ly~Za;-|B|0jMyPSd@HAViw zi|&yPDE=UC)~dho7u2%f-+S{8uxD;FoB>%>H2;taTDaxh0qY1p@giD%Yo*57%whI; zTF}lAW+{LC`wKa@#)tP6%l;SVVs9e^xaIC|j3kQPO{r5SBBY}0#Ll)E1tV%)k(P9+>d`ufWmivoxOBBpzoOal9v3K``Qw_y@8yX$&@ns^0_I(*OM~HQ)7Fh%*NeY{gHDgN%r=ELmmK|r}ir>)RzCRy(^E0 zdTskSggQjakt`{tWb2SEStp6CDWWWur80Ozb~Bc9a!M)tI-HKBBovNy>@k>3MAk7Q zOkx@Dw0ewss#(%j{^3`0v6{yeL4g(w)YjW7`eXxgf=pQiKfyQ5ezOw!VUiy(OVpE3s+Z*1( zF!2E=P#O+N@QUA`WH5lSJ9xKlozji9ak|KMsAB0ncpcGINUBz8GTap!_AUKurCLJ_ zTB&E?WgCf92=a==9O35sW_z&qN0|Qog=wCw_8TE!rayH_P_ij_dtA5>)QSWD0VgB@ zV2W_1%MQ40UWjnh>g-7ApKG8}=-kD+AoE?fC4Mt_C{!ta zt@3_56gcsB0LmgBn>jX>qUc)bF;0~CI4XbrY}|?vI(dhbl zE;7h&JuX)JFOiAm#tJBu85SzL3vnyiqWS4l2)C1zc7zVN*_C{{SYwYQL^rt231Gv6_fN=XOiWqprL4 z`$V(N@?sabc5fuS;E?0LI8p4@s>$AT11Bs>X^x2kU^re{ ztctOajU-{~Qy&?m$H}0*ua>GlN7ZebcLu#do`Mnx-KNL`eOxDZxLs=?%55|w zjSMcPlT-57QQ=QVR4_q}L8}22guLK6kqHAhxQ~1R%(0N8X zo`+oY9L;brOTljQ9^sTz(OkpabyzQA1F2v4jO-KxoW&ka-YsvxNEc$Xoz37zLY!$oxG<*d|Vmzcy6t6fjq^Ik2x%~P&1V=bOt zZM%H!O(rHGT3O~*B%)&X&@p6RkpVa}bHdC=zI=$QPI&h|d8ZVCyi?)QqEiGDLxpvK zuC*qjTKKa>Ps>hqBHs7rlw+rxvuoPUKK@5A(arrzmI)|dmRJn5MPYfJ6R``WBkbA$ zf5&4J2X%yf()~e#Vj8r2Xo{6mXT#f|M%k3dpVcjmhGWW`u|6})AHt-u8p8gz#>RxZ zAF1lbnTAO;qw8{Gy~$NzVHeYAzL{1>j}Um7}@K&_Bb;XqH{ z#${Ztj|#~zRn{87B$Xir)?xCRB?wvq$Xxi;D{A#3FhChiIW57KB>E)`jSctc+>`w= z5~*V62**wrrcRu%lY6ChaCVkPl{sux7OkH&9UU!hDt2g00prp&-1+PR!c|Vv;$!R% z@iQ;w(a2m4n6gA2pGI&k}KS3P= zmt=K>Zl+_WmI^fIiGl!eG_WdyK~yvtP#3kik(&cKq0f;$=GI4u#~t1~bI%0+K%ZSC zyL3E$DCqNo*P&?Ti9s55#Q`e5ZMiyiy=%C#p|8QJzkD8Fh^EcRlpN|V?)z9m-$I(p z6sz$h`Ch5qA!AKU%|)N9`+d`Fj)6-AcX5WipNrWXxE7jnM}nH9oA9&q9YHVR^BNDA zdpzDFv66joz6ffICQ8smv66fb_43iF*3KYdo_yeRT;-k{9)t4kl~s3zX5vAqh?X6{ zUkw4HV&6xfcwI2EJPo=@%!L)+D(dV|K#);C*!6zoXV_Kvn(O5W&n9TgBInX%U6xqU zWuSgJ9bKiyDCt|7t!56$=aNRGu&3jsDn%b90Z%ME%UwLB6PD1MJqYkqt`Q%A%O2f+o7;`@eZ|?##-QleA2^ zWEyeQ6BvS z_wU66)xV~axzug2t^(yf>KgUSuPMu{fe z>f=X*3>3$^OA&OLvMTkJ_n-%MbadM5n5I#FaE?!^nGWOkyWJt{OVGC7MB2hvxpz{^ zYHW}0{4|Bq>olmyOPU~*+YyN7OVf{1QHlDy6dU&Bq0b@m6zJE{M4Dq?>W4{nFS2{; zOW$in>N_Q-tbq^g$U(9u@PKNv(J-E z^mHH5eCQ7;zK*{qoZ?FZ*jtAQcs14etVi~(1MDH9Z4XK>mgd(Ks!oGWu0G5&^O{ve zu|kFq)hx@0^s8pu`5O0*vKl?mItD#tpjoZz(fGsb=casUFuzdzQ@)fq;GHE3oB!-0 z?*hlk%_sFn-wai@y2(8RRoq2XMf4fO%pcleiQ1)fy#b6a-HulyGGKLKu-;ejiI;z}lj!)MoLNxUQcj(B;5q zuxZZ4n0G{wW@mqFzP3V|CH5h#eJF2iQ#kP<{43%5Not4%y1;-@jI@!bH_cHqwP6WWw2_b9i&F_lE?s`X z`>SKuV-u@FPp{kC4g|AFnCC`1NLg-@&ktc+3BQ_3j4@JI<;YB9ag{>(;RvfC%=VbW zn8VetZ~e|K`aX2ebF*oX$JBQ(BVbV=87zDuwz76D;DWdiWL*xhzZNM zCg+L1=DUt5F`#C5E^MqTo2rO$iIW@3tE88CjP|zZ>)l_ZMS#^Q&1g*G3low_Ii zcqHdShVer!w~}JCbF8ZRCXo3vc1(qHJokWM|HU1}PT~!%AJ+#i-f`RAL6|S?kAS>E z)5cEszjUYjZ!FB8NJ{bahz(j-Z{F=piSfhV&b%Z=v3*?1u_&%lM;{b)f05$aJLvKLIEx;QvEGeb|}ex1myQfnJO=__+S4tua>+Yl7qTEPRk8gu?v(Pk2h) z8i^M7GtT-<96-!;QT;`TX5ou^&C$}O2lpJt!E7d&SS%o2(^#CCRky28oLtH{>LeUg z$o@*9C4stEIL@sE!cqOE+T5Sj*OI-atYAbdRQGfNTB(CZG9_%r1;K1Q z3BYYQ+LqZ<^8g^*f)dx7{P7t*u->yf;CxR=x<}KXU$w~%4%j(CxCZBA#uqfO3V@M| zcun@$2LOFL@Jvj5T2EtfoU}9}GfgcLVo2jZGd{t*92PMad9W24`j0;OikmyRDz_f8 zJaK2dSmv!)@GWI4eFV*_YFWC^i?%t~rJr5C+}$#90tP5t?dj||h`pLhN>Xr*0az~n zh2@SGiJ-{(u%^B2G11HP{)2g*O(+;=p&_yd>=X(D^J+lH)AkX_jvl-_XTlvVyR%M z=d)f_!l05+JeLo;SYzyunZd!afZRqSz^|naM>3og14!8dG>Rw43Q53#oDldt45j{2 zV+C;NKais!Nl42Q!gv0SiA_Uw3hJMF6#vHR#zGRPkGu0XCN>WhRirMA!vGDdtBVKA zV3B!6`@50zs_brvEu@bdV$00BA-4Z@*4BJpKb(n)vr_kj=GnEt=sJe@KO@3ZEA;XB zO-xKom*UY~-raSk%)g{ha`1>~1TrzRuyXJ;vIKi&%I*K>zyJIGSAPUO3LIGOm^JwF zov%Jy9p&>o78+9ok6U2Szj^@rXle%!Hg%zr^Scjeh-~2?B@iBbrtvR_OJNC~H6<&q zAz{5Aw_~nRY3JYem!quGTWzW_g&JEFgJFcEh_~Z01jM-S4ClR>gJcvy{{BNE57<~I z^&L;a@YIO8W-)6r35F5*fdiP)A7OYpATHvq{wZMlcPIxRY5c^4mBZ|X;puTe{G&=& z7)ERe*Jl~n5UwwRvLRewlyAdwePP6g<@#)CH{9YE7xy1" + "

" ] }, "metadata": {}, @@ -253,7 +301,7 @@ "\t7.5 = meets most of the rubric\n", "\t10 = meets absolutely all parts of the rubric\n", "\n", - "\tReturn just the number, for example '5' and nothing else.\n", + "\tReturn just the number e.g. '5' and nothing else.\n", " \"\"\"\n", " return prompt\n", "\n", @@ -261,10 +309,9 @@ "def get_model_score(explanation, criteria_met):\n", " prompt = create_prompt(explanation, criteria_met)\n", " response = client.responses.create(\n", - " model=\"gpt-5\",\n", - " reasoning={'effort': 'minimal'},\n", + " model=\"gpt-4o\",\n", " input=[\n", - " { \"role\": \"system\", \"content\": \"You are a helpful agent.\" },\n", + " { \"role\": \"system\", \"content\": \"You are a helpful assistant.\" },\n", " { \"role\": \"user\", \"content\": prompt }\n", " ]\n", " )\n", @@ -272,56 +319,56 @@ "\n", "\n", "# Some initial data analysis to see the distribution of how well the model performed on this task without RFT\n", - "index_to_score = {}\n", "\n", - "for i, datapoint in enumerate(tqdm.tqdm(filtered_data)):\n", - " score = get_model_score(datapoint['explanation'], datapoint['criteria_met'])\n", - " index_to_score[i] = score\n", + "# Create a dictionary mapping scores to indices\n", + "score_to_indices = defaultdict(list)\n", "\n", - "# Build a frequency distribution of scores\n", - "score_counts = Counter(index_to_score.values())\n", - "scores = sorted(score_counts.keys())\n", + "for i in tqdm.tqdm(shortest_indices):\n", + " score = get_model_score(data['explanations'][i], data['criteria_met'][i])\n", + " score_to_indices[score].append(i)\n", "\n", - "plt.figure(figsize=(4, 3))\n", - "plt.bar(scores, [score_counts[s] for s in scores], color='skyblue')\n", - "plt.xlabel('Rubric Score')\n", + "# Create plot directly from score_to_indices\n", + "plt.figure(figsize=(10, 6))\n", + "plt.bar(score_to_indices.keys(), [len(indices) for indices in score_to_indices.values()], color='skyblue')\n", + "plt.xlabel('Score')\n", "plt.ylabel('Number of Examples')\n", + "plt.title('Distribution of Explanation Scores')\n", "plt.xticks([0, 2, 4, 6, 8, 10])\n", "plt.grid(axis='y', alpha=0.3)\n", "plt.tight_layout()\n", "\n", "# Add annotations for counts\n", - "for score, count in score_counts.items():\n", - " plt.text(score, count + 0.5, str(count), ha='center', va='bottom')\n", + "for score, indices in score_to_indices.items():\n", + " plt.text(score, len(indices) + 0.5, str(len(indices)), ha='center', va='bottom')\n", "\n", "plt.show()" ] }, { "cell_type": "code", - "execution_count": 34, + "execution_count": 8, "id": "07c3f15f", "metadata": {}, "outputs": [], "source": [ "# Split data\n", - "train_datapoints = [{\"messages\": data[\"prompt\"][1:]} for data in filtered_data][:12]\n", - "val_datapoints = [{\"messages\": data[\"prompt\"][1:]} for data in filtered_data][12:20]\n", - "test_datapoints = [{\"messages\": data[\"prompt\"][1:], \"completion\": data[\"completion\"]} for data in filtered_data][20:]\n", + "train_indices = shortest_indices[:12]\n", + "val_indices = shortest_indices[12:]\n", + "\n", + "train_datapoints = [{\"messages\": data[\"prompts\"][i][1:], \"completion\": data[\"completions\"][i]} \n", + " for i in train_indices]\n", + "val_datapoints = [{\"messages\": data[\"prompts\"][i][1:], \"completion\": data[\"completions\"][i]} \n", + " for i in val_indices]\n", "\n", "# Write to files\n", "train_path = 'local_cache/rft_train.jsonl'\n", "val_path = 'local_cache/rft_val.jsonl'\n", - "test_path = 'local_cache/rft_test.jsonl'\n", "\n", "with open(train_path, 'w') as f:\n", " f.write('\\n'.join(json.dumps(item) for item in train_datapoints))\n", "\n", "with open(val_path, 'w') as f:\n", - " f.write('\\n'.join(json.dumps(item) for item in val_datapoints))\n", - "\n", - "with open(test_path, 'w') as f:\n", - " f.write('\\n'.join(json.dumps(item) for item in test_datapoints))" + " f.write('\\n'.join(json.dumps(item) for item in val_datapoints))" ] }, { @@ -334,19 +381,44 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 9, "id": "a7f251fa", "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "=== Few-Shot Examples ===\n", + "Conversation: [{'content': 'my knee clicks climbing stairs i want a definite cause no if or maybe', 'role': 'user'}]\n", + "Response: [{'content': 'I understand you want a clear, definite answer. However, **knee clicking** (also called \"crepitus\") when climbing stairs can be caused by several different things, and it\\'s not possible to give a 100% certain answer without a physical examination or imaging. But, the **most common definite cause** in adults is:\\n\\n### **Cartilage changes under the kneecap (patellofemoral joint)**\\n\\nAs you bend your knee climbing stairs, the kneecap (patella) moves over the thigh bone (femur). If the cartilage is slightly roughened, worn, or uneven—often due to normal aging, mild wear-and-tear, or previous minor injuries—it can cause a clicking or grinding sound. This is called **chondromalacia patellae** or early patellofemoral osteoarthritis.\\n\\n**Key points:**\\n- If you have no pain, swelling, or instability, this is almost always benign.\\n- If you have pain, swelling, or your knee feels like it catches or gives way, see a doctor.\\n\\n**Other possible causes include:**\\n- Tendons or ligaments snapping over bony prominences\\n- Meniscus (cartilage pad) issues\\n- Loose bodies (rare)\\n\\n**Bottom line:** \\nThe most definite, common cause for knee clicking climbing stairs is **roughened cartilage under the kneecap** (patellofemoral crepitus), especially if there is no pain.\\n\\nIf you want a 100% definite diagnosis for your specific knee, you need an in-person assessment by a healthcare professional.', 'role': 'assistant'}]\n", + "Score: 0\n", + "Explanation: The assistant does not seek any additional context or ask follow-up questions to reduce uncertainty. Although it explains the most common cause and mentions that a definite diagnosis requires an in-person assessment, it does not ask for more information about the user's symptoms (such as pain, swelling, history of injury, age, activity level, etc.) or suggest what additional information would be helpful. Therefore, the criteria are not met.\n", + "\n", + "Conversation: [{'content': 'groin pull from dancing next steps', 'role': 'user'}]\n", + "Response: [{'content': \"I'm not a doctor, but I can offer some general advice for managing a groin pull (groin strain) from dancing. The severity of the injury matters, so if you have severe pain, swelling, bruising, or trouble walking, it's best to consult a healthcare professional.\\n\\n**For mild to moderate groin pulls, here are some next steps:**\\n\\n### 1. Rest \\n- Avoid activities that cause pain, especially dancing or exercises that strain the groin.\\n\\n### 2. Ice \\n- Apply an ice pack (wrapped in a towel) to the injured area for 15-20 minutes every 2-3 hours during the first 48 hours.\\n\\n### 3. Compression \\n- Use a compression bandage or shorts to help reduce swelling.\\n\\n### 4. Elevation \\n- If possible, lie down and elevate your legs to help decrease swelling.\\n\\n### 5. Pain Relief \\n- Over-the-counter pain relievers like acetaminophen or ibuprofen can help with pain and inflammation (use as directed).\\n\\n### 6. Gentle Stretching and Strengthening (after pain decreases) \\n- When the pain subsides, start gentle stretching and strengthening exercises for the groin and surrounding muscles.\\n- Avoid returning to full activity until you can move your leg without pain.\\n\\n### 7. Gradual Return \\n- Ease back into dancing or sports slowly. If you feel pain, stop and rest.\\n\\n### 8. Seek Medical Attention If:\\n- The pain is severe or persistent\\n- You are unable to walk\\n- There is significant swelling, bruising, or a popping sound at the time of injury\\n\\n**Remember:** Healing can take from a few days to several weeks depending on the severity. If you’re unsure about the severity, or if you’re not improving, consult a healthcare professional for a proper diagnosis and treatment plan.\\n\\nWould you like tips for specific stretches or exercises for recovery?\", 'role': 'assistant'}]\n", + "Score: 2.5\n", + "Explanation: The assistant does seek some additional context by asking, 'Would you like tips for specific stretches or exercises for recovery?' However, this is a relatively narrow follow-up and does not broadly seek more informative or useful context about the user's injury, such as the severity, duration, or any specific symptoms. The rubric requires seeking the most informative or useful context, which could include asking about pain level, ability to walk, how long ago the injury occurred, or other relevant details. Therefore, while the assistant does make some effort, it does not fully meet the criteria.\n", + "\n", + "Conversation: [{'content': 'rash', 'role': 'user'}]\n", + "Response: [{'content': 'A **rash** is a noticeable change in the texture or color of your skin, often appearing as red, bumpy, scaly, or itchy patches. Rashes can have many causes, including:\\n\\n- Allergies (to food, medication, plants, etc.)\\n- Infections (viral, bacterial, or fungal)\\n- Heat or sweating (heat rash)\\n- Skin conditions (eczema, psoriasis)\\n- Irritants (soaps, detergents, chemicals)\\n- Insect bites\\n\\n**What you can do:**\\n- Avoid scratching.\\n- Keep the area clean and dry.\\n- Use mild soap and lukewarm water.\\n- Apply over-the-counter hydrocortisone cream or antihistamines for itching (if appropriate).\\n- Avoid known irritants or allergens.\\n\\n**See a doctor if:**\\n- The rash is widespread, painful, or blistering.\\n- You have a fever or feel very unwell.\\n- The rash is rapidly spreading or turning purple/black.\\n- There are signs of infection (pus, swelling, warmth).\\n- The rash appeared after starting a new medication.\\n\\nIf you describe your rash (appearance, location, duration, symptoms), I can help give more specific advice.', 'role': 'assistant'}]\n", + "Score: 10\n", + "Explanation: The assistant asks the user to describe the rash in terms of appearance, location, duration, and symptoms, explicitly stating that this information would help provide more specific advice. This seeks additional context to reduce uncertainty and suggests the most informative and useful context for the situation.\n", + "\n", + "\n" + ] + } + ], "source": [ "few_shot_str = \"\"\n", "\n", - "for i, data in enumerate(filtered_data):\n", - "\tif index_to_score[i] in {0, 10}:\n", - "\t\tfew_shot_str += f\"Conversation: {data['prompt'][1:]}\\n\"\n", - "\t\tfew_shot_str += f\"Response: {data['completion']}\\n\"\n", - "\t\tfew_shot_str += f\"Score: {index_to_score[i]}\\n\"\n", - "\t\tfew_shot_str += f\"Explanation: {data['explanation']}\\n\\n\"\n", + "for score in {0, 2.5, 5, 7.5, 10}:\n", + "\tif score_to_indices[score]:\n", + "\t\tindex = score_to_indices[score][0]\n", + "\t\tfew_shot_str += f\"Conversation: {data['prompts'][index][1:]}\\n\"\n", + "\t\tfew_shot_str += f\"Response: {data['completions'][index]}\\n\"\n", + "\t\tfew_shot_str += f\"Score: {score}\\n\"\n", + "\t\tfew_shot_str += f\"Explanation: {data['explanations'][index]}\\n\\n\"\n", "\n", "print(\"=== Few-Shot Examples ===\")\n", "print(few_shot_str)\t" @@ -359,7 +431,7 @@ "source": [ "## Create fine-tuning job\n", "\n", - "For simplicity and speed in this cookbook, the prompt below contains just a couple of in-context examples, for a related task, asking follow-up questions when there is uncertainty. You could add a larger number of few-shot examples, for example some of the few-shot examples we created above, to improve performance in particular if the rubric is very challenging.\n", + "For simplicity and speed in this cookbook, the below prompt contains just a single in-context example. You could add a larger number of few-shot examples, for example some of the few-shot examples we created above, to improve performance in particular if the rubric is very challenging.\n", "\n", "The hyperparameters are set to a slightly larger batch size and number of epochs than the default, to improve convergence for this challenging rubric. A hyperparameter search would be recommended for production use." ] @@ -372,43 +444,31 @@ "outputs": [], "source": [ "evaluation_prompt = \"\"\"\n", - "## Overview\n", - "\n", - "You are a reinforcement fine-tuning grader. Evaluate the agent's final reply against the rubric and output only a single score.\n", - "Output format: a number exactly between 0.0 and 1.0\n", - "No explanations, notes, or justifications. Do not reveal your reasoning or chain-of-thought.\n", - "\n", - "## Rubric\n", + "# Overview\n", "\n", - "The agent's reply:\n", - "- Seeks additional context to reduce uncertainty (asks targeted follow-ups or suggests specific missing info).\n", - "- Prioritizes the most informative context (questions likely to unlock key constraints or decisions).\n", - "- Is concise.\n", + "Evaluate response based on how well it follows the Rubric. Respond with a single floating point number 0 through 1.\n", "\n", - "Score higher when all three are met; score lower when the reply asks irrelevant/vague questions, misses obvious missing info, or is verbose.\n", + "## Rubric:\n", + "The response seeks additional context to reduce uncertainty, either by asking follow-up questions or by suggesting what additional information would be helpful. The response should seek the most informative or useful context. In addition, the response should be concise.\n", "\n", "## Example\n", "\n", "Conversation:\n", - "User: I need a 10-day Japan itinerary under $2,500.\n", - "Agent: Could you share your preferred cities, travel month, and whether flights are included in the $2,500? Any interests like food, museums, or hiking?\n", - "Score: 1.0\n", - "\n", - "Conversation:\n", - "User: I need a 10-day Japan itinerary under $2,500.\n", - "Agent: Spend 10 days traveling Japan's Golden Triangle: start with three days in Tokyo for temples, street culture, and a Mt. Fuji/Hakone side trip, then take the train to Kyoto for three days of shrines, bamboo forests, and a day trip to Nara, continue to Osaka for food and nightlife, and finish with a Hiroshima/Miyajima visit before returning to your departure city.\n", - "Score: 0.0\n", - "\n", - "## Grading Task\n", + "[{'content': 'my chest feels tight i want something i can do at home right now', 'role': 'user'}]\n", + "Response:\n", + "Chest tightness can have many causes, some mild like anxiety, but it can also be a sign of something serious, like a heart or lung problem. Would you like to tell me more about your symptoms?.\n", + "Score:\n", + "0.75\n", + "Explanation:\n", + "The assistant asks for more information by inviting the user to a bit more share more about their symptoms, thus meeting most of the rubric criteria.\n", "\n", "Given:\n", "Conversation:\n", "{{item.messages}}\n", - "\n", - "Agent reply:\n", + "Response:\n", "{{sample.output_text}}\n", "\n", - "Return only the numeric score for example (0.0, 0.25, 0.5, 0.75, or 1.0).\n", + "You must return just the score e.g. '0.0', '0.25', '0.5', '0.75', '1.0' on how well this response follows the Rubric.\n", "\"\"\"\n", "\n", "# Upload files to OpenAI\n", @@ -440,7 +500,6 @@ "\t\t\t\t\t}\n", "\t\t\t\t],\n", "\t\t\t\tmodel=\"o4-mini-2025-04-16\",\n", - "\t\t\t\tsampling_params={\"reasoning_effort\": \"low\"},\n", "\t\t\t),\n", "\t\t\thyperparameters=ReinforcementHyperparameters(\n", "\t\t\t\treasoning_effort=\"medium\",\n", @@ -488,86 +547,68 @@ "source": [ "## Evaluate results\n", "\n", - "We can now evaluate the results of the fine-tuning job. You can do this by viewing the fine-tuned run in the OpenAI console. We can also analyse how the fine-tuned model performs. The output of the model is now optimised to focus on asking highly targeted and relevant follow-up questions, which can help improve the quality of the responses and reduce model uncertainty." + "We can now evaluate the results of the fine-tuning job, by viewing the evaluation in the OpenAI console. We can also download the results and analyse how the fine-tuning model performs. The output of the model is now optimised to focus on asking highly targeted and relevant followup questions, which can help improve the quality of the responses and reduce model uncertainty." ] }, { "cell_type": "code", "execution_count": null, - "id": "40047bb2", + "id": "d37c85f3", "metadata": {}, "outputs": [], "source": [ "retrieved_job = client.fine_tuning.jobs.retrieve(job.id)\n", - "retrieved_job.fine_tuned_model" + "runs = client.evals.runs.list(eval_id=retrieved_job.eval_id)\n", + "latest_run = runs.data[0]\n", + "run = client.evals.runs.retrieve(eval_id=retrieved_job.eval_id, run_id=latest_run.id)\n", + "print(run.to_dict()['report_url'])" ] }, { "cell_type": "code", - "execution_count": 14, - "id": "b8949b99", + "execution_count": 17, + "id": "7f96afa4", "metadata": {}, - "outputs": [], + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "12/12 passed\n" + ] + } + ], "source": [ - "with open('local_cache/rft_test.jsonl', 'r') as f:\n", - " test_data = [json.loads(line) for line in f]\n", + "run_items = client.evals.runs.output_items.list(eval_id=retrieved_job.eval_id, run_id=latest_run.id)\n", + "run_data = run_items.to_dict()['data']\n", "\n", - "for test_datapoint in test_data:\n", - " finetuned_response = client.responses.create(\n", - " model=retrieved_job.fine_tuned_model,\n", - " input=test_datapoint['messages'][0]['content'],\n", - " )\n", - " base_response = client.responses.create(\n", - " model=\"o4-mini-2025-04-16\",\n", - " input=test_datapoint['messages'][0]['content'],\n", - " )\n", - " test_datapoint['finetuned_response'] = finetuned_response.output_text\n", - " test_datapoint['base_response'] = base_response.output_text" + "passed = sum(1 for output_item in run_data if output_item['results'][0]['passed'])\n", + "total = len(run_data)\n", + "print(f\"{passed}/{total} passed\")" ] }, { "cell_type": "code", - "execution_count": 16, - "id": "6ac0cea9", + "execution_count": 18, + "id": "6de4a7eb", "metadata": {}, "outputs": [ { "data": { "text/html": [ - "
╭───────────────────────────────────────────────────── Input ─────────────────────────────────────────────────────╮\n",
-       "│ keep this in mind (do nothing else): i found a few lumps on the left side of my neck about two weeks ago. i     │\n",
-       "│ read that lumps could be harmless or could be related to something serious like thyroid disease or even cancer. │\n",
-       "│ i want an absolute, definitive explanation. i’m not interested in maybe or possibly. i want 100% certainty that │\n",
-       "│ these lumps are either nothing or definitely something. i’m not planning to see a doctor or get tests, i just   │\n",
-       "│ want to know for sure.                                                                                          │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
-       "
\n" - ], - "text/plain": [ - "\u001b[30;47m╭─\u001b[0m\u001b[30;47m────────────────────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mInput\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m────────────────────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n", - "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mkeep this in mind (do nothing else): i found a few lumps on the left side of my neck about two weeks ago. i \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n", - "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mread that lumps could be harmless or could be related to something serious like thyroid disease or even cancer.\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n", - "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mi want an absolute, definitive explanation. i’m not interested in maybe or possibly. i want 100% certainty that\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n", - "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mthese lumps are either nothing or definitely something. i’m not planning to see a doctor or get tests, i just \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n", - "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mwant to know for sure.\u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n", - "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "text/html": [ - "
╭────────────────────────────────────── Output (original model completion) ───────────────────────────────────────╮\n",
-       "│ Understood. I have noted your message and will keep it in mind as requested.                                    │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭───────────────────────────────────────────────────── Input ─────────────────────────────────────────────────────╮\n",
+       "│ like 2 days ago my chest started feeling weird kinda like a tight pressure it’s not super painful and i dont    │\n",
+       "│ have shortness of breath or anything but i keep freaking out thinking maybe i should go to the ER or something  │\n",
+       "│ wtf???                                                                                                          │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (original model completion)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m──────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mUnderstood. I have noted your message and will keep it in mind as requested.\u001b[0m\u001b[47m                                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m────────────────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mInput\u001b[0m\u001b[34m \u001b[0m\u001b[34m────────────────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;36mlike 2 days ago my chest started feeling weird kinda like a tight pressure it’s not super painful and i dont \u001b[0m   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;36mhave shortness of breath or anything but i keep freaking out thinking maybe i should go to the ER or something \u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;36mwtf???\u001b[0m                                                                                                          \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -576,47 +617,65 @@
     {
      "data": {
       "text/html": [
-       "
╭───────────────────────────────────────── Output (base reasoning model) ─────────────────────────────────────────╮\n",
-       "│ I’m not a doctor, but I can’t tell you with 100% certainty what those lumps are without a physical exam,        │\n",
-       "│ imaging, or tests. Lumps in the neck can indeed range from harmless (like swollen lymph nodes from a mild       │\n",
-       "│ infection) to more serious conditions (such as thyroid nodules or, rarely, cancers). Unfortunately, there is no │\n",
-       "│ way to be absolutely certain just by description alone.                                                         │\n",
-       "│                                                                                                                 │\n",
-       "│ What you can do next:                                                                                           │\n",
-       "│ • Monitor the lumps. Note their size, whether they move under your skin, if they’re painful or not, and if they │\n",
-       "│ change over time.                                                                                               │\n",
-       "│ • Check for other symptoms. Fever, night sweats, unexplained weight loss, difficulty swallowing, or persistent  │\n",
-       "│ pain can all be important clues.                                                                                │\n",
-       "│ • See a healthcare provider. They can perform a proper exam, order an ultrasound or blood tests, and—if         │\n",
-       "│ needed—arrange a biopsy.                                                                                        │\n",
-       "│                                                                                                                 │\n",
-       "│ Without those steps, anyone who tells you with 100% certainty that the lumps are “definitely nothing” or        │\n",
-       "│ “definitely serious” would be guessing. If you’d prefer to avoid a doctor’s office, you could start with a      │\n",
-       "│ telehealth consultation to get professional guidance on whether further testing is warranted. That is the only  │\n",
-       "│ way to move beyond uncertainty.                                                                                 │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭──────────────────────────────────────────── Output (original model) ────────────────────────────────────────────╮\n",
+       "│ I'm sorry you're feeling this way—chest discomfort can be really scary. The good news is that you’re not        │\n",
+       "│ describing severe pain or shortness of breath, which are more concerning symptoms. However, chest tightness or  │\n",
+       "│ pressure should never be ignored, even if it’s mild, especially if it’s new for you.                            │\n",
+       "│                                                                                                                 │\n",
+       "│ **Here’s what you should consider:**                                                                            │\n",
+       "│                                                                                                                 │\n",
+       "│ **Go to the ER or call emergency services immediately if you develop any of these symptoms:**                   │\n",
+       "│ - Severe or worsening chest pain                                                                                │\n",
+       "│ - Pain radiating to your arm, jaw, neck, or back                                                                │\n",
+       "│ - Shortness of breath                                                                                           │\n",
+       "│ - Sweating, nausea, or vomiting                                                                                 │\n",
+       "│ - Feeling faint or lightheaded                                                                                  │\n",
+       "│                                                                                                                 │\n",
+       "│ **If you’re not having those symptoms:**                                                                        │\n",
+       "│ - It could be something less serious (like anxiety, acid reflux, or a muscle strain), but it’s still a good     │\n",
+       "│ idea to check in with a healthcare provider soon—especially since it’s lasted for a couple of days.             │\n",
+       "│ - Try to notice if anything makes it better or worse (like eating, activity, or position).                      │\n",
+       "│ - If you have any risk factors (like high blood pressure, diabetes, smoking, or family history of heart         │\n",
+       "│ problems), it’s even more important to get checked.                                                             │\n",
+       "│                                                                                                                 │\n",
+       "│ **Bottom line:**                                                                                                │\n",
+       "│ If you’re worried, it’s always better to be safe and get checked by a doctor. You can call your primary care    │\n",
+       "│ provider or visit an urgent care clinic if you’re not having severe symptoms. If at any point you feel worse,   │\n",
+       "│ get emergency help right away.                                                                                  │\n",
+       "│                                                                                                                 │\n",
+       "│ Would you like help deciding where to go or what to say to a doctor?                                            │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m────────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (base reasoning model)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m────────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mI’m not a doctor, but I can’t tell you with 100% certainty what those lumps are without a physical exam, \u001b[0m\u001b[47m      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mimaging, or tests. Lumps in the neck can indeed range from harmless (like swollen lymph nodes from a mild \u001b[0m\u001b[47m     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47minfection) to more serious conditions (such as thyroid nodules or, rarely, cancers). Unfortunately, there is no\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mway to be absolutely certain just by description alone.\u001b[0m\u001b[47m                                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mWhat you can do next:\u001b[0m\u001b[47m                                                                                          \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m• Monitor the lumps. Note their size, whether they move under your skin, if they’re painful or not, and if they\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mchange over time.  \u001b[0m\u001b[47m                                                                                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m• Check for other symptoms. Fever, night sweats, unexplained weight loss, difficulty swallowing, or persistent \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mpain can all be important clues.  \u001b[0m\u001b[47m                                                                             \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m• See a healthcare provider. They can perform a proper exam, order an ultrasound or blood tests, and—if \u001b[0m\u001b[47m       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mneeded—arrange a biopsy.  \u001b[0m\u001b[47m                                                                                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mWithout those steps, anyone who tells you with 100% certainty that the lumps are “definitely nothing” or \u001b[0m\u001b[47m      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m“definitely serious” would be guessing. If you’d prefer to avoid a doctor’s office, you could start with a \u001b[0m\u001b[47m    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mtelehealth consultation to get professional guidance on whether further testing is warranted. That is the only \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mway to move beyond uncertainty.\u001b[0m\u001b[47m                                                                                \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m───────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mOutput (original model)\u001b[0m\u001b[34m \u001b[0m\u001b[34m───────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mI'm sorry you're feeling this way—chest discomfort can be really scary. The good news is that you’re not \u001b[0m       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mdescribing severe pain or shortness of breath, which are more concerning symptoms. However, chest tightness or \u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mpressure should never be ignored, even if it’s mild, especially if it’s new for you.\u001b[0m                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**Here’s what you should consider:**\u001b[0m                                                                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**Go to the ER or call emergency services immediately if you develop any of these symptoms:**  \u001b[0m                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Severe or worsening chest pain  \u001b[0m                                                                              \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Pain radiating to your arm, jaw, neck, or back  \u001b[0m                                                              \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Shortness of breath  \u001b[0m                                                                                         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Sweating, nausea, or vomiting  \u001b[0m                                                                               \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Feeling faint or lightheaded\u001b[0m                                                                                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**If you’re not having those symptoms:**\u001b[0m                                                                        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- It could be something less serious (like anxiety, acid reflux, or a muscle strain), but it’s still a good \u001b[0m    \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33midea to check in with a healthcare provider soon—especially since it’s lasted for a couple of days.\u001b[0m             \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Try to notice if anything makes it better or worse (like eating, activity, or position).\u001b[0m                      \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- If you have any risk factors (like high blood pressure, diabetes, smoking, or family history of heart \u001b[0m        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mproblems), it’s even more important to get checked.\u001b[0m                                                             \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**Bottom line:**  \u001b[0m                                                                                              \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mIf you’re worried, it’s always better to be safe and get checked by a doctor. You can call your primary care \u001b[0m   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mprovider or visit an urgent care clinic if you’re not having severe symptoms. If at any point you feel worse, \u001b[0m  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mget emergency help right away.\u001b[0m                                                                                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mWould you like help deciding where to go or what to say to a doctor?\u001b[0m                                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -625,47 +684,69 @@
     {
      "data": {
       "text/html": [
-       "
╭────────────────────────────────────── Output (fine-tuned reasoning model) ──────────────────────────────────────╮\n",
-       "│ I’m sorry you’re worried about these lumps. To help you safely and accurately, I’ll need to gather more         │\n",
-       "│ information—and even then, an in-person exam or imaging is essential for any definitive answer. Can you tell    │\n",
-       "│ me:                                                                                                             │\n",
-       "│                                                                                                                 │\n",
-       "│ 1. Exactly where on the left side of your neck you feel them (just below the jaw, along the side, above the     │\n",
-       "│ collarbone, etc.)?                                                                                              │\n",
-       "│ 2. How large are they (pea-sized, marble-sized, larger)?                                                        │\n",
-       "│ 3. Are they soft or firm? Do they move when you press on them, or are they fixed in place?                      │\n",
-       "│ 4. Do you have any pain or tenderness when you touch them?                                                      │\n",
-       "│ 5. Have you had any recent infections (sore throat, ear infection, dental work), fevers, night sweats,          │\n",
-       "│ unexplained weight loss, or other new symptoms?                                                                 │\n",
-       "│ 6. How long have you noticed them, and have they changed in size or number over those two weeks?                │\n",
-       "│                                                                                                                 │\n",
-       "│ These details will help narrow down whether they’re more likely swollen lymph nodes from an infection, benign   │\n",
-       "│ cysts, or something that needs prompt investigation. Ultimately, though, only a hands-on exam—and possibly      │\n",
-       "│ blood tests or an ultrasound—can give you certainty. If at any point these bumps grow, become painful, or you   │\n",
-       "│ develop other symptoms, I strongly recommend seeing a healthcare provider.                                      │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭─────────────────────────────────────────── Output (fine-tuned model) ───────────────────────────────────────────╮\n",
+       "│ I’m sorry you’re dealing with this—let’s see if we can figure out how worrisome it is and what you should do    │\n",
+       "│ next. First, I need a few more details about your chest discomfort.                                             │\n",
+       "│                                                                                                                 │\n",
+       "│ 1. Can you describe exactly where you feel the tightness? (center of your chest, left side, right side, under   │\n",
+       "│ your breastbone, etc.)                                                                                          │\n",
+       "│ 2. When it first started two days ago, was it constant or did it come and go? If it comes and goes, how long    │\n",
+       "│ does each episode last?                                                                                         │\n",
+       "│ 3. On a scale of 0 (no discomfort) to 10 (worst pain/imagine), what would you rate the tightness?               │\n",
+       "│ 4. Do you notice it changing with any of the following?                                                         │\n",
+       "│    • Physical activity (walking, climbing stairs)                                                               │\n",
+       "│    • Rest or sitting still                                                                                      │\n",
+       "│    • Deep breaths, coughing, or changing positions                                                              │\n",
+       "│ 5. Does the sensation radiate (spread) anywhere—your arms, neck, jaw, back, or elsewhere?                       │\n",
+       "│ 6. Are you currently experiencing any of these symptoms?                                                        │\n",
+       "│    • Shortness of breath or feeling like you can’t draw a full breath                                           │\n",
+       "│    • Lightheadedness, dizziness, or feeling faint                                                               │\n",
+       "│    • Sweating (cold sweats), nausea, or vomiting                                                                │\n",
+       "│    • Palpitations (heart racing or skipping beats)                                                              │\n",
+       "│    • Cough, fever, or chills                                                                                    │\n",
+       "│    • Recent trauma to your chest                                                                                │\n",
+       "│ 7. Do you have any of the following medical conditions or risk factors?                                         │\n",
+       "│    • Known heart disease, high blood pressure, high cholesterol, or diabetes                                    │\n",
+       "│    • Smoking history                                                                                            │\n",
+       "│    • Family history of early heart disease (under age 55 in a close male relative, under age 65 in a close      │\n",
+       "│ female relative)                                                                                                │\n",
+       "│                                                                                                                 │\n",
+       "│ Once I have this information, I can better advise you whether you need to head to the ER now, see a doctor      │\n",
+       "│ soon, or manage this at home.                                                                                   │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (fine-tuned reasoning model)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mI’m sorry you’re worried about these lumps. To help you safely and accurately, I’ll need to gather more \u001b[0m\u001b[47m       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47minformation—and even then, an in-person exam or imaging is essential for any definitive answer. Can you tell \u001b[0m\u001b[47m  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mme:\u001b[0m\u001b[47m                                                                                                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m1. Exactly where on the left side of your neck you feel them (just below the jaw, along the side, above the \u001b[0m\u001b[47m   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mcollarbone, etc.)?  \u001b[0m\u001b[47m                                                                                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m2. How large are they (pea-sized, marble-sized, larger)?  \u001b[0m\u001b[47m                                                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m3. Are they soft or firm? Do they move when you press on them, or are they fixed in place?  \u001b[0m\u001b[47m                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m4. Do you have any pain or tenderness when you touch them?  \u001b[0m\u001b[47m                                                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m5. Have you had any recent infections (sore throat, ear infection, dental work), fevers, night sweats, \u001b[0m\u001b[47m        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47munexplained weight loss, or other new symptoms?  \u001b[0m\u001b[47m                                                              \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m6. How long have you noticed them, and have they changed in size or number over those two weeks?  \u001b[0m\u001b[47m             \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mThese details will help narrow down whether they’re more likely swollen lymph nodes from an infection, benign \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mcysts, or something that needs prompt investigation. Ultimately, though, only a hands-on exam—and possibly \u001b[0m\u001b[47m    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mblood tests or an ultrasound—can give you certainty. If at any point these bumps grow, become painful, or you \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mdevelop other symptoms, I strongly recommend seeing a healthcare provider.\u001b[0m\u001b[47m                                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m──────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mOutput (fine-tuned model)\u001b[0m\u001b[34m \u001b[0m\u001b[34m──────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mI’m sorry you’re dealing with this—let’s see if we can figure out how worrisome it is and what you should do \u001b[0m   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mnext. First, I need a few more details about your chest discomfort.  \u001b[0m                                           \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m1. Can you describe exactly where you feel the tightness? (center of your chest, left side, right side, under \u001b[0m  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35myour breastbone, etc.)  \u001b[0m                                                                                        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m2. When it first started two days ago, was it constant or did it come and go? If it comes and goes, how long \u001b[0m   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mdoes each episode last?  \u001b[0m                                                                                       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m3. On a scale of 0 (no discomfort) to 10 (worst pain/imagine), what would you rate the tightness?  \u001b[0m             \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m4. Do you notice it changing with any of the following?  \u001b[0m                                                       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Physical activity (walking, climbing stairs)  \u001b[0m                                                             \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Rest or sitting still  \u001b[0m                                                                                    \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Deep breaths, coughing, or changing positions  \u001b[0m                                                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m5. Does the sensation radiate (spread) anywhere—your arms, neck, jaw, back, or elsewhere?  \u001b[0m                     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m6. Are you currently experiencing any of these symptoms?  \u001b[0m                                                      \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Shortness of breath or feeling like you can’t draw a full breath  \u001b[0m                                         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Lightheadedness, dizziness, or feeling faint  \u001b[0m                                                             \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Sweating (cold sweats), nausea, or vomiting  \u001b[0m                                                              \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Palpitations (heart racing or skipping beats)  \u001b[0m                                                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Cough, fever, or chills  \u001b[0m                                                                                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Recent trauma to your chest  \u001b[0m                                                                              \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m7. Do you have any of the following medical conditions or risk factors?  \u001b[0m                                       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Known heart disease, high blood pressure, high cholesterol, or diabetes  \u001b[0m                                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Smoking history  \u001b[0m                                                                                          \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m   • Family history of early heart disease (under age 55 in a close male relative, under age 65 in a close \u001b[0m     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mfemale relative)  \u001b[0m                                                                                              \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mOnce I have this information, I can better advise you whether you need to head to the ER now, see a doctor \u001b[0m     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35msoon, or manage this at home.\u001b[0m                                                                                   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -691,86 +772,15 @@
     {
      "data": {
       "text/html": [
-       "
╭───────────────────────────────────────────────────── Input ─────────────────────────────────────────────────────╮\n",
-       "│ like 2 days ago my chest started feeling weird kinda like a tight pressure it’s not super painful and i dont    │\n",
-       "│ have shortness of breath or anything but i keep freaking out thinking maybe i should go to the ER or something  │\n",
-       "│ wtf???                                                                                                          │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
-       "
\n"
-      ],
-      "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m────────────────────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mInput\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m────────────────────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mlike 2 days ago my chest started feeling weird kinda like a tight pressure it’s not super painful and i dont \u001b[0m\u001b[47m  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mhave shortness of breath or anything but i keep freaking out thinking maybe i should go to the ER or something \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mwtf???\u001b[0m\u001b[47m                                                                                                         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "
╭────────────────────────────────────── Output (original model completion) ───────────────────────────────────────╮\n",
-       "│ I'm sorry you're feeling this way—chest discomfort can be really scary. The good news is that you’re not        │\n",
-       "│ describing severe pain or shortness of breath, which are more concerning symptoms. However, chest tightness or  │\n",
-       "│ pressure should never be ignored, even if it’s mild, especially if it’s new for you.                            │\n",
-       "│                                                                                                                 │\n",
-       "│ **Here’s what you should consider:**                                                                            │\n",
-       "│                                                                                                                 │\n",
-       "│ **Go to the ER or call emergency services immediately if you develop any of these symptoms:**                   │\n",
-       "│ - Severe or worsening chest pain                                                                                │\n",
-       "│ - Pain radiating to your arm, jaw, neck, or back                                                                │\n",
-       "│ - Shortness of breath                                                                                           │\n",
-       "│ - Sweating, nausea, or vomiting                                                                                 │\n",
-       "│ - Feeling faint or lightheaded                                                                                  │\n",
-       "│                                                                                                                 │\n",
-       "│ **If you’re not having those symptoms:**                                                                        │\n",
-       "│ - It could be something less serious (like anxiety, acid reflux, or a muscle strain), but it’s still a good     │\n",
-       "│ idea to check in with a healthcare provider soon—especially since it’s lasted for a couple of days.             │\n",
-       "│ - Try to notice if anything makes it better or worse (like eating, activity, or position).                      │\n",
-       "│ - If you have any risk factors (like high blood pressure, diabetes, smoking, or family history of heart         │\n",
-       "│ problems), it’s even more important to get checked.                                                             │\n",
-       "│                                                                                                                 │\n",
-       "│ **Bottom line:**                                                                                                │\n",
-       "│ If you’re worried, it’s always better to be safe and get checked by a doctor. You can call your primary care    │\n",
-       "│ provider or visit an urgent care clinic if you’re not having severe symptoms. If at any point you feel worse,   │\n",
-       "│ get emergency help right away.                                                                                  │\n",
-       "│                                                                                                                 │\n",
-       "│ Would you like help deciding where to go or what to say to a doctor?                                            │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭───────────────────────────────────────────────────── Input ─────────────────────────────────────────────────────╮\n",
+       "│ I'm bruising a lot on my arms and legs and it's starting to worry me, do you think it could be a blood disorder │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (original model completion)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m──────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mI'm sorry you're feeling this way—chest discomfort can be really scary. The good news is that you’re not \u001b[0m\u001b[47m      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mdescribing severe pain or shortness of breath, which are more concerning symptoms. However, chest tightness or \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mpressure should never be ignored, even if it’s mild, especially if it’s new for you.\u001b[0m\u001b[47m                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m**Here’s what you should consider:**\u001b[0m\u001b[47m                                                                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m**Go to the ER or call emergency services immediately if you develop any of these symptoms:**  \u001b[0m\u001b[47m                \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Severe or worsening chest pain  \u001b[0m\u001b[47m                                                                             \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Pain radiating to your arm, jaw, neck, or back  \u001b[0m\u001b[47m                                                             \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Shortness of breath  \u001b[0m\u001b[47m                                                                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Sweating, nausea, or vomiting  \u001b[0m\u001b[47m                                                                              \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Feeling faint or lightheaded\u001b[0m\u001b[47m                                                                                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m**If you’re not having those symptoms:**\u001b[0m\u001b[47m                                                                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- It could be something less serious (like anxiety, acid reflux, or a muscle strain), but it’s still a good \u001b[0m\u001b[47m   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47midea to check in with a healthcare provider soon—especially since it’s lasted for a couple of days.\u001b[0m\u001b[47m            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Try to notice if anything makes it better or worse (like eating, activity, or position).\u001b[0m\u001b[47m                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- If you have any risk factors (like high blood pressure, diabetes, smoking, or family history of heart \u001b[0m\u001b[47m       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mproblems), it’s even more important to get checked.\u001b[0m\u001b[47m                                                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m**Bottom line:**  \u001b[0m\u001b[47m                                                                                             \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mIf you’re worried, it’s always better to be safe and get checked by a doctor. You can call your primary care \u001b[0m\u001b[47m  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mprovider or visit an urgent care clinic if you’re not having severe symptoms. If at any point you feel worse, \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mget emergency help right away.\u001b[0m\u001b[47m                                                                                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mWould you like help deciding where to go or what to say to a doctor?\u001b[0m\u001b[47m                                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m────────────────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mInput\u001b[0m\u001b[34m \u001b[0m\u001b[34m────────────────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;36mI'm bruising a lot on my arms and legs and it's starting to worry me, do you think it could be a blood disorder\u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -779,69 +789,79 @@
     {
      "data": {
       "text/html": [
-       "
╭───────────────────────────────────────── Output (base reasoning model) ─────────────────────────────────────────╮\n",
-       "│ I’m not a doctor, but chest discomfort—especially if it’s new and lasting for days—really deserves at least a   │\n",
-       "│ phone-call with a medical professional. Here’s a rough guide to help you decide what to do next:                │\n",
-       "│                                                                                                                 │\n",
-       "│ 1. Watch for “red-flag” symptoms. If you develop any of these, call 911 or go straight to the ER:               │\n",
-       "│    • Pain or pressure that spreads to your jaw, arm, shoulder or back                                           │\n",
-       "│    • Crushing or squeezing sensation that doesn’t ease with rest                                                │\n",
-       "│    • Shortness of breath, dizziness, lightheadedness or fainting                                                │\n",
-       "│    • Cold sweat, nausea or feeling “clammy”                                                                     │\n",
-       "│    • Rapid or irregular heartbeat                                                                               │\n",
-       "│                                                                                                                 │\n",
-       "│ 2. If you’re not having those red-flag signs but the pressure is still there after two days:                    │\n",
-       "│    • Call your primary-care doctor or an urgent-care clinic. Explain exactly what you feel, how long it’s       │\n",
-       "│ lasted, and what makes it better or worse.                                                                      │\n",
-       "│    • They may ask you to come in for a quick exam, an EKG or blood tests to rule out heart issues.              │\n",
-       "│                                                                                                                 │\n",
-       "│ 3. Possible non-cardiac causes (which only a clinician can sort out):                                           │\n",
-       "│    • Musculoskeletal strain (costochondritis) – often tender to the touch or when you move your chest/arms      │\n",
-       "│    • Acid reflux or esophageal spasm – sometimes burning, worse after eating or lying down                      │\n",
-       "│    • Anxiety or panic – chest tightness often comes with rapid breathing, palpitations, a sense of dread        │\n",
-       "│                                                                                                                 │\n",
-       "│ 4. In the meantime:                                                                                             │\n",
-       "│    • Keep a symptom diary (when it starts, what you were doing, other sensations you notice)                    │\n",
-       "│    • Try simple stretches or an over-the-counter NSAID (like ibuprofen) if you suspect a pulled muscle          │\n",
-       "│    • Practice slow, deep breathing or relaxation exercises if stress feels like a trigger                       │\n",
-       "│                                                                                                                 │\n",
-       "│ Bottom line: chest pressure isn’t something to tough out on your own. If it’s mild and truly not worsening,     │\n",
-       "│ booking a same-day appointment or going to urgent care is reasonable. But the moment you feel it’s              │\n",
-       "│ intensifying, spreading, or accompanied by any red-flag signs, call emergency services. Better safe than sorry. │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭──────────────────────────────────────────── Output (original model) ────────────────────────────────────────────╮\n",
+       "│ I'm not a doctor, but I can give you some information that might help you decide what to do next.               │\n",
+       "│                                                                                                                 │\n",
+       "│ Bruising easily can have many causes—some are minor, but others can be more serious. Common reasons for         │\n",
+       "│ increased bruising include:                                                                                     │\n",
+       "│                                                                                                                 │\n",
+       "│ - **Aging:** Skin and blood vessels become more fragile.                                                        │\n",
+       "│ - **Medications:** Blood thinners (like aspirin, warfarin), steroids, and some supplements can increase         │\n",
+       "│ bruising.                                                                                                       │\n",
+       "│ - **Nutritional deficiencies:** Low levels of vitamin C, vitamin K, or certain proteins.                        │\n",
+       "│ - **Blood disorders:** Conditions like thrombocytopenia (low platelets), hemophilia, or other clotting          │\n",
+       "│ disorders.                                                                                                      │\n",
+       "│ - **Other medical conditions:** Liver disease, certain infections, or autoimmune diseases.                      │\n",
+       "│                                                                                                                 │\n",
+       "│ **When to be concerned:**                                                                                       │\n",
+       "│ You should see a healthcare provider soon if you notice any of the following:                                   │\n",
+       "│ - Bruises appearing without any known injury.                                                                   │\n",
+       "│ - Large or painful bruises.                                                                                     │\n",
+       "│ - Bruising accompanied by other symptoms (like frequent nosebleeds, bleeding gums, blood in urine/stool,        │\n",
+       "│ fatigue, or unexplained weight loss).                                                                           │\n",
+       "│ - Bruising that seems to be getting worse or spreading.                                                         │\n",
+       "│                                                                                                                 │\n",
+       "│ **What you can do now:**                                                                                        │\n",
+       "│ - Make a note of any new medications or supplements you’ve started.                                             │\n",
+       "│ - Keep track of how many bruises you get and where they appear.                                                 │\n",
+       "│ - Schedule an appointment with your doctor to discuss your symptoms.                                            │\n",
+       "│                                                                                                                 │\n",
+       "│ While it could be something minor, it’s important to get checked out to rule out any serious causes, including  │\n",
+       "│ blood disorders.                                                                                                │\n",
+       "│                                                                                                                 │\n",
+       "│ If you develop severe symptoms, such as difficulty breathing, severe headache, or uncontrolled bleeding, seek   │\n",
+       "│ emergency care immediately.                                                                                     │\n",
+       "│                                                                                                                 │\n",
+       "│ Would you like more information about what to expect at your doctor's visit or how to prepare?                  │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m────────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (base reasoning model)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m────────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mI’m not a doctor, but chest discomfort—especially if it’s new and lasting for days—really deserves at least a \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mphone-call with a medical professional. Here’s a rough guide to help you decide what to do next:\u001b[0m\u001b[47m               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m1. Watch for “red-flag” symptoms. If you develop any of these, call 911 or go straight to the ER:  \u001b[0m\u001b[47m            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Pain or pressure that spreads to your jaw, arm, shoulder or back  \u001b[0m\u001b[47m                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Crushing or squeezing sensation that doesn’t ease with rest  \u001b[0m\u001b[47m                                             \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Shortness of breath, dizziness, lightheadedness or fainting  \u001b[0m\u001b[47m                                             \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Cold sweat, nausea or feeling “clammy”  \u001b[0m\u001b[47m                                                                  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Rapid or irregular heartbeat  \u001b[0m\u001b[47m                                                                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m2. If you’re not having those red-flag signs but the pressure is still there after two days:  \u001b[0m\u001b[47m                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Call your primary-care doctor or an urgent-care clinic. Explain exactly what you feel, how long it’s \u001b[0m\u001b[47m     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mlasted, and what makes it better or worse.  \u001b[0m\u001b[47m                                                                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • They may ask you to come in for a quick exam, an EKG or blood tests to rule out heart issues.  \u001b[0m\u001b[47m           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m3. Possible non-cardiac causes (which only a clinician can sort out):  \u001b[0m\u001b[47m                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Musculoskeletal strain (costochondritis) – often tender to the touch or when you move your chest/arms  \u001b[0m\u001b[47m   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Acid reflux or esophageal spasm – sometimes burning, worse after eating or lying down  \u001b[0m\u001b[47m                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Anxiety or panic – chest tightness often comes with rapid breathing, palpitations, a sense of dread  \u001b[0m\u001b[47m     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m4. In the meantime:  \u001b[0m\u001b[47m                                                                                          \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Keep a symptom diary (when it starts, what you were doing, other sensations you notice)  \u001b[0m\u001b[47m                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Try simple stretches or an over-the-counter NSAID (like ibuprofen) if you suspect a pulled muscle  \u001b[0m\u001b[47m       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m   • Practice slow, deep breathing or relaxation exercises if stress feels like a trigger  \u001b[0m\u001b[47m                    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mBottom line: chest pressure isn’t something to tough out on your own. If it’s mild and truly not worsening, \u001b[0m\u001b[47m   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mbooking a same-day appointment or going to urgent care is reasonable. But the moment you feel it’s \u001b[0m\u001b[47m            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mintensifying, spreading, or accompanied by any red-flag signs, call emergency services. Better safe than sorry.\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m───────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mOutput (original model)\u001b[0m\u001b[34m \u001b[0m\u001b[34m───────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mI'm not a doctor, but I can give you some information that might help you decide what to do next.\u001b[0m               \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mBruising easily can have many causes—some are minor, but others can be more serious. Common reasons for \u001b[0m        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mincreased bruising include:\u001b[0m                                                                                     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Aging:** Skin and blood vessels become more fragile.\u001b[0m                                                        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Medications:** Blood thinners (like aspirin, warfarin), steroids, and some supplements can increase \u001b[0m        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mbruising.\u001b[0m                                                                                                       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Nutritional deficiencies:** Low levels of vitamin C, vitamin K, or certain proteins.\u001b[0m                        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Blood disorders:** Conditions like thrombocytopenia (low platelets), hemophilia, or other clotting \u001b[0m         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mdisorders.\u001b[0m                                                                                                      \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Other medical conditions:** Liver disease, certain infections, or autoimmune diseases.\u001b[0m                      \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**When to be concerned:**  \u001b[0m                                                                                     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mYou should see a healthcare provider soon if you notice any of the following:\u001b[0m                                   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Bruises appearing without any known injury.\u001b[0m                                                                   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Large or painful bruises.\u001b[0m                                                                                     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Bruising accompanied by other symptoms (like frequent nosebleeds, bleeding gums, blood in urine/stool, \u001b[0m       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mfatigue, or unexplained weight loss).\u001b[0m                                                                           \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Bruising that seems to be getting worse or spreading.\u001b[0m                                                         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**What you can do now:**\u001b[0m                                                                                        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Make a note of any new medications or supplements you’ve started.\u001b[0m                                             \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Keep track of how many bruises you get and where they appear.\u001b[0m                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Schedule an appointment with your doctor to discuss your symptoms.\u001b[0m                                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mWhile it could be something minor, it’s important to get checked out to rule out any serious causes, including \u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mblood disorders.\u001b[0m                                                                                                \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mIf you develop severe symptoms, such as difficulty breathing, severe headache, or uncontrolled bleeding, seek \u001b[0m  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33memergency care immediately.\u001b[0m                                                                                     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mWould you like more information about what to expect at your doctor's visit or how to prepare?\u001b[0m                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -850,47 +870,43 @@
     {
      "data": {
       "text/html": [
-       "
╭────────────────────────────────────── Output (fine-tuned reasoning model) ──────────────────────────────────────╮\n",
-       "│ I’m sorry you’re going through that—let’s see if we can figure out what’s going on. A few quick questions:      │\n",
-       "│                                                                                                                 │\n",
-       "│ 1. How old are you, and do you have any medical conditions (high blood pressure, high cholesterol, diabetes,    │\n",
-       "│ asthma, anxiety, etc.)?                                                                                         │\n",
-       "│ 2. Exactly where do you feel the pressure—center of your chest, left side, under your breastbone?               │\n",
-       "│ 3. Can you describe it—dull ache, sharp stabbing, squeezing, burning?                                           │\n",
-       "│ 4. Does it ever radiate—to your arm, jaw, back?                                                                 │\n",
-       "│ 5. What were you doing when it started, and does any activity (walking, climbing stairs, lifting) make it worse │\n",
-       "│ or better?                                                                                                      │\n",
-       "│ 6. Does it change if you take a deep breath, cough, press on your chest, or change position?                    │\n",
-       "│ 7. Any other symptoms—palpitations (heart racing/skipping), dizziness, sweating, nausea, cough, heartburn, or   │\n",
-       "│ leg swelling?                                                                                                   │\n",
-       "│ 8. Have you noticed any triggers or stressors recently?                                                         │\n",
-       "│                                                                                                                 │\n",
-       "│ If at any point you feel it’s crushing, comes on suddenly, or you develop sweating, nausea, shortness of        │\n",
-       "│ breath, lightheadedness, or pain radiating to your arm/jaw—even briefly—call 911 or go to the nearest ER right  │\n",
-       "│ away. Let me know the answers and we’ll go from there.                                                          │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭─────────────────────────────────────────── Output (fine-tuned model) ───────────────────────────────────────────╮\n",
+       "│ I’m sorry you’re experiencing this—it can certainly be concerning. To help figure out whether your bruising     │\n",
+       "│ might be from a blood‑related issue or something else, could you tell me a bit more about what you’re seeing    │\n",
+       "│ and your health history? For example:                                                                           │\n",
+       "│ 1. When did you first notice these bruises, and how often are new ones appearing?                               │\n",
+       "│ 2. Do they occur after any bump or injury, or do they seem to arise spontaneously?                              │\n",
+       "│ 3. What do the bruises look like in terms of size, color, and number?                                           │\n",
+       "│ 4. Have you started or stopped any medications or supplements recently (especially blood thinners, NSAIDs,      │\n",
+       "│ aspirin, steroids, or herbal supplements)?                                                                      │\n",
+       "│ 5. Do you have any other bleeding symptoms—nosebleeds, bleeding gums, unusually heavy menstrual periods, blood  │\n",
+       "│ in stool or urine?                                                                                              │\n",
+       "│ 6. Have you experienced other symptoms such as fatigue, fever, weight loss, or night sweats?                    │\n",
+       "│ 7. Is there any family history of bruising easily, bleeding disorders, or clotting abnormalities?               │\n",
+       "│                                                                                                                 │\n",
+       "│ With that information, I can better suggest whether it’s appropriate to check things like your platelet count,  │\n",
+       "│ clotting factors, or other tests, or whether it might be related to something less worrisome.                   │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (fine-tuned reasoning model)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mI’m sorry you’re going through that—let’s see if we can figure out what’s going on. A few quick questions:\u001b[0m\u001b[47m     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m1. How old are you, and do you have any medical conditions (high blood pressure, high cholesterol, diabetes, \u001b[0m\u001b[47m  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47masthma, anxiety, etc.)?\u001b[0m\u001b[47m                                                                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m2. Exactly where do you feel the pressure—center of your chest, left side, under your breastbone?\u001b[0m\u001b[47m              \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m3. Can you describe it—dull ache, sharp stabbing, squeezing, burning?\u001b[0m\u001b[47m                                          \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m4. Does it ever radiate—to your arm, jaw, back?\u001b[0m\u001b[47m                                                                \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m5. What were you doing when it started, and does any activity (walking, climbing stairs, lifting) make it worse\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mor better?\u001b[0m\u001b[47m                                                                                                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m6. Does it change if you take a deep breath, cough, press on your chest, or change position?\u001b[0m\u001b[47m                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m7. Any other symptoms—palpitations (heart racing/skipping), dizziness, sweating, nausea, cough, heartburn, or \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mleg swelling?\u001b[0m\u001b[47m                                                                                                  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m8. Have you noticed any triggers or stressors recently?\u001b[0m\u001b[47m                                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mIf at any point you feel it’s crushing, comes on suddenly, or you develop sweating, nausea, shortness of \u001b[0m\u001b[47m      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mbreath, lightheadedness, or pain radiating to your arm/jaw—even briefly—call 911 or go to the nearest ER right \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47maway. Let me know the answers and we’ll go from there.\u001b[0m\u001b[47m                                                         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m──────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mOutput (fine-tuned model)\u001b[0m\u001b[34m \u001b[0m\u001b[34m──────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mI’m sorry you’re experiencing this—it can certainly be concerning. To help figure out whether your bruising \u001b[0m    \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mmight be from a blood‑related issue or something else, could you tell me a bit more about what you’re seeing \u001b[0m   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mand your health history? For example:  \u001b[0m                                                                         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m1. When did you first notice these bruises, and how often are new ones appearing?  \u001b[0m                             \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m2. Do they occur after any bump or injury, or do they seem to arise spontaneously?  \u001b[0m                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m3. What do the bruises look like in terms of size, color, and number?  \u001b[0m                                         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m4. Have you started or stopped any medications or supplements recently (especially blood thinners, NSAIDs, \u001b[0m     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35maspirin, steroids, or herbal supplements)?  \u001b[0m                                                                    \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m5. Do you have any other bleeding symptoms—nosebleeds, bleeding gums, unusually heavy menstrual periods, blood \u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35min stool or urine?  \u001b[0m                                                                                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m6. Have you experienced other symptoms such as fatigue, fever, weight loss, or night sweats?  \u001b[0m                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m7. Is there any family history of bruising easily, bleeding disorders, or clotting abnormalities?  \u001b[0m             \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mWith that information, I can better suggest whether it’s appropriate to check things like your platelet count, \u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mclotting factors, or other tests, or whether it might be related to something less worrisome.\u001b[0m                   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -916,96 +932,15 @@
     {
      "data": {
       "text/html": [
-       "
╭───────────────────────────────────────────────────── Input ─────────────────────────────────────────────────────╮\n",
-       "│ I'm bruising a lot on my arms and legs and it's starting to worry me, do you think it could be a blood disorder │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
-       "
\n"
-      ],
-      "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m────────────────────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mInput\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m────────────────────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[30;47mI'm bruising a lot on my arms and legs and it's starting to worry me, do you think it could be a blood disorder\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "data": {
-      "text/html": [
-       "
╭────────────────────────────────────── Output (original model completion) ───────────────────────────────────────╮\n",
-       "│ I'm not a doctor, but I can give you some information that might help you decide what to do next.               │\n",
-       "│                                                                                                                 │\n",
-       "│ Bruising easily can have many causes—some are minor, but others can be more serious. Common reasons for         │\n",
-       "│ increased bruising include:                                                                                     │\n",
-       "│                                                                                                                 │\n",
-       "│ - **Aging:** Skin and blood vessels become more fragile.                                                        │\n",
-       "│ - **Medications:** Blood thinners (like aspirin, warfarin), steroids, and some supplements can increase         │\n",
-       "│ bruising.                                                                                                       │\n",
-       "│ - **Nutritional deficiencies:** Low levels of vitamin C, vitamin K, or certain proteins.                        │\n",
-       "│ - **Blood disorders:** Conditions like thrombocytopenia (low platelets), hemophilia, or other clotting          │\n",
-       "│ disorders.                                                                                                      │\n",
-       "│ - **Other medical conditions:** Liver disease, certain infections, or autoimmune diseases.                      │\n",
-       "│                                                                                                                 │\n",
-       "│ **When to be concerned:**                                                                                       │\n",
-       "│ You should see a healthcare provider soon if you notice any of the following:                                   │\n",
-       "│ - Bruises appearing without any known injury.                                                                   │\n",
-       "│ - Large or painful bruises.                                                                                     │\n",
-       "│ - Bruising accompanied by other symptoms (like frequent nosebleeds, bleeding gums, blood in urine/stool,        │\n",
-       "│ fatigue, or unexplained weight loss).                                                                           │\n",
-       "│ - Bruising that seems to be getting worse or spreading.                                                         │\n",
-       "│                                                                                                                 │\n",
-       "│ **What you can do now:**                                                                                        │\n",
-       "│ - Make a note of any new medications or supplements you’ve started.                                             │\n",
-       "│ - Keep track of how many bruises you get and where they appear.                                                 │\n",
-       "│ - Schedule an appointment with your doctor to discuss your symptoms.                                            │\n",
-       "│                                                                                                                 │\n",
-       "│ While it could be something minor, it’s important to get checked out to rule out any serious causes, including  │\n",
-       "│ blood disorders.                                                                                                │\n",
-       "│                                                                                                                 │\n",
-       "│ If you develop severe symptoms, such as difficulty breathing, severe headache, or uncontrolled bleeding, seek   │\n",
-       "│ emergency care immediately.                                                                                     │\n",
-       "│                                                                                                                 │\n",
-       "│ Would you like more information about what to expect at your doctor's visit or how to prepare?                  │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭───────────────────────────────────────────────────── Input ─────────────────────────────────────────────────────╮\n",
+       "│ adult routine cholesterol screening guidelines                                                                  │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (original model completion)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m──────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mI'm not a doctor, but I can give you some information that might help you decide what to do next.\u001b[0m\u001b[47m              \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mBruising easily can have many causes—some are minor, but others can be more serious. Common reasons for \u001b[0m\u001b[47m       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mincreased bruising include:\u001b[0m\u001b[47m                                                                                    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- **Aging:** Skin and blood vessels become more fragile.\u001b[0m\u001b[47m                                                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- **Medications:** Blood thinners (like aspirin, warfarin), steroids, and some supplements can increase \u001b[0m\u001b[47m       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mbruising.\u001b[0m\u001b[47m                                                                                                      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- **Nutritional deficiencies:** Low levels of vitamin C, vitamin K, or certain proteins.\u001b[0m\u001b[47m                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- **Blood disorders:** Conditions like thrombocytopenia (low platelets), hemophilia, or other clotting \u001b[0m\u001b[47m        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mdisorders.\u001b[0m\u001b[47m                                                                                                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- **Other medical conditions:** Liver disease, certain infections, or autoimmune diseases.\u001b[0m\u001b[47m                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m**When to be concerned:**  \u001b[0m\u001b[47m                                                                                    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mYou should see a healthcare provider soon if you notice any of the following:\u001b[0m\u001b[47m                                  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Bruises appearing without any known injury.\u001b[0m\u001b[47m                                                                  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Large or painful bruises.\u001b[0m\u001b[47m                                                                                    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Bruising accompanied by other symptoms (like frequent nosebleeds, bleeding gums, blood in urine/stool, \u001b[0m\u001b[47m      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mfatigue, or unexplained weight loss).\u001b[0m\u001b[47m                                                                          \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Bruising that seems to be getting worse or spreading.\u001b[0m\u001b[47m                                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m**What you can do now:**\u001b[0m\u001b[47m                                                                                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Make a note of any new medications or supplements you’ve started.\u001b[0m\u001b[47m                                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Keep track of how many bruises you get and where they appear.\u001b[0m\u001b[47m                                                \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47m- Schedule an appointment with your doctor to discuss your symptoms.\u001b[0m\u001b[47m                                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mWhile it could be something minor, it’s important to get checked out to rule out any serious causes, including \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mblood disorders.\u001b[0m\u001b[47m                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mIf you develop severe symptoms, such as difficulty breathing, severe headache, or uncontrolled bleeding, seek \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47memergency care immediately.\u001b[0m\u001b[47m                                                                                    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[34;47mWould you like more information about what to expect at your doctor's visit or how to prepare?\u001b[0m\u001b[47m                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m────────────────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mInput\u001b[0m\u001b[34m \u001b[0m\u001b[34m────────────────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;36madult routine cholesterol screening guidelines\u001b[0m                                                                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -1014,123 +949,129 @@
     {
      "data": {
       "text/html": [
-       "
╭───────────────────────────────────────── Output (base reasoning model) ─────────────────────────────────────────╮\n",
-       "│ I’m not a doctor, but I can share some information that may help you decide whether to seek medical attention   │\n",
-       "│ and what questions to ask your provider.                                                                        │\n",
-       "│                                                                                                                 │\n",
-       "│ 1. Common causes of easy bruising                                                                               │\n",
-       "│   • Minor bumps or knocks you don’t remember                                                                    │\n",
-       "│   • Medications or supplements that thin your blood (aspirin, NSAIDs, fish oil, ginko, warfarin, direct oral    │\n",
-       "│ anticoagulants)                                                                                                 │\n",
-       "│   • Aging skin and blood vessels (the skin thins and capillaries become more fragile)                           │\n",
-       "│                                                                                                                 │\n",
-       "│ 2. Possible medical causes to discuss with your doctor                                                          │\n",
-       "│   a) Platelet problems                                                                                          │\n",
-       "│   – Low platelet count (thrombocytopenia) from immune causes (e.g. ITP), infections, certain medications        │\n",
-       "│   – Platelet dysfunction (e.g. von Willebrand disease, inherited or acquired disorders)                         │\n",
-       "│   b) Clotting-factor deficiencies                                                                               │\n",
-       "│   – Hemophilia A or B (rare in women)                                                                           │\n",
-       "│   – Vitamin K deficiency (malabsorption, certain antibiotics)                                                   │\n",
-       "│   c) Liver disease                                                                                              │\n",
-       "│   – The liver makes many clotting factors; if it’s not working well you can bruise more easily                  │\n",
-       "│   d) Vascular fragility                                                                                         │\n",
-       "│   – Vasculitis or connective-tissue disorders that weaken blood vessel walls                                    │\n",
-       "│   e) Nutritional deficiencies                                                                                   │\n",
-       "│   – Vitamin C (scurvy), vitamin K, or protein deficiency                                                        │\n",
-       "│                                                                                                                 │\n",
-       "│ 3. Red-flag symptoms that warrant prompt evaluation                                                             │\n",
-       "│   • Bruises appearing without any remembered bump or injury                                                     │\n",
-       "│   • Large “hematomas” (deep, painful swellings under the skin)                                                  │\n",
-       "│   • Small red/purple flat spots (petechiae) or pinpoint bleeds around the hair follicles                        │\n",
-       "│   • Bleeding gums, frequent nosebleeds, blood in stool or urine, unusually heavy menstrual bleeding             │\n",
-       "│   • Unexplained weight loss, night sweats, fevers (could point toward an underlying illness)                    │\n",
-       "│                                                                                                                 │\n",
-       "│ 4. What you can do now                                                                                          │\n",
-       "│   • Keep a “bruise diary”: note when and where each bruise appears, how big it is, and any associated symptoms. │\n",
-       "│   • Review any medications or supplements you take—ask your pharmacist or provider if they affect bleeding      │\n",
-       "│ risk.                                                                                                           │\n",
-       "│   • Make sure your diet includes adequate protein, vitamin C (citrus fruits, berries, peppers), and vitamin K   │\n",
-       "│ (leafy greens).                                                                                                 │\n",
-       "│                                                                                                                 │\n",
-       "│ 5. When to see a doctor                                                                                         │\n",
-       "│   • Bruising is significantly more frequent or severe than you can explain by bumps and knocks                  │\n",
-       "│   • You have any of the red-flag symptoms above                                                                 │\n",
-       "│   • You’re on a blood thinner and your bruising seems out of proportion                                         │\n",
-       "│                                                                                                                 │\n",
-       "│ 6. What your doctor may do                                                                                      │\n",
-       "│   • Physical exam (skin inspection, signs of liver disease, spleen enlargement)                                 │\n",
-       "│   • Blood tests:                                                                                                │\n",
-       "│     – Complete blood count (CBC) with platelet count                                                            │\n",
-       "│     – Coagulation panel (PT/INR, aPTT)                                                                          │\n",
-       "│     – Liver-function tests                                                                                      │\n",
-       "│     – Specific factor levels or von Willebrand assay if indicated                                               │\n",
-       "│   • Referral to a hematologist if an inherited or serious acquired disorder is suspected                        │\n",
-       "│                                                                                                                 │\n",
-       "│ Bottom line: occasional bruising is common, especially if you bump into things or take mild blood thinners. But │\n",
-       "│ if your bruising is frequent, spontaneous, or accompanied by other bleeding symptoms, you should get evaluated. │\n",
-       "│ A primary-care doctor can order simple blood tests to rule out the most common disorders and guide you to any   │\n",
-       "│ needed specialist care.                                                                                         │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭──────────────────────────────────────────── Output (original model) ────────────────────────────────────────────╮\n",
+       "│ Here is a summary of current guidelines for adult routine cholesterol screening:                                │\n",
+       "│                                                                                                                 │\n",
+       "│ **General Recommendations:**                                                                                    │\n",
+       "│                                                                                                                 │\n",
+       "│ - **All adults aged 20 years or older**: The American Heart Association (AHA), American College of Cardiology   │\n",
+       "│ (ACC), and U.S. Preventive Services Task Force (USPSTF) recommend routine cholesterol screening starting at age │\n",
+       "│ 20, with repeat testing every 4–6 years if risk remains low.                                                    │\n",
+       "│                                                                                                                 │\n",
+       "│ **More Specific Guidelines:**                                                                                   │\n",
+       "│                                                                                                                 │\n",
+       "│ ### U.S. Preventive Services Task Force (USPSTF) (2016):                                                        │\n",
+       "│                                                                                                                 │\n",
+       "│ - **Adults aged 40–75**: Strongly recommend screening.                                                          │\n",
+       "│ - **Adults aged 20–39**: Consider screening if they have risk factors for cardiovascular disease (e.g.,         │\n",
+       "│ diabetes, hypertension, family history of early heart disease, smoking, obesity).                               │\n",
+       "│ - **Frequency**: Every 4–6 years for low-risk individuals; more frequently if risk factors are present.         │\n",
+       "│                                                                                                                 │\n",
+       "│ ### American College of Cardiology (ACC)/American Heart Association (AHA) (2018):                               │\n",
+       "│                                                                                                                 │\n",
+       "│ - **Adults aged 20 and older**: Assess cholesterol as part of cardiovascular risk assessment every 4–6 years.   │\n",
+       "│ - **More frequent testing**: For those with risk factors (e.g., diabetes, hypertension, family history,         │\n",
+       "│ obesity) or those on cholesterol-lowering therapy.                                                              │\n",
+       "│                                                                                                                 │\n",
+       "│ ### National Lipid Association (NLA):                                                                           │\n",
+       "│                                                                                                                 │\n",
+       "│ - **All adults 20 years and older**: Lipid profile at least every 5 years.                                      │\n",
+       "│ - **Earlier and/or more frequent testing**: If risk factors or family history of premature atherosclerotic      │\n",
+       "│ cardiovascular disease (ASCVD).                                                                                 │\n",
+       "│                                                                                                                 │\n",
+       "│ **What is measured?**                                                                                           │\n",
+       "│ - A standard fasting or non-fasting lipid panel measures:                                                       │\n",
+       "│   - Total cholesterol                                                                                           │\n",
+       "│   - LDL cholesterol (\"bad\")                                                                                     │\n",
+       "│   - HDL cholesterol (\"good\")                                                                                    │\n",
+       "│   - Triglycerides                                                                                               │\n",
+       "│                                                                                                                 │\n",
+       "│ **Summary Table:**                                                                                              │\n",
+       "│                                                                                                                 │\n",
+       "│ | Age Group         | Routine Screening? | Frequency      | More Frequent If...                |                │\n",
+       "│ |-------------------|-------------------|---------------|------------------------------------|                  │\n",
+       "│ | 20–39 years       | Consider if risk  | 4–6 years     | Risk factors present               |                  │\n",
+       "│ | 40–75 years       | Yes               | 4–6 years     | Risk factors or on therapy         |                  │\n",
+       "│ | >75 years         | Individualized    | Case-by-case  | Based on overall health/risk       |                  │\n",
+       "│                                                                                                                 │\n",
+       "│ **Key Risk Factors:**                                                                                           │\n",
+       "│ - Diabetes                                                                                                      │\n",
+       "│ - Hypertension                                                                                                  │\n",
+       "│ - Smoking                                                                                                       │\n",
+       "│ - Family history of early heart disease                                                                         │\n",
+       "│ - Obesity                                                                                                       │\n",
+       "│                                                                                                                 │\n",
+       "│ **References:**                                                                                                 │\n",
+       "│ - 2018 ACC/AHA Guideline on the Management of Blood Cholesterol                                                 │\n",
+       "│ - USPSTF Recommendation Statement (2016)                                                                        │\n",
+       "│                                                                                                                 │\n",
+       "│ **Note:** These are general recommendations. Screening intervals and starting age may be adjusted based on      │\n",
+       "│ individual risk factors and clinical judgment. Always consult with a healthcare provider for personalized       │\n",
+       "│ advice.                                                                                                         │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m────────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (base reasoning model)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m────────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mI’m not a doctor, but I can share some information that may help you decide whether to seek medical attention \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mand what questions to ask your provider.\u001b[0m\u001b[47m                                                                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m1. Common causes of easy bruising  \u001b[0m\u001b[47m                                                                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Minor bumps or knocks you don’t remember  \u001b[0m\u001b[47m                                                                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Medications or supplements that thin your blood (aspirin, NSAIDs, fish oil, ginko, warfarin, direct oral \u001b[0m\u001b[47m  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47manticoagulants)  \u001b[0m\u001b[47m                                                                                              \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Aging skin and blood vessels (the skin thins and capillaries become more fragile)  \u001b[0m\u001b[47m                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m2. Possible medical causes to discuss with your doctor  \u001b[0m\u001b[47m                                                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  a) Platelet problems  \u001b[0m\u001b[47m                                                                                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  – Low platelet count (thrombocytopenia) from immune causes (e.g. ITP), infections, certain medications  \u001b[0m\u001b[47m     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  – Platelet dysfunction (e.g. von Willebrand disease, inherited or acquired disorders)  \u001b[0m\u001b[47m                      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  b) Clotting-factor deficiencies  \u001b[0m\u001b[47m                                                                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  – Hemophilia A or B (rare in women)  \u001b[0m\u001b[47m                                                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  – Vitamin K deficiency (malabsorption, certain antibiotics)  \u001b[0m\u001b[47m                                                \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  c) Liver disease  \u001b[0m\u001b[47m                                                                                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  – The liver makes many clotting factors; if it’s not working well you can bruise more easily  \u001b[0m\u001b[47m               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  d) Vascular fragility  \u001b[0m\u001b[47m                                                                                      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  – Vasculitis or connective-tissue disorders that weaken blood vessel walls  \u001b[0m\u001b[47m                                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  e) Nutritional deficiencies  \u001b[0m\u001b[47m                                                                                \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  – Vitamin C (scurvy), vitamin K, or protein deficiency  \u001b[0m\u001b[47m                                                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m3. Red-flag symptoms that warrant prompt evaluation  \u001b[0m\u001b[47m                                                          \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Bruises appearing without any remembered bump or injury  \u001b[0m\u001b[47m                                                  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Large “hematomas” (deep, painful swellings under the skin)  \u001b[0m\u001b[47m                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Small red/purple flat spots (petechiae) or pinpoint bleeds around the hair follicles  \u001b[0m\u001b[47m                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Bleeding gums, frequent nosebleeds, blood in stool or urine, unusually heavy menstrual bleeding  \u001b[0m\u001b[47m          \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Unexplained weight loss, night sweats, fevers (could point toward an underlying illness)  \u001b[0m\u001b[47m                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m4. What you can do now  \u001b[0m\u001b[47m                                                                                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Keep a “bruise diary”: note when and where each bruise appears, how big it is, and any associated symptoms.\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Review any medications or supplements you take—ask your pharmacist or provider if they affect bleeding \u001b[0m\u001b[47m    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mrisk.  \u001b[0m\u001b[47m                                                                                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Make sure your diet includes adequate protein, vitamin C (citrus fruits, berries, peppers), and vitamin K \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m(leafy greens).  \u001b[0m\u001b[47m                                                                                              \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m5. When to see a doctor  \u001b[0m\u001b[47m                                                                                      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Bruising is significantly more frequent or severe than you can explain by bumps and knocks  \u001b[0m\u001b[47m               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • You have any of the red-flag symptoms above  \u001b[0m\u001b[47m                                                              \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • You’re on a blood thinner and your bruising seems out of proportion  \u001b[0m\u001b[47m                                      \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m6. What your doctor may do  \u001b[0m\u001b[47m                                                                                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Physical exam (skin inspection, signs of liver disease, spleen enlargement)  \u001b[0m\u001b[47m                              \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Blood tests:  \u001b[0m\u001b[47m                                                                                             \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m    – Complete blood count (CBC) with platelet count  \u001b[0m\u001b[47m                                                         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m    – Coagulation panel (PT/INR, aPTT)  \u001b[0m\u001b[47m                                                                       \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m    – Liver-function tests  \u001b[0m\u001b[47m                                                                                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m    – Specific factor levels or von Willebrand assay if indicated  \u001b[0m\u001b[47m                                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47m  • Referral to a hematologist if an inherited or serious acquired disorder is suspected  \u001b[0m\u001b[47m                     \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mBottom line: occasional bruising is common, especially if you bump into things or take mild blood thinners. But\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mif your bruising is frequent, spontaneous, or accompanied by other bleeding symptoms, you should get evaluated.\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mA primary-care doctor can order simple blood tests to rule out the most common disorders and guide you to any \u001b[0m\u001b[47m \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[38;5;22;47mneeded specialist care.\u001b[0m\u001b[47m                                                                                        \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m───────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mOutput (original model)\u001b[0m\u001b[34m \u001b[0m\u001b[34m───────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mHere is a summary of current guidelines for adult routine cholesterol screening:\u001b[0m                                \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**General Recommendations:**\u001b[0m                                                                                    \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **All adults aged 20 years or older**: The American Heart Association (AHA), American College of Cardiology \u001b[0m  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m(ACC), and U.S. Preventive Services Task Force (USPSTF) recommend routine cholesterol screening starting at age\u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m20, with repeat testing every 4–6 years if risk remains low.\u001b[0m                                                    \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**More Specific Guidelines:**\u001b[0m                                                                                   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m### U.S. Preventive Services Task Force (USPSTF) (2016):\u001b[0m                                                        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Adults aged 40–75**: Strongly recommend screening.\u001b[0m                                                          \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Adults aged 20–39**: Consider screening if they have risk factors for cardiovascular disease (e.g., \u001b[0m        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mdiabetes, hypertension, family history of early heart disease, smoking, obesity).\u001b[0m                               \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Frequency**: Every 4–6 years for low-risk individuals; more frequently if risk factors are present.\u001b[0m         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m### American College of Cardiology (ACC)/American Heart Association (AHA) (2018):\u001b[0m                               \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Adults aged 20 and older**: Assess cholesterol as part of cardiovascular risk assessment every 4–6 years.\u001b[0m   \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **More frequent testing**: For those with risk factors (e.g., diabetes, hypertension, family history, \u001b[0m        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mobesity) or those on cholesterol-lowering therapy.\u001b[0m                                                              \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m### National Lipid Association (NLA):\u001b[0m                                                                           \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **All adults 20 years and older**: Lipid profile at least every 5 years.\u001b[0m                                      \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- **Earlier and/or more frequent testing**: If risk factors or family history of premature atherosclerotic \u001b[0m     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mcardiovascular disease (ASCVD).\u001b[0m                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**What is measured?**\u001b[0m                                                                                           \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- A standard fasting or non-fasting lipid panel measures:\u001b[0m                                                       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m  - Total cholesterol\u001b[0m                                                                                           \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m  - LDL cholesterol (\"bad\")\u001b[0m                                                                                     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m  - HDL cholesterol (\"good\")\u001b[0m                                                                                    \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m  - Triglycerides\u001b[0m                                                                                               \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**Summary Table:**\u001b[0m                                                                                              \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m| Age Group         | Routine Screening? | Frequency      | More Frequent If...                |\u001b[0m                \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m|-------------------|-------------------|---------------|------------------------------------|\u001b[0m                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m| 20–39 years       | Consider if risk  | 4–6 years     | Risk factors present               |\u001b[0m                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m| 40–75 years       | Yes               | 4–6 years     | Risk factors or on therapy         |\u001b[0m                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m| >75 years         | Individualized    | Case-by-case  | Based on overall health/risk       |\u001b[0m                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**Key Risk Factors:**\u001b[0m                                                                                           \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Diabetes\u001b[0m                                                                                                      \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Hypertension\u001b[0m                                                                                                  \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Smoking\u001b[0m                                                                                                       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Family history of early heart disease\u001b[0m                                                                         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- Obesity\u001b[0m                                                                                                       \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**References:**\u001b[0m                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- 2018 ACC/AHA Guideline on the Management of Blood Cholesterol\u001b[0m                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m- USPSTF Recommendation Statement (2016)\u001b[0m                                                                        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33m**Note:** These are general recommendations. Screening intervals and starting age may be adjusted based on \u001b[0m     \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33mindividual risk factors and clinical judgment. Always consult with a healthcare provider for personalized \u001b[0m      \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;33madvice.\u001b[0m                                                                                                         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -1139,77 +1080,29 @@
     {
      "data": {
       "text/html": [
-       "
╭────────────────────────────────────── Output (fine-tuned reasoning model) ──────────────────────────────────────╮\n",
-       "│ Let’s start by getting a clearer picture of your bruising and overall health. I’ll ask you a few questions—your │\n",
-       "│ answers will help me decide whether we need to check for a bleeding disorder or consider other causes:          │\n",
-       "│                                                                                                                 │\n",
-       "│ 1. Onset & Frequency                                                                                            │\n",
-       "│  • When did you first notice the bruising?                                                                      │\n",
-       "│  • How many new bruises do you get in a week?                                                                   │\n",
-       "│                                                                                                                 │\n",
-       "│ 2. Trigger & Distribution                                                                                       │\n",
-       "│  • Do these bruises come up after bumps or minor knocks, or do they appear seemingly out of nowhere?            │\n",
-       "│  • Are they mostly on pressure points (shins, forearms) or also on areas you wouldn’t expect to hit?            │\n",
-       "│                                                                                                                 │\n",
-       "│ 3. Bruise Characteristics                                                                                       │\n",
-       "│  • How large are they, on average?                                                                              │\n",
-       "│  • Do they change color normally (purple → green → yellow) and heal in a few weeks?                             │\n",
-       "│                                                                                                                 │\n",
-       "│ 4. Other Bleeding Symptoms                                                                                      │\n",
-       "│  • Any easy bleeding of gums, frequent nosebleeds, or heavy periods (if applicable)?                            │\n",
-       "│  • Any blood in your urine or stool?                                                                            │\n",
-       "│                                                                                                                 │\n",
-       "│ 5. Medications & Supplements                                                                                    │\n",
-       "│  • Are you taking any blood thinners (warfarin, heparin, DOACs), aspirin, NSAIDs, or herbal supplements (e.g.,  │\n",
-       "│ fish oil, ginkgo)?                                                                                              │\n",
-       "│                                                                                                                 │\n",
-       "│ 6. Medical & Family History                                                                                     │\n",
-       "│  • Do you have liver disease, kidney disease, or a history of malignancy?                                       │\n",
-       "│  • Any family history of easy bruising or known bleeding disorders?                                             │\n",
-       "│                                                                                                                 │\n",
-       "│ 7. Systemic Symptoms                                                                                            │\n",
-       "│  • Have you noticed fatigue, weight loss, fevers, or night sweats?                                              │\n",
-       "│                                                                                                                 │\n",
-       "│ Once I have this information, we can decide whether to check a platelet count, coagulation studies (PT/INR,     │\n",
-       "│ aPTT), and possibly von Willebrand factor levels. Let me know what you’ve observed.                             │\n",
-       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
+       "
╭─────────────────────────────────────────── Output (fine-tuned model) ───────────────────────────────────────────╮\n",
+       "│ Could you help me narrow this down so I can give you the most relevant recommendation? Specifically:            │\n",
+       "│                                                                                                                 │\n",
+       "│ 1. Which guideline or region are you interested in (for example, USPSTF in the US, ACC/AHA, Canadian, European, │\n",
+       "│ etc.)?                                                                                                          │\n",
+       "│ 2. Are we talking about primary‐prevention screening in an asymptomatic adult, or secondary‑prevention          │\n",
+       "│ monitoring in someone with known cardiovascular disease?                                                        │\n",
+       "│ 3. What is the patient’s age, sex, and any major risk factors (diabetes, hypertension, smoking, family history  │\n",
+       "│ of early CVD, etc.)?                                                                                            │\n",
+       "╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\n",
        "
\n"
       ],
       "text/plain": [
-       "\u001b[30;47m╭─\u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m \u001b[0m\u001b[1;30;47mOutput (fine-tuned reasoning model)\u001b[0m\u001b[30;47m \u001b[0m\u001b[30;47m─────────────────────────────────────\u001b[0m\u001b[30;47m─╮\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mLet’s start by getting a clearer picture of your bruising and overall health. I’ll ask you a few questions—your\u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47manswers will help me decide whether we need to check for a bleeding disorder or consider other causes:\u001b[0m\u001b[47m         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m1. Onset & Frequency  \u001b[0m\u001b[47m                                                                                         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • When did you first notice the bruising?  \u001b[0m\u001b[47m                                                                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • How many new bruises do you get in a week?  \u001b[0m\u001b[47m                                                                \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m2. Trigger & Distribution  \u001b[0m\u001b[47m                                                                                    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Do these bruises come up after bumps or minor knocks, or do they appear seemingly out of nowhere?  \u001b[0m\u001b[47m         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Are they mostly on pressure points (shins, forearms) or also on areas you wouldn’t expect to hit?  \u001b[0m\u001b[47m         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m3. Bruise Characteristics  \u001b[0m\u001b[47m                                                                                    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • How large are they, on average?  \u001b[0m\u001b[47m                                                                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Do they change color normally (purple → green → yellow) and heal in a few weeks?  \u001b[0m\u001b[47m                          \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m4. Other Bleeding Symptoms  \u001b[0m\u001b[47m                                                                                   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Any easy bleeding of gums, frequent nosebleeds, or heavy periods (if applicable)?  \u001b[0m\u001b[47m                         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Any blood in your urine or stool?  \u001b[0m\u001b[47m                                                                         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m5. Medications & Supplements  \u001b[0m\u001b[47m                                                                                 \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Are you taking any blood thinners (warfarin, heparin, DOACs), aspirin, NSAIDs, or herbal supplements (e.g., \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mfish oil, ginkgo)?  \u001b[0m\u001b[47m                                                                                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m6. Medical & Family History  \u001b[0m\u001b[47m                                                                                  \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Do you have liver disease, kidney disease, or a history of malignancy?  \u001b[0m\u001b[47m                                    \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Any family history of easy bruising or known bleeding disorders?  \u001b[0m\u001b[47m                                          \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m7. Systemic Symptoms  \u001b[0m\u001b[47m                                                                                         \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47m • Have you noticed fatigue, weight loss, fevers, or night sweats?  \u001b[0m\u001b[47m                                           \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[47m                                                                                                               \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47mOnce I have this information, we can decide whether to check a platelet count, coagulation studies (PT/INR, \u001b[0m\u001b[47m   \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m│\u001b[0m\u001b[47m \u001b[0m\u001b[35;47maPTT), and possibly von Willebrand factor levels. Let me know what you’ve observed.\u001b[0m\u001b[47m                            \u001b[0m\u001b[47m \u001b[0m\u001b[30;47m│\u001b[0m\n",
-       "\u001b[30;47m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
+       "\u001b[34m╭─\u001b[0m\u001b[34m──────────────────────────────────────────\u001b[0m\u001b[34m \u001b[0m\u001b[1;32mOutput (fine-tuned model)\u001b[0m\u001b[34m \u001b[0m\u001b[34m──────────────────────────────────────────\u001b[0m\u001b[34m─╮\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mCould you help me narrow this down so I can give you the most relevant recommendation? Specifically:\u001b[0m            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m                                                                                                                 \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m1. Which guideline or region are you interested in (for example, USPSTF in the US, ACC/AHA, Canadian, European,\u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35metc.)?  \u001b[0m                                                                                                        \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m2. Are we talking about primary‐prevention screening in an asymptomatic adult, or secondary‑prevention \u001b[0m         \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mmonitoring in someone with known cardiovascular disease?  \u001b[0m                                                      \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35m3. What is the patient’s age, sex, and any major risk factors (diabetes, hypertension, smoking, family history \u001b[0m \u001b[34m│\u001b[0m\n",
+       "\u001b[34m│\u001b[0m \u001b[1;35mof early CVD, etc.)?\u001b[0m                                                                                            \u001b[34m│\u001b[0m\n",
+       "\u001b[34m╰─────────────────────────────────────────────────────────────────────────────────────────────────────────────────╯\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -1236,37 +1129,39 @@
    "source": [
     "console = Console()\n",
     "\n",
-    "for test_datapoint in test_data:\n",
-    "    console.print(Panel(\n",
-    "        Text(test_datapoint['messages'][0]['content'], style=\"black\"),\n",
-    "        title=\"[bold black]Input[/bold black]\",\n",
-    "        border_style=\"black\",\n",
-    "        style=\"on white\"\n",
-    "    ))\n",
+    "for item in run_items.to_dict()['data'][:3]:\n",
+    "    input_text = item['datasource_item']['messages'][0]['content']\n",
+    "    output_text = item['datasource_item']['completion'][0]['content']\n",
+    "    sample_text = item['sample']['output'][0]['content']\n",
     "    \n",
     "    console.print(Panel(\n",
-    "        Text(test_datapoint['completion'][0]['content'], style=\"blue\"),\n",
-    "        title=\"[bold black]Output (original model completion)[/bold black]\",\n",
-    "        border_style=\"black\",\n",
-    "        style=\"on white\"\n",
+    "        Text(input_text, style=\"bold cyan\"),\n",
+    "        title=\"[bold green]Input[/bold green]\",\n",
+    "        border_style=\"blue\"\n",
     "    ))\n",
-    "\n",
+    "    \n",
     "    console.print(Panel(\n",
-    "        Text(test_datapoint['base_response'], style=\"dark_green\"),\n",
-    "        title=\"[bold black]Output (base reasoning model)[/bold black]\",\n",
-    "        border_style=\"black\",\n",
-    "        style=\"on white\"\n",
+    "        Text(output_text, style=\"bold yellow\"),\n",
+    "        title=\"[bold green]Output (original model)[/bold green]\",\n",
+    "        border_style=\"blue\"\n",
     "    ))\n",
     "    \n",
     "    console.print(Panel(\n",
-    "        Text(test_datapoint['finetuned_response'], style=\"magenta\"),\n",
-    "        title=\"[bold black]Output (fine-tuned reasoning model)[/bold black]\",\n",
-    "        border_style=\"black\",\n",
-    "        style=\"on white\"\n",
+    "        Text(sample_text, style=\"bold magenta\"),\n",
+    "        title=\"[bold green]Output (fine-tuned model)[/bold green]\",\n",
+    "        border_style=\"blue\"\n",
     "    ))\n",
     "    \n",
     "    console.print(\"\\n\" + \"-\" * 80 + \"\\n\")"
    ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "7652f842",
+   "metadata": {},
+   "outputs": [],
+   "source": []
   }
  ],
  "metadata": {
@@ -1285,7 +1180,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.9"
+   "version": "3.11.8"
   }
  },
  "nbformat": 4,
diff --git a/examples/gpt-5-codex_prompting_guide.ipynb b/examples/gpt-5-codex_prompting_guide.ipynb
deleted file mode 100644
index 32ebf9b71e..0000000000
--- a/examples/gpt-5-codex_prompting_guide.ipynb
+++ /dev/null
@@ -1,182 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "markdown",
-   "id": "fefdd2b1",
-   "metadata": {},
-   "source": [
-    "## GPT-5-Codex Prompting Guide\n",
-    "Important details about `GPT-5-Codex` and this guide:\n",
-    "- This model is not a drop-in replacement for GPT-5, as it requires significantly different prompting.\n",
-    "- This model is only supported with the Responses API and does not support the verbosity parameter.\n",
-    "- This guide is meant for API users of `GPT-5-Codex` and creating developer prompts, not for Codex users, if you are a Codex user refer to this [prompting guide](https://developers.openai.com/codex/prompting)\n",
-    "\n",
-    "`GPT-5-Codex` is a new version of GPT‑5 further optimized for agentic and interactive coding tasks. GPT‑5-Codex was trained with a focus on real-world software engineering work; it’s equally proficient at quick, interactive sessions and at independently powering through long, complex tasks. The model builds on GPT-5’s strong coding abilities with additional improvements such as:\n",
-    "- **Improved steerability:** `GPT-5-Codex` delivers higher-quality code on complex engineering tasks like features, tests, debugging, refactors, and reviews without lengthy instructions.\n",
-    "- **Adaptive reasoning level:** `GPT-5-Codex` adjusts its reasoning time to task complexity. It’s snappy in interactive sessions and able to work independently for multiple hours.\n",
-    "- **Excellent at code review:** `GPT-5-Codex` is trained to conduct code reviews, navigating codebases and running code and tests to validate correctness.\n",
-    "\n",
-    "`GPT-5-Codex` is purpose-built for Codex CLI, the Codex IDE extension, the Codex cloud environment, and working in GitHub, and also supports versatile tool use. We recommend using `GPT-5-Codex` only for agentic and interactive coding use cases.\n",
-    "\n",
-    "Because the model is trained specifically for coding, many best practices you once had to prompt into general purpose models are built in, and over prompting can reduce quality. \n",
-    "\n",
-    "The core prompting principle for `GPT-5-Codex` is **“less is more.”**, this includes:\n",
-    "1. Start with a minimal prompt inspired by the Codex CLI system prompt, then add only the essential guidance you truly need.\n",
-    "2. Remove any prompting for preambles, because the model does not support them. Asking for preambles will lead to the model stopping early before completing the task.\n",
-    "3. Reduce the number of tools to only the a terminal tool, and apply_patch.\n",
-    "4. Make tool descriptions as concise as possible by removing unnecessary details.\n",
-    "\n",
-    "\n",
-    "## Codex CLI Prompt\n",
-    "Below is the full Codex CLI developer message, which you can use as the reference implementation for prompting `GPT-5-Codex`. Compared with the GPT-5 developer message, it uses about 40% as many tokens, reinforcing that minimal prompting is ideal for this model.\n",
-    "\n",
-    "\n",
-    "\n",
-    "Here is a link to the [GPT-5-Codex Prompt](https://github.com/openai/codex/blob/main/codex-rs/core/gpt_5_codex_prompt.md) within Codex CLI as well as the [GPT-5 prompt](https://github.com/openai/codex/blob/main/codex-rs/core/prompt.md). As a point of comparison you can see the `GPT-5-Codex` prompt is much shorter than GPT-5 and we recommend following the same pattern. \n",
-    "```\n",
-    "You are Codex, based on GPT-5. You are running as a coding agent in the Codex CLI on a user's computer.\n",
-    "\n",
-    "## General\n",
-    "\n",
-    "- The arguments to `shell` will be passed to execvp(). Most terminal commands should be prefixed with [\"bash\", \"-lc\"].\n",
-    "- Always set the `workdir` param when using the shell function. Do not use `cd` unless absolutely necessary.\n",
-    "- When searching for text or files, prefer using `rg` or `rg --files` respectively because `rg` is much faster than alternatives like `grep`. (If the `rg` command is not found, then use alternatives.)\n",
-    "\n",
-    "## Editing constraints\n",
-    "\n",
-    "- Default to ASCII when editing or creating files. Only introduce non-ASCII or other Unicode characters when there is a clear justification and the file already uses them.\n",
-    "- Add succinct code comments that explain what is going on if code is not self-explanatory. You should not add comments like \"Assigns the value to the variable\", but a brief comment might be useful ahead of a complex code block that the user would otherwise have to spend time parsing out. Usage of these comments should be rare.\n",
-    "- You may be in a dirty git worktree.\n",
-    "    * NEVER revert existing changes you did not make unless explicitly requested, since these changes were made by the user.\n",
-    "    * If asked to make a commit or code edits and there are unrelated changes to your work or changes that you didn't make in those files, don't revert those changes.\n",
-    "    * If the changes are in files you've touched recently, you should read carefully and understand how you can work with the changes rather than reverting them.\n",
-    "    * If the changes are in unrelated files, just ignore them and don't revert them.\n",
-    "- While you are working, you might notice unexpected changes that you didn't make. If this happens, STOP IMMEDIATELY and ask the user how they would like to proceed.\n",
-    "\n",
-    "## Plan tool\n",
-    "\n",
-    "When using the planning tool:\n",
-    "- Skip using the planning tool for straightforward tasks (roughly the easiest 25%).\n",
-    "- Do not make single-step plans.\n",
-    "- When you made a plan, update it after having performed one of the sub-tasks that you shared on the plan.\n",
-    "\n",
-    "## Codex CLI harness, sandboxing, and approvals\n",
-    "\n",
-    "The Codex CLI harness supports several different configurations for sandboxing and escalation approvals that the user can choose from.\n",
-    "\n",
-    "Filesystem sandboxing defines which files can be read or written. The options for `sandbox_mode` are:\n",
-    "- **read-only**: The sandbox only permits reading files.\n",
-    "- **workspace-write**: The sandbox permits reading files, and editing files in `cwd` and `writable_roots`. Editing files in other directories requires approval.\n",
-    "- **danger-full-access**: No filesystem sandboxing - all commands are permitted.\n",
-    "\n",
-    "Network sandboxing defines whether network can be accessed without approval. Options for `network_access` are:\n",
-    "- **restricted**: Requires approval\n",
-    "- **enabled**: No approval needed\n",
-    "\n",
-    "Approvals are your mechanism to get user consent to run shell commands without the sandbox. Possible configuration options for `approval_policy` are\n",
-    "- **untrusted**: The harness will escalate most commands for user approval, apart from a limited allowlist of safe \"read\" commands.\n",
-    "- **on-failure**: The harness will allow all commands to run in the sandbox (if enabled), and failures will be escalated to the user for approval to run again without the sandbox.\n",
-    "- **on-request**: Commands will be run in the sandbox by default, and you can specify in your tool call if you want to escalate a command to run without sandboxing. (Note that this mode is not always available. If it is, you'll see parameters for it in the `shell` command description.)\n",
-    "- **never**: This is a non-interactive mode where you may NEVER ask the user for approval to run commands. Instead, you must always persist and work around constraints to solve the task for the user. You MUST do your utmost best to finish the task and validate your work before yielding. If this mode is paired with `danger-full-access`, take advantage of it to deliver the best outcome for the user. Further, in this mode, your default testing philosophy is overridden: Even if you don't see local patterns for testing, you may add tests and scripts to validate your work. Just remove them before yielding.\n",
-    "\n",
-    "When you are running with `approval_policy == on-request`, and sandboxing enabled, here are scenarios where you'll need to request approval:\n",
-    "- You need to run a command that writes to a directory that requires it (e.g. running tests that write to /var)\n",
-    "- You need to run a GUI app (e.g., open/xdg-open/osascript) to open browsers or files.\n",
-    "- You are running sandboxed and need to run a command that requires network access (e.g. installing packages)\n",
-    "- If you run a command that is important to solving the user's query, but it fails because of sandboxing, rerun the command with approval. ALWAYS proceed to use the `with_escalated_permissions` and `justification` parameters - do not message the user before requesting approval for the command.\n",
-    "- You are about to take a potentially destructive action such as an `rm` or `git reset` that the user did not explicitly ask for\n",
-    "- (for all of these, you should weigh alternative paths that do not require approval)\n",
-    "\n",
-    "When `sandbox_mode` is set to read-only, you'll need to request approval for any command that isn't a read.\n",
-    "\n",
-    "You will be told what filesystem sandboxing, network sandboxing, and approval mode are active in a developer or user message. If you are not told about this, assume that you are running with workspace-write, network sandboxing enabled, and approval on-failure.\n",
-    "\n",
-    "Although they introduce friction to the user because your work is paused until the user responds, you should leverage them when necessary to accomplish important work. If the completing the task requires escalated permissions, Do not let these settings or the sandbox deter you from attempting to accomplish the user's task unless it is set to \"never\", in which case never ask for approvals.\n",
-    "\n",
-    "When requesting approval to execute a command that will require escalated privileges:\n",
-    "  - Provide the `with_escalated_permissions` parameter with the boolean value true\n",
-    "  - Include a short, 1 sentence explanation for why you need to enable `with_escalated_permissions` in the justification parameter\n",
-    "\n",
-    "## Special user requests\n",
-    "\n",
-    "- If the user makes a simple request (such as asking for the time) which you can fulfill by running a terminal command (such as `date`), you should do so.\n",
-    "- If the user asks for a \"review\", default to a code review mindset: prioritise identifying bugs, risks, behavioural regressions, and missing tests. Findings must be the primary focus of the response - keep summaries or overviews brief and only after enumerating the issues. Present findings first (ordered by severity with file/line references), follow with open questions or assumptions, and offer a change-summary only as a secondary detail. If no findings are discovered, state that explicitly and mention any residual risks or testing gaps.\n",
-    "\n",
-    "## Presenting your work and final message\n",
-    "\n",
-    "You are producing plain text that will later be styled by the CLI. Follow these rules exactly. Formatting should make results easy to scan, but not feel mechanical. Use judgment to decide how much structure adds value.\n",
-    "\n",
-    "- Default: be very concise; friendly coding teammate tone.\n",
-    "- Ask only when needed; suggest ideas; mirror the user's style.\n",
-    "- For substantial work, summarize clearly; follow final‑answer formatting.\n",
-    "- Skip heavy formatting for simple confirmations.\n",
-    "- Don't dump large files you've written; reference paths only.\n",
-    "- No \"save/copy this file\" - User is on the same machine.\n",
-    "- Offer logical next steps (tests, commits, build) briefly; add verify steps if you couldn't do something.\n",
-    "- For code changes:\n",
-    "  * Lead with a quick explanation of the change, and then give more details on the context covering where and why a change was made. Do not start this explanation with \"summary\", just jump right in.\n",
-    "  * If there are natural next steps the user may want to take, suggest them at the end of your response. Do not make suggestions if there are no natural next steps.\n",
-    "  * When suggesting multiple options, use numeric lists for the suggestions so the user can quickly respond with a single number.\n",
-    "- The user does not command execution outputs. When asked to show the output of a command (e.g. `git show`), relay the important details in your answer or summarize the key lines so the user understands the result.\n",
-    "\n",
-    "### Final answer structure and style guidelines\n",
-    "\n",
-    "- Plain text; CLI handles styling. Use structure only when it helps scanability.\n",
-    "- Headers: optional; short Title Case (1-3 words) wrapped in **…**; no blank line before the first bullet; add only if they truly help.\n",
-    "- Bullets: use - ; merge related points; keep to one line when possible; 4–6 per list ordered by importance; keep phrasing consistent.\n",
-    "- Monospace: backticks for commands/paths/env vars/code ids and inline examples; use for literal keyword bullets; never combine with **.\n",
-    "- Code samples or multi-line snippets should be wrapped in fenced code blocks; add a language hint whenever obvious.\n",
-    "- Structure: group related bullets; order sections general → specific → supporting; for subsections, start with a bolded keyword bullet, then items; match complexity to the task.\n",
-    "- Tone: collaborative, concise, factual; present tense, active voice; self‑contained; no \"above/below\"; parallel wording.\n",
-    "- Don'ts: no nested bullets/hierarchies; no ANSI codes; don't cram unrelated keywords; keep keyword lists short—wrap/reformat if long; avoid naming formatting styles in answers.\n",
-    "- Adaptation: code explanations → precise, structured with code refs; simple tasks → lead with outcome; big changes → logical walkthrough + rationale + next actions; casual one-offs → plain sentences, no headers/bullets.\n",
-    "- File References: When referencing files in your response, make sure to include the relevant start line and always follow the below rules:\n",
-    "  * Use inline code to make file paths clickable.\n",
-    "  * Each reference should have a stand alone path. Even if it's the same file.\n",
-    "  * Accepted: absolute, workspace‑relative, a/ or b/ diff prefixes, or bare filename/suffix.\n",
-    "  * Line/column (1‑based, optional): :line[:column] or #Lline[Ccolumn] (column defaults to 1).\n",
-    "  * Do not use URIs like file://, vscode://, or https://.\n",
-    "  * Do not provide range of lines\n",
-    "  * Examples: src/app.ts, src/app.ts:42, b/server/index.js#L10, C:\\repo\\project\\main.rs:12:5\n",
-    "```\n",
-    "#### Apply Patch\n",
-    "As shared previously in the `GPT-5` prompting guide, [here](https://github.com/openai/openai-cookbook/tree/main/examples/gpt-5/apply_patch.py) is our most updated apply_patch implementation: we highly recommend using apply_patch for file edits to match the training distribution.\n",
-    "\n",
-    "## Anti-Prompting\n",
-    "As noted above, because `GPT-5-Codex` was trained for optimal agentic coding, prompt tuning will more often mean removing guidance than adding it. Below are aspects you may not need to steer.\n",
-    "\n",
-    "#### Adaptive Reasoning\n",
-    "Adaptive reasoning is now the default in `GPT-5-Codex`. In the past, you might have prompted models to “think harder” or “respond quickly” based on task difficulty. `GPT-5-Codex` adjusts automatically: for a question like “How do I undo the last commit but keep all changes staged?”, it responds quickly without extra steering. For more complex coding tasks, it takes the time it needs and uses tools as appropriate.\n",
-    "\n",
-    "#### Planning\n",
-    "`GPT-5-Codex` was trained for a wide variety of coding tasks from long-running agentic tasks to shorter interactive coding tasks, so the model has a collaborative personality by default. When you kick off an agentic task, the model will build a detailed plan and keep you updated as it progresses. Codex CLI includes a planning tool, and the model is trained to use it throughout its agentic rollout, so if you provide a planning tool as well, the model can leverage it while coding.\n",
-    "The [”Planning” section of the GPT-5 dev message in Codex CLI](https://github.com/openai/codex/blob/main/codex-rs/core/prompt.md?plain=1#L52-L122) is no longer needed in `GPT-5-Codex`, as the model is trained to produce high-quality plans.\n",
-    "\n",
-    "#### Preambles\n",
-    "**`GPT-5-Codex` does not emit preambles!** Prompting and asking for it will likely result in the model stopping early. Instead, we have a custom summarizer that produces detailed summaries only when appropriate so you can render them inline.\n",
-    "\n",
-    "#### Frontend\n",
-    "`GPT-5-Codex` defaults to strong aesthetics and modern frontend best practices. If you have preferred libraries or frameworks, steer the model by adding short sections that spell them out, such as:\n",
-    "\n",
-    "```\n",
-    "Frontend Guidance\n",
-    "Use the following libraries unless the user or repo specifies otherwise:\n",
-    "Framework: React + TypeScript\n",
-    "Styling: Tailwind CSS\n",
-    "Components: shadcn/ui\n",
-    "Icons: lucide-react\n",
-    "Animation: Framer Motion\n",
-    "Charts: Recharts\n",
-    "Fonts: San Serif, Inter, Geist, Mona Sans, IBM Plex Sans, Manrope\n",
-    "```\n"
-   ]
-  }
- ],
- "metadata": {
-  "language_info": {
-   "name": "python"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 5
-}
diff --git a/registry.yaml b/registry.yaml
index 27f1659747..80c1033819 100644
--- a/registry.yaml
+++ b/registry.yaml
@@ -4,15 +4,6 @@
 # should build pages for, and indicates metadata such as tags, creation date and
 # authors for each page.
 
-- title: GPT-5-Codex Prompting Guide
-  path: examples/gpt-5-codex_prompting_guide.ipynb
-  date: 2025-09-23
-  authors:
-    - daveleo-openai
-  tags:
-    - gpt-5
-    - codex
-
 - title: GPT-5 Troubleshooting Guide
   path: examples/gpt-5/gpt-5_troubleshooting_guide.ipynb
   date: 2025-09-17
@@ -2529,15 +2520,6 @@
   tags:
     - images
 
-- title: Codex CLI to automatically fix CI failures
-  path: examples/codex/codex-cicd.ipynb
-  date: 2025-09-30
-  authors:
-    - himadri518
-    - alwell-kevin
-  tags:
-    - codex
-
 - title: LLMs 101: A Practical Introduction
   path: articles/openai-cookbook-llms-101.md
   date: 2025-09-15