update

Author: fochan

docs/class1/class1.rst

@@ -5,7 +5,7 @@ Class 1: The fundamental of Generative Artificial Intelligent (AI)
 
 AI Primer
 ---------
-As GenAI Practitioner, you need to minimally understand what it means with the following term and able to articulate the concept introduced. The following not an exhasive list. The purpose if to get you started. For details, please refer to various contents. Content below were extracted from collection of resources from Internet.
+As GenAI Practitioner, you need to minimally understand what it means with the following term and able to articulate the concept introduced. The following not an exhaustive list. The purpose is to get you started. For details, please refer to various contents. Content below were extracted from collection of resources from Internet.
 
 
 What is AI and how different with GenAI?
@@ -27,12 +27,13 @@ In summary, LLMs are a specialized application of machine learning, focusing on
 What is SLM?
 ~~~~~~~~~~~~
 SLM (Small Language Model) is a type of artificial intelligence model designed for natural language processing that has fewer parameters (between million to few billion) and computational requirements compared to large language models.
-Parameter Scale Examples
 
-Small language model: 100 million to 1 billion parameters
-Medium model: 1-10 billion parameters
-Large models like GPT-3: 175 billion parameters
-Very large models: 500 billion to 1 trillion parameters
+**Parameter Scale Examples**
+
+- Small language model: 100 million to 1 billion parameters
+- Medium model: 1-10 billion parameters
+- Large models like GPT-3: 175 billion parameters
+- Very large models: 500 billion to 1 trillion parameters
 
 
 What is ML?
@@ -43,13 +44,15 @@ What hallucination means in AI?
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Hallucination in AI is when an AI model generates information that is false, inaccurate, or completely made up, even though it might sound convincing. It's like the AI "imagining" things that aren't real or aren't supported by its training data.
 
-For instance, if you ask an AI about a person named "Olivia Smith" it might confidently generate a detailed biography about a specific Oliver Smith, complete with birth date and achievements, even though it's not referring to any real person – it's just combining patterns it learned during training
+For instance, if you ask an AI about a person named "Olivia Smith" it might confidently generate a detailed biography about a specific Olivia Smith, complete with birth date and achievements, even though it's not referring to any real person – it's just combining patterns it learned during training
 
 
 What "token" means in context in AI?
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-A token is the basic unit of text that an AI model processes and understands. A token can be a word, part of a word, or even a punctuation mark. Tokenization is the process of breaking down text into these individual tokens
+A token is the basic unit of text that an AI model processes and understands. A token can be a word, part of a word, or even a punctuation mark. Tokenization is the process of breaking down text into these individual tokens.
+
 Typically, 1 token is approximately:
+
 - 4 characters of English text
 - About 3/4 of a word
 - Varies slightly between different AI models
@@ -94,7 +97,7 @@ What is "temperature" in AI?
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 Temperature controlled “the creativity of the response. It is a hyperparameter that controls the randomness or creativity of the model's output during text generation. 
 
-Low temperature (close to 0) makes the output more deterministic and focused. Model selects the most probable words. Responses become more percise and consistent. Lesss creative and more conservative outputs.
+Low temperature (close to 0) makes the output more deterministic and focused. Model selects the most probable words. Responses become more percise and consistent. Less creative and more conservative outputs.
 
 High temperature (closer to 1 or above) increases randomness and creativity. Model is more likely to choose less probable words. Outputs become more diverse and unpredictable. Can generate more unique and imaginative responses.
 

docs/class2/_static/class2-0.png

@@ -6,7 +6,9 @@ Class 2: Deploy and Secure a modern application
 1 - Deploy Nginx Ingress Controller with App Protect to protect Web Apps Server
 --------------------------------------------------------------------------------
 
-After login to linux Jumphost, change directory to **webapps**. Jumphost server was installed with utilities called 'direnv' - https://direnv.net/. Its a tools that will load the environment file (kubeconfig) when you switch to that directory. Its an efficient tools to switch K8S context from one cluster to the other just by changing directory.
+..  image:: ./_static/class2-0.png
+
+After login/putty to linux Jumphost (from Windows10 Jumphost), change directory to **webapps**. Jumphost server was installed with utilities called 'direnv' - https://direnv.net/. Its a tools that will load the environment file (kubeconfig) when you switch to that directory. Its an efficient tools to switch K8S context from one cluster to the other just by changing directory.
 
 .. NOTE::
        Refer to **Prerequsite** section to find the password for the Windows10 and Linux Jumphost.
@@ -96,6 +98,8 @@ values-plus-nap.yaml ::
 2 - Deploy Arcadia Financial Modern Apps
 ----------------------------------------
 
+..  image:: ./_static/class2-3-0.png
+
 Deploy Arcadia Financial application on Kubernetes. Arcadia Trading consist of multiple microservices.
 
 - frontend

docs/class2/_static/class2-3-0.png

@@ -43,6 +43,8 @@ With the growing popularity of Generative AI, your organization has decided to u
 2 - Deploy Nginx Ingress Controller for AIGW K8S
 ------------------------------------------------
 
+..  image:: ./_static/class3-1-1.png
+
 .. code-block:: bash
 
    cd ~/ai-gateway/nginx-ingress
@@ -67,6 +69,8 @@ With the growing popularity of Generative AI, your organization has decided to u
 3 - Deploy Open-WebUI with Ollama Service
 -----------------------------------------
 
+..  image:: ./_static/class3-4-0.png
+
 **Open-Webui** is a self-hosted WebUI that allow user to interact with AI models. It allow user to download respective language model that use by Ollama.
 
 **Ollama** is a open-source tools that allow user to run large language model. User can expose model access via Ollama inference API.
@@ -199,7 +203,7 @@ Test interacting with LLM model. Feel free to test with different language model
 .. attention:: 
    Please do notes that GenAI is hallucinating and providing wrong information - about F5 Inc headquarters. Please ignore as smaller model (smaller parameter, less intelligent) tend to hallucinate more compare to a larger model. Large models with more parameters are more capable and intelligent than smaller models, but require expensive machines with multiple GPUs to run. 
 
-   Its also depends on dataset use for the training - "Garbage In, Garbage Out".
+   It also depends on dataset use for the training - "Garbage In, Garbage Out".
 
 
 5 - Deploy LLM model service (Ollama)
@@ -238,6 +242,9 @@ ollama-ingress-http.yaml ::
 
 6 - Deploy LLM orchestrator service (Flowise AI)
 ------------------------------------------------
+
+..  image:: ./_static/class3-13-0.png
+
 Deploy LLM Orchstrator to facilitate AI component communication. Flowise AI - an open source low-code tool for developer to build customized LLM orchstration flow and AI agent is used. (https://flowiseai.com/). Flowise complements LangChain by offering a visual interface.
 
 .. code-block:: bash
@@ -271,7 +278,7 @@ Deploy LLM Orchstrator to facilitate AI component communication. Flowise AI - an
 
 ..  image:: ./_static/class3-13.png
 
-Flowise is installed with the following custom values. You can login with the password shown in the value file.
+Flowise is installed with the following custom values. Plese take notes of the password as you may need it for the next section.
 
 values.yaml ::
 
@@ -327,15 +334,15 @@ Confirm that you can login and access to LLM orchestrator (flowise)
 
 ..  image:: ./_static/class3-15.png
 
-Import arcadia RAG chatflow into flowise. Select **Add New**, settings icons and **Load Chatflow**
+Import arcadia RAG chatflow into flowise. Select **Add New**, click **Settings icons** and **Load Chatflow**
 
 ..  image:: ./_static/class3-16.png
 
 A copy of the chatflow located on the jumphost **Documents** directory. Select the chatflow json file.
 
 ..  image:: ./_static/class3-17.png
 
-Save the chatflow with a name as shown.
+Save the chatflow (arcadia-rag)
 
 ..  image:: ./_static/class3-18.png
 
@@ -354,6 +361,8 @@ Save the chatflow with a name as shown.
 7 - Deploy Vector Database (Qdrant)
 -----------------------------------
 
+..  image:: ./_static/class3-20-0.png
+
 **Qdrant** is a vector similarity search engine and vector database. It provides a production-ready service with a convenient API to store, search, and manage vectors points.
 
 |
@@ -426,7 +435,7 @@ Load the imported "arcadia-rag" chatflow.
 ..  image:: ./_static/class3-24.png
 
 
-Here is the full RAG pipeline implmented in a low-code platform.
+Here is the full RAG pipeline implemented in a low-code platform.
 
 ..  image:: ./_static/class3-25.png
 
@@ -437,7 +446,7 @@ Here are some of the node/chain used.
 +=============================================+=======================================================================+
 |  **Recursive Character Text Splitter**      |  Split documents recursively by different characters -                |
 |                                             |  starting with "\n\n", then "\n", then " ".                           |
-|     Chunck Size: 250                        |                                                                       |
+|     Chunk Size: 250                         |                                                                       |
 |                                             |                                                                       |
 |     Chunk Overlap : 20                      |                                                                       |
 |                                             |                                                                       |