diff --git a/examples/community/README.md b/examples/community/README.md
index e51124e75956..267c8f4bb904 100755
--- a/examples/community/README.md
+++ b/examples/community/README.md
@@ -73,7 +73,8 @@ Please also check out our [Community Scripts](https://github.com/huggingface/dif
| Stable Diffusion BoxDiff Pipeline | Training-free controlled generation with bounding boxes using [BoxDiff](https://github.com/showlab/BoxDiff) | [Stable Diffusion BoxDiff Pipeline](#stable-diffusion-boxdiff) | - | [Jingyang Zhang](https://github.com/zjysteven/) |
| FRESCO V2V Pipeline | Implementation of [[CVPR 2024] FRESCO: Spatial-Temporal Correspondence for Zero-Shot Video Translation](https://arxiv.org/abs/2403.12962) | [FRESCO V2V Pipeline](#fresco) | - | [Yifan Zhou](https://github.com/SingleZombie) |
| AnimateDiff IPEX Pipeline | Accelerate AnimateDiff inference pipeline with BF16/FP32 precision on Intel Xeon CPUs with [IPEX](https://github.com/intel/intel-extension-for-pytorch) | [AnimateDiff on IPEX](#animatediff-on-ipex) | - | [Dan Li](https://github.com/ustcuna/) |
-| HunyuanDiT Differential Diffusion Pipeline | Applies [Differential Diffsuion](https://github.com/exx8/differential-diffusion) to [HunyuanDiT](https://github.com/huggingface/diffusers/pull/8240). | [HunyuanDiT with Differential Diffusion](#hunyuandit-with-differential-diffusion) | [](https://colab.research.google.com/drive/1v44a5fpzyr4Ffr4v2XBQ7BajzG874N4P?usp=sharing) | [Monjoy Choudhury](https://github.com/MnCSSJ4x) |
+| HunyuanDiT Differential Diffusion Pipeline | Applies [Differential Diffusion](https://github.com/exx8/differential-diffusion) to [HunyuanDiT](https://github.com/huggingface/diffusers/pull/8240). | [HunyuanDiT with Differential Diffusion](#hunyuandit-with-differential-diffusion) | [](https://colab.research.google.com/drive/1v44a5fpzyr4Ffr4v2XBQ7BajzG874N4P?usp=sharing) | [Monjoy Choudhury](https://github.com/MnCSSJ4x) |
+| [🪆Matryoshka Diffusion Models](https://huggingface.co/papers/2310.15111) | A diffusion process that denoises inputs at multiple resolutions jointly and uses a NestedUNet architecture where features and parameters for small scale inputs are nested within those of the large scales. See [original codebase](https://github.com/apple/ml-mdm). | [🪆Matryoshka Diffusion Models](#matryoshka-diffusion-models) | [](https://huggingface.co/spaces/pcuenq/mdm) [](https://colab.research.google.com/gist/tolgacangoz/1f54875fc7aeaabcf284ebde64820966/matryoshka_hf.ipynb) | [M. Tolga Cangöz](https://github.com/tolgacangoz) |
To load a custom pipeline you just need to pass the `custom_pipeline` argument to `DiffusionPipeline`, as one of the files in `diffusers/examples/community`. Feel free to send a PR with your own pipelines, we will merge them quickly.
@@ -85,17 +86,17 @@ pipe = DiffusionPipeline.from_pretrained("stable-diffusion-v1-5/stable-diffusion
### Flux with CFG
-Know more about Flux [here](https://blackforestlabs.ai/announcing-black-forest-labs/). Since Flux doesn't use CFG, this implementation provides one, inspired by the [PuLID Flux adaptation](https://github.com/ToTheBeginning/PuLID/blob/main/docs/pulid_for_flux.md).
+Know more about Flux [here](https://blackforestlabs.ai/announcing-black-forest-labs/). Since Flux doesn't use CFG, this implementation provides one, inspired by the [PuLID Flux adaptation](https://github.com/ToTheBeginning/PuLID/blob/main/docs/pulid_for_flux.md).
Example usage:
```py
from diffusers import DiffusionPipeline
-import torch
+import torch
pipeline = DiffusionPipeline.from_pretrained(
- "black-forest-labs/FLUX.1-dev",
- torch_dtype=torch.bfloat16,
+ "black-forest-labs/FLUX.1-dev",
+ torch_dtype=torch.bfloat16,
custom_pipeline="pipeline_flux_with_cfg"
)
pipeline.enable_model_cpu_offload()
@@ -103,10 +104,10 @@ prompt = "a watercolor painting of a unicorn"
negative_prompt = "pink"
img = pipeline(
- prompt=prompt,
- negative_prompt=negative_prompt,
- true_cfg=1.5,
- guidance_scale=3.5,
+ prompt=prompt,
+ negative_prompt=negative_prompt,
+ true_cfg=1.5,
+ guidance_scale=3.5,
num_images_per_prompt=1,
generator=torch.manual_seed(0)
).images[0]
@@ -2656,7 +2657,7 @@ image with mask mech_painted.png
-result:
+result:
@@ -4324,6 +4325,51 @@ image = pipe(
A colab notebook demonstrating all results can be found [here](https://colab.research.google.com/drive/1v44a5fpzyr4Ffr4v2XBQ7BajzG874N4P?usp=sharing). Depth Maps have also been added in the same colab.
+### 🪆Matryoshka Diffusion Models
+
+
+
+The Abstract of the paper:
+>Diffusion models are the _de-facto_ approach for generating high-quality images and videos but learning high-dimensional models remains a formidable task due to computational and optimization challenges. Existing methods often resort to training cascaded models in pixel space, or using a downsampled latent space of a separately trained auto-encoder. In this paper, we introduce Matryoshka Diffusion (MDM), **a novel framework for high-resolution image and video synthesis**. We propose a diffusion process that denoises inputs at multiple resolutions jointly and uses a **NestedUNet** architecture where features and parameters for small scale inputs are nested within those of the large scales. In addition, MDM enables a progressive training schedule from lower to higher resolutions which leads to significant improvements in optimization for high-resolution generation. We demonstrate the effectiveness of our approach on various benchmarks, including class-conditioned image generation, high-resolution text-to-image, and text-to-video applications. Remarkably, we can train a **_single pixel-space model_ at resolutions of up to 1024 × 1024 pixels**, demonstrating strong zero shot generalization using the **CC12M dataset, which contains only 12 million images**. Code and pre-trained checkpoints are released at https://github.com/apple/ml-mdm.
+
+- `64×64, nesting_level=0`: 1.719 GiB. With `50` DDIM inference steps:
+
+**64x64**
+:-------------------------:
+| 
|
+
+- `256×256, nesting_level=1`: 1.776 GiB. With `150` DDIM inference steps:
+
+**64x64** | **256x256**
+:-------------------------:|:-------------------------:
+| 
| 
|
+
+- `1024×1024, nesting_level=2`: 1.792 GiB. As one can realize the cost of adding another layer is really negligible. With `250` DDIM inference steps:
+
+**64x64** | **256x256** | **1024x1024**
+:-------------------------:|:-------------------------:|:-------------------------:
+| 
| 
| 
|
+
+```py
+from diffusers import DiffusionPipeline
+from diffusers.utils import make_image_grid
+
+# nesting_level=0 -> 64x64; nesting_level=1 -> 256x256 - 64x64; nesting_level=2 -> 1024x1024 - 256x256 - 64x64
+pipe = DiffusionPipeline.from_pretrained("tolgacangoz/matryoshka-diffusion-models",
+ nesting_level=0,
+ trust_remote_code=False, # One needs to give permission for this code to run
+ ).to("cuda")
+
+prompt0 = "a blue jay stops on the top of a helmet of Japanese samurai, background with sakura tree"
+prompt = f"breathtaking {prompt0}. award-winning, professional, highly detailed"
+negative_prompt = "deformed, mutated, ugly, disfigured, blur, blurry, noise, noisy"
+image = pipe(prompt=prompt, negative_prompt=negative_prompt, num_inference_steps=50).images
+make_image_grid(image, rows=1, cols=len(image))
+
+# pipe.change_nesting_level() # 0, 1, or 2
+# 50+, 100+, and 250+ num_inference_steps are recommended for nesting levels 0, 1, and 2 respectively.
+```
+
# Perturbed-Attention Guidance
[Project](https://ku-cvlab.github.io/Perturbed-Attention-Guidance/) / [arXiv](https://arxiv.org/abs/2403.17377) / [GitHub](https://github.com/KU-CVLAB/Perturbed-Attention-Guidance)
diff --git a/examples/community/matryoshka.py b/examples/community/matryoshka.py
new file mode 100644
index 000000000000..7ef1438f7204
--- /dev/null
+++ b/examples/community/matryoshka.py
@@ -0,0 +1,4641 @@
+# Copyright 2024 The HuggingFace Team. All rights reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+# http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+#
+# Based on [🪆Matryoshka Diffusion Models](https://huggingface.co/papers/2310.15111).
+# Authors: Jiatao Gu, Shuangfei Zhai, Yizhe Zhang, Josh Susskind, Navdeep Jaitly
+# Code: https://github.com/apple/ml-mdm with MIT license
+#
+# Adapted to Diffusers by [M. Tolga Cangöz](https://github.com/tolgacangoz).
+
+
+import gc
+import inspect
+import math
+from dataclasses import dataclass
+from typing import Any, Callable, Dict, List, Optional, Tuple, Union
+
+import numpy as np
+import torch
+import torch.nn.functional as F
+import torch.utils.checkpoint
+from packaging import version
+from PIL import Image
+from torch import nn
+from transformers import CLIPImageProcessor, CLIPVisionModelWithProjection, T5EncoderModel, T5TokenizerFast
+
+from diffusers.callbacks import MultiPipelineCallbacks, PipelineCallback
+from diffusers.configuration_utils import ConfigMixin, FrozenDict, LegacyConfigMixin, register_to_config
+from diffusers.image_processor import PipelineImageInput, VaeImageProcessor
+from diffusers.loaders import (
+ FromSingleFileMixin,
+ IPAdapterMixin,
+ PeftAdapterMixin,
+ StableDiffusionLoraLoaderMixin,
+ TextualInversionLoaderMixin,
+ UNet2DConditionLoadersMixin,
+)
+from diffusers.loaders.single_file_model import FromOriginalModelMixin
+from diffusers.models.activations import GELU, get_activation
+from diffusers.models.attention_processor import (
+ ADDED_KV_ATTENTION_PROCESSORS,
+ CROSS_ATTENTION_PROCESSORS,
+ Attention,
+ AttentionProcessor,
+ AttnAddedKVProcessor,
+ AttnProcessor,
+ FusedAttnProcessor2_0,
+)
+from diffusers.models.downsampling import Downsample2D
+from diffusers.models.embeddings import (
+ GaussianFourierProjection,
+ GLIGENTextBoundingboxProjection,
+ ImageHintTimeEmbedding,
+ ImageProjection,
+ ImageTimeEmbedding,
+ TextImageProjection,
+ TextImageTimeEmbedding,
+ TextTimeEmbedding,
+ TimestepEmbedding,
+ Timesteps,
+)
+from diffusers.models.lora import adjust_lora_scale_text_encoder
+from diffusers.models.modeling_utils import LegacyModelMixin, ModelMixin
+from diffusers.models.resnet import ResnetBlock2D
+from diffusers.models.unets.unet_2d_blocks import DownBlock2D, UpBlock2D
+from diffusers.models.upsampling import Upsample2D
+from diffusers.pipelines.pipeline_utils import DiffusionPipeline, StableDiffusionMixin
+from diffusers.schedulers.scheduling_utils import SchedulerMixin
+from diffusers.utils import (
+ USE_PEFT_BACKEND,
+ BaseOutput,
+ deprecate,
+ is_torch_version,
+ is_torch_xla_available,
+ logging,
+ replace_example_docstring,
+ scale_lora_layers,
+ unscale_lora_layers,
+)
+from diffusers.utils.torch_utils import apply_freeu, randn_tensor
+
+
+if is_torch_xla_available():
+ import torch_xla.core.xla_model as xm # type: ignore
+
+ XLA_AVAILABLE = True
+else:
+ XLA_AVAILABLE = False
+
+logger = logging.get_logger(__name__) # pylint: disable=invalid-name
+
+EXAMPLE_DOC_STRING = """
+ Examples:
+ ```py
+ >>> from diffusers import DiffusionPipeline
+ >>> from diffusers.utils import make_image_grid
+
+ >>> # nesting_level=0 -> 64x64; nesting_level=1 -> 256x256 - 64x64; nesting_level=2 -> 1024x1024 - 256x256 - 64x64
+ >>> pipe = DiffusionPipeline.from_pretrained("tolgacangoz/matryoshka-diffusion-models",
+ >>> custom_pipeline="matryoshka").to("cuda")
+
+ >>> prompt0 = "a blue jay stops on the top of a helmet of Japanese samurai, background with sakura tree"
+ >>> prompt = f"breathtaking {prompt0}. award-winning, professional, highly detailed"
+ >>> negative_prompt = "deformed, mutated, ugly, disfigured, blur, blurry, noise, noisy"
+ >>> image = pipe(prompt=prompt, negative_prompt=negative_prompt, num_inference_steps=50).images
+ >>> make_image_grid(image, rows=1, cols=len(image))
+
+ >>> pipe.change_nesting_level() # 0, 1, or 2
+ >>> # 50+, 100+, and 250+ num_inference_steps are recommended for nesting levels 0, 1, and 2 respectively.
+ ```
+"""
+
+
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.rescale_noise_cfg
+def rescale_noise_cfg(noise_cfg, noise_pred_text, guidance_rescale=0.0):
+ """
+ Rescale `noise_cfg` according to `guidance_rescale`. Based on findings of [Common Diffusion Noise Schedules and
+ Sample Steps are Flawed](https://arxiv.org/pdf/2305.08891.pdf). See Section 3.4
+ """
+ std_text = noise_pred_text.std(dim=list(range(1, noise_pred_text.ndim)), keepdim=True)
+ std_cfg = noise_cfg.std(dim=list(range(1, noise_cfg.ndim)), keepdim=True)
+ # rescale the results from guidance (fixes overexposure)
+ noise_pred_rescaled = noise_cfg * (std_text / std_cfg)
+ # mix with the original results from guidance by factor guidance_rescale to avoid "plain looking" images
+ noise_cfg = guidance_rescale * noise_pred_rescaled + (1 - guidance_rescale) * noise_cfg
+ return noise_cfg
+
+
+# Copied from diffusers.pipelines.stable_diffusion.pipeline_stable_diffusion.retrieve_timesteps
+def retrieve_timesteps(
+ scheduler,
+ num_inference_steps: Optional[int] = None,
+ device: Optional[Union[str, torch.device]] = None,
+ timesteps: Optional[List[int]] = None,
+ sigmas: Optional[List[float]] = None,
+ **kwargs,
+):
+ """
+ Calls the scheduler's `set_timesteps` method and retrieves timesteps from the scheduler after the call. Handles
+ custom timesteps. Any kwargs will be supplied to `scheduler.set_timesteps`.
+
+ Args:
+ scheduler (`SchedulerMixin`):
+ The scheduler to get timesteps from.
+ num_inference_steps (`int`):
+ The number of diffusion steps used when generating samples with a pre-trained model. If used, `timesteps`
+ must be `None`.
+ device (`str` or `torch.device`, *optional*):
+ The device to which the timesteps should be moved to. If `None`, the timesteps are not moved.
+ timesteps (`List[int]`, *optional*):
+ Custom timesteps used to override the timestep spacing strategy of the scheduler. If `timesteps` is passed,
+ `num_inference_steps` and `sigmas` must be `None`.
+ sigmas (`List[float]`, *optional*):
+ Custom sigmas used to override the timestep spacing strategy of the scheduler. If `sigmas` is passed,
+ `num_inference_steps` and `timesteps` must be `None`.
+
+ Returns:
+ `Tuple[torch.Tensor, int]`: A tuple where the first element is the timestep schedule from the scheduler and the
+ second element is the number of inference steps.
+ """
+ if timesteps is not None and sigmas is not None:
+ raise ValueError("Only one of `timesteps` or `sigmas` can be passed. Please choose one to set custom values")
+ if timesteps is not None:
+ accepts_timesteps = "timesteps" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+ if not accepts_timesteps:
+ raise ValueError(
+ f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+ f" timestep schedules. Please check whether you are using the correct scheduler."
+ )
+ scheduler.set_timesteps(timesteps=timesteps, device=device, **kwargs)
+ timesteps = scheduler.timesteps
+ num_inference_steps = len(timesteps)
+ elif sigmas is not None:
+ accept_sigmas = "sigmas" in set(inspect.signature(scheduler.set_timesteps).parameters.keys())
+ if not accept_sigmas:
+ raise ValueError(
+ f"The current scheduler class {scheduler.__class__}'s `set_timesteps` does not support custom"
+ f" sigmas schedules. Please check whether you are using the correct scheduler."
+ )
+ scheduler.set_timesteps(sigmas=sigmas, device=device, **kwargs)
+ timesteps = scheduler.timesteps
+ num_inference_steps = len(timesteps)
+ else:
+ scheduler.set_timesteps(num_inference_steps, device=device, **kwargs)
+ timesteps = scheduler.timesteps
+ return timesteps, num_inference_steps
+
+
+# Copied from diffusers.models.attention._chunked_feed_forward
+def _chunked_feed_forward(ff: nn.Module, hidden_states: torch.Tensor, chunk_dim: int, chunk_size: int):
+ # "feed_forward_chunk_size" can be used to save memory
+ if hidden_states.shape[chunk_dim] % chunk_size != 0:
+ raise ValueError(
+ f"`hidden_states` dimension to be chunked: {hidden_states.shape[chunk_dim]} has to be divisible by chunk size: {chunk_size}. Make sure to set an appropriate `chunk_size` when calling `unet.enable_forward_chunking`."
+ )
+
+ num_chunks = hidden_states.shape[chunk_dim] // chunk_size
+ ff_output = torch.cat(
+ [ff(hid_slice) for hid_slice in hidden_states.chunk(num_chunks, dim=chunk_dim)],
+ dim=chunk_dim,
+ )
+ return ff_output
+
+
+@dataclass
+class MatryoshkaDDIMSchedulerOutput(BaseOutput):
+ """
+ Output class for the scheduler's `step` function output.
+
+ Args:
+ prev_sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)` for images):
+ Computed sample `(x_{t-1})` of previous timestep. `prev_sample` should be used as next model input in the
+ denoising loop.
+ pred_original_sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)` for images):
+ The predicted denoised sample `(x_{0})` based on the model output from the current timestep.
+ `pred_original_sample` can be used to preview progress or for guidance.
+ """
+
+ prev_sample: Union[torch.Tensor, List[torch.Tensor]]
+ pred_original_sample: Optional[Union[torch.Tensor, List[torch.Tensor]]] = None
+
+
+# Copied from diffusers.schedulers.scheduling_ddpm.betas_for_alpha_bar
+def betas_for_alpha_bar(
+ num_diffusion_timesteps,
+ max_beta=0.999,
+ alpha_transform_type="cosine",
+):
+ """
+ Create a beta schedule that discretizes the given alpha_t_bar function, which defines the cumulative product of
+ (1-beta) over time from t = [0,1].
+
+ Contains a function alpha_bar that takes an argument t and transforms it to the cumulative product of (1-beta) up
+ to that part of the diffusion process.
+
+
+ Args:
+ num_diffusion_timesteps (`int`): the number of betas to produce.
+ max_beta (`float`): the maximum beta to use; use values lower than 1 to
+ prevent singularities.
+ alpha_transform_type (`str`, *optional*, default to `cosine`): the type of noise schedule for alpha_bar.
+ Choose from `cosine` or `exp`
+
+ Returns:
+ betas (`np.ndarray`): the betas used by the scheduler to step the model outputs
+ """
+ if alpha_transform_type == "cosine":
+
+ def alpha_bar_fn(t):
+ return math.cos((t + 0.008) / 1.008 * math.pi / 2) ** 2
+
+ elif alpha_transform_type == "exp":
+
+ def alpha_bar_fn(t):
+ return math.exp(t * -12.0)
+
+ else:
+ raise ValueError(f"Unsupported alpha_transform_type: {alpha_transform_type}")
+
+ betas = []
+ for i in range(num_diffusion_timesteps):
+ t1 = i / num_diffusion_timesteps
+ t2 = (i + 1) / num_diffusion_timesteps
+ betas.append(min(1 - alpha_bar_fn(t2) / alpha_bar_fn(t1), max_beta))
+ return torch.tensor(betas, dtype=torch.float32)
+
+
+# Copied from diffusers.schedulers.scheduling_ddim.rescale_zero_terminal_snr
+def rescale_zero_terminal_snr(betas):
+ """
+ Rescales betas to have zero terminal SNR Based on https://arxiv.org/pdf/2305.08891.pdf (Algorithm 1)
+
+
+ Args:
+ betas (`torch.Tensor`):
+ the betas that the scheduler is being initialized with.
+
+ Returns:
+ `torch.Tensor`: rescaled betas with zero terminal SNR
+ """
+ # Convert betas to alphas_bar_sqrt
+ alphas = 1.0 - betas
+ alphas_cumprod = torch.cumprod(alphas, dim=0)
+ alphas_bar_sqrt = alphas_cumprod.sqrt()
+
+ # Store old values.
+ alphas_bar_sqrt_0 = alphas_bar_sqrt[0].clone()
+ alphas_bar_sqrt_T = alphas_bar_sqrt[-1].clone()
+
+ # Shift so the last timestep is zero.
+ alphas_bar_sqrt -= alphas_bar_sqrt_T
+
+ # Scale so the first timestep is back to the old value.
+ alphas_bar_sqrt *= alphas_bar_sqrt_0 / (alphas_bar_sqrt_0 - alphas_bar_sqrt_T)
+
+ # Convert alphas_bar_sqrt to betas
+ alphas_bar = alphas_bar_sqrt**2 # Revert sqrt
+ alphas = alphas_bar[1:] / alphas_bar[:-1] # Revert cumprod
+ alphas = torch.cat([alphas_bar[0:1], alphas])
+ betas = 1 - alphas
+
+ return betas
+
+
+class MatryoshkaDDIMScheduler(SchedulerMixin, ConfigMixin):
+ """
+ `DDIMScheduler` extends the denoising procedure introduced in denoising diffusion probabilistic models (DDPMs) with
+ non-Markovian guidance.
+
+ This model inherits from [`SchedulerMixin`] and [`ConfigMixin`]. Check the superclass documentation for the generic
+ methods the library implements for all schedulers such as loading and saving.
+
+ Args:
+ num_train_timesteps (`int`, defaults to 1000):
+ The number of diffusion steps to train the model.
+ beta_start (`float`, defaults to 0.0001):
+ The starting `beta` value of inference.
+ beta_end (`float`, defaults to 0.02):
+ The final `beta` value.
+ beta_schedule (`str`, defaults to `"linear"`):
+ The beta schedule, a mapping from a beta range to a sequence of betas for stepping the model. Choose from
+ `linear`, `scaled_linear`, or `squaredcos_cap_v2`.
+ trained_betas (`np.ndarray`, *optional*):
+ Pass an array of betas directly to the constructor to bypass `beta_start` and `beta_end`.
+ clip_sample (`bool`, defaults to `True`):
+ Clip the predicted sample for numerical stability.
+ clip_sample_range (`float`, defaults to 1.0):
+ The maximum magnitude for sample clipping. Valid only when `clip_sample=True`.
+ set_alpha_to_one (`bool`, defaults to `True`):
+ Each diffusion step uses the alphas product value at that step and at the previous one. For the final step
+ there is no previous alpha. When this option is `True` the previous alpha product is fixed to `1`,
+ otherwise it uses the alpha value at step 0.
+ steps_offset (`int`, defaults to 0):
+ An offset added to the inference steps, as required by some model families.
+ prediction_type (`str`, defaults to `epsilon`, *optional*):
+ Prediction type of the scheduler function; can be `epsilon` (predicts the noise of the diffusion process),
+ `sample` (directly predicts the noisy sample`) or `v_prediction` (see section 2.4 of [Imagen
+ Video](https://imagen.research.google/video/paper.pdf) paper).
+ thresholding (`bool`, defaults to `False`):
+ Whether to use the "dynamic thresholding" method. This is unsuitable for latent-space diffusion models such
+ as Stable Diffusion.
+ dynamic_thresholding_ratio (`float`, defaults to 0.995):
+ The ratio for the dynamic thresholding method. Valid only when `thresholding=True`.
+ sample_max_value (`float`, defaults to 1.0):
+ The threshold value for dynamic thresholding. Valid only when `thresholding=True`.
+ timestep_spacing (`str`, defaults to `"leading"`):
+ The way the timesteps should be scaled. Refer to Table 2 of the [Common Diffusion Noise Schedules and
+ Sample Steps are Flawed](https://huggingface.co/papers/2305.08891) for more information.
+ rescale_betas_zero_snr (`bool`, defaults to `False`):
+ Whether to rescale the betas to have zero terminal SNR. This enables the model to generate very bright and
+ dark samples instead of limiting it to samples with medium brightness. Loosely related to
+ [`--offset_noise`](https://github.com/huggingface/diffusers/blob/74fd735eb073eb1d774b1ab4154a0876eb82f055/examples/dreambooth/train_dreambooth.py#L506).
+ """
+
+ order = 1
+
+ @register_to_config
+ def __init__(
+ self,
+ num_train_timesteps: int = 1000,
+ beta_start: float = 0.0001,
+ beta_end: float = 0.02,
+ beta_schedule: str = "linear",
+ trained_betas: Optional[Union[np.ndarray, List[float]]] = None,
+ clip_sample: bool = True,
+ set_alpha_to_one: bool = True,
+ steps_offset: int = 0,
+ prediction_type: str = "epsilon",
+ thresholding: bool = False,
+ dynamic_thresholding_ratio: float = 0.995,
+ clip_sample_range: float = 1.0,
+ sample_max_value: float = 1.0,
+ timestep_spacing: str = "leading",
+ rescale_betas_zero_snr: bool = False,
+ ):
+ if trained_betas is not None:
+ self.betas = torch.tensor(trained_betas, dtype=torch.float32)
+ elif beta_schedule == "linear":
+ self.betas = torch.linspace(beta_start, beta_end, num_train_timesteps, dtype=torch.float32)
+ elif beta_schedule == "scaled_linear":
+ # this schedule is very specific to the latent diffusion model.
+ self.betas = torch.linspace(beta_start**0.5, beta_end**0.5, num_train_timesteps, dtype=torch.float32) ** 2
+ elif beta_schedule == "squaredcos_cap_v2":
+ if self.config.timestep_spacing == "matryoshka_style":
+ self.betas = torch.cat((torch.tensor([0]), betas_for_alpha_bar(num_train_timesteps)))
+ else:
+ # Glide cosine schedule
+ self.betas = betas_for_alpha_bar(num_train_timesteps)
+ else:
+ raise NotImplementedError(f"{beta_schedule} is not implemented for {self.__class__}")
+
+ # Rescale for zero SNR
+ if rescale_betas_zero_snr:
+ self.betas = rescale_zero_terminal_snr(self.betas)
+
+ self.alphas = 1.0 - self.betas
+ self.alphas_cumprod = torch.cumprod(self.alphas, dim=0)
+
+ # At every step in ddim, we are looking into the previous alphas_cumprod
+ # For the final step, there is no previous alphas_cumprod because we are already at 0
+ # `set_alpha_to_one` decides whether we set this parameter simply to one or
+ # whether we use the final alpha of the "non-previous" one.
+ self.final_alpha_cumprod = torch.tensor(1.0) if set_alpha_to_one else self.alphas_cumprod[0]
+
+ # standard deviation of the initial noise distribution
+ self.init_noise_sigma = 1.0
+
+ # setable values
+ self.num_inference_steps = None
+ self.timesteps = torch.from_numpy(np.arange(0, num_train_timesteps)[::-1].copy().astype(np.int64))
+
+ self.scales = None
+
+ def scale_model_input(self, sample: torch.Tensor, timestep: Optional[int] = None) -> torch.Tensor:
+ """
+ Ensures interchangeability with schedulers that need to scale the denoising model input depending on the
+ current timestep.
+
+ Args:
+ sample (`torch.Tensor`):
+ The input sample.
+ timestep (`int`, *optional*):
+ The current timestep in the diffusion chain.
+
+ Returns:
+ `torch.Tensor`:
+ A scaled input sample.
+ """
+ return sample
+
+ def _get_variance(self, timestep, prev_timestep):
+ alpha_prod_t = self.alphas_cumprod[timestep]
+ alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
+ beta_prod_t = 1 - alpha_prod_t
+ beta_prod_t_prev = 1 - alpha_prod_t_prev
+
+ variance = (beta_prod_t_prev / beta_prod_t) * (1 - alpha_prod_t / alpha_prod_t_prev)
+
+ return variance
+
+ # Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler._threshold_sample
+ def _threshold_sample(self, sample: torch.Tensor) -> torch.Tensor:
+ """
+ "Dynamic thresholding: At each sampling step we set s to a certain percentile absolute pixel value in xt0 (the
+ prediction of x_0 at timestep t), and if s > 1, then we threshold xt0 to the range [-s, s] and then divide by
+ s. Dynamic thresholding pushes saturated pixels (those near -1 and 1) inwards, thereby actively preventing
+ pixels from saturation at each step. We find that dynamic thresholding results in significantly better
+ photorealism as well as better image-text alignment, especially when using very large guidance weights."
+
+ https://arxiv.org/abs/2205.11487
+ """
+ dtype = sample.dtype
+ batch_size, channels, *remaining_dims = sample.shape
+
+ if dtype not in (torch.float32, torch.float64):
+ sample = sample.float() # upcast for quantile calculation, and clamp not implemented for cpu half
+
+ # Flatten sample for doing quantile calculation along each image
+ sample = sample.reshape(batch_size, channels * np.prod(remaining_dims))
+
+ abs_sample = sample.abs() # "a certain percentile absolute pixel value"
+
+ s = torch.quantile(abs_sample, self.config.dynamic_thresholding_ratio, dim=1)
+ s = torch.clamp(
+ s, min=1, max=self.config.sample_max_value
+ ) # When clamped to min=1, equivalent to standard clipping to [-1, 1]
+ s = s.unsqueeze(1) # (batch_size, 1) because clamp will broadcast along dim=0
+ sample = torch.clamp(sample, -s, s) / s # "we threshold xt0 to the range [-s, s] and then divide by s"
+
+ sample = sample.reshape(batch_size, channels, *remaining_dims)
+ sample = sample.to(dtype)
+
+ return sample
+
+ def set_timesteps(self, num_inference_steps: int, device: Union[str, torch.device] = None):
+ """
+ Sets the discrete timesteps used for the diffusion chain (to be run before inference).
+
+ Args:
+ num_inference_steps (`int`):
+ The number of diffusion steps used when generating samples with a pre-trained model.
+ """
+
+ if num_inference_steps > self.config.num_train_timesteps:
+ raise ValueError(
+ f"`num_inference_steps`: {num_inference_steps} cannot be larger than `self.config.train_timesteps`:"
+ f" {self.config.num_train_timesteps} as the unet model trained with this scheduler can only handle"
+ f" maximal {self.config.num_train_timesteps} timesteps."
+ )
+
+ self.num_inference_steps = num_inference_steps
+
+ # "linspace", "leading", "trailing" corresponds to annotation of Table 2. of https://arxiv.org/abs/2305.08891
+ if self.config.timestep_spacing == "linspace":
+ timesteps = (
+ np.linspace(0, self.config.num_train_timesteps - 1, num_inference_steps)
+ .round()[::-1]
+ .copy()
+ .astype(np.int64)
+ )
+ elif self.config.timestep_spacing == "leading":
+ step_ratio = self.config.num_train_timesteps // self.num_inference_steps
+ # creates integer timesteps by multiplying by ratio
+ # casting to int to avoid issues when num_inference_step is power of 3
+ timesteps = (np.arange(0, num_inference_steps) * step_ratio).round()[::-1].copy().astype(np.int64)
+ timesteps += self.config.steps_offset
+ elif self.config.timestep_spacing == "trailing":
+ step_ratio = self.config.num_train_timesteps / self.num_inference_steps
+ # creates integer timesteps by multiplying by ratio
+ # casting to int to avoid issues when num_inference_step is power of 3
+ timesteps = np.round(np.arange(self.config.num_train_timesteps, 0, -step_ratio)).astype(np.int64)
+ timesteps -= 1
+ elif self.config.timestep_spacing == "matryoshka_style":
+ step_ratio = (self.config.num_train_timesteps + 1) / (num_inference_steps + 1)
+ timesteps = (np.arange(0, num_inference_steps + 1) * step_ratio).round()[::-1].copy().astype(np.int64)
+ else:
+ raise ValueError(
+ f"{self.config.timestep_spacing} is not supported. Please make sure to choose one of 'leading' or 'trailing'."
+ )
+
+ self.timesteps = torch.from_numpy(timesteps).to(device)
+
+ def get_schedule_shifted(self, alpha_prod, scale_factor=None):
+ if (scale_factor is not None) and (scale_factor > 1): # rescale noise schedule
+ snr = alpha_prod / (1 - alpha_prod)
+ scaled_snr = snr / scale_factor
+ alpha_prod = 1 / (1 + 1 / scaled_snr)
+ return alpha_prod
+
+ def step(
+ self,
+ model_output: torch.Tensor,
+ timestep: int,
+ sample: torch.Tensor,
+ eta: float = 0.0,
+ use_clipped_model_output: bool = False,
+ generator=None,
+ variance_noise: Optional[torch.Tensor] = None,
+ return_dict: bool = True,
+ ) -> Union[MatryoshkaDDIMSchedulerOutput, Tuple]:
+ """
+ Predict the sample from the previous timestep by reversing the SDE. This function propagates the diffusion
+ process from the learned model outputs (most often the predicted noise).
+
+ Args:
+ model_output (`torch.Tensor`):
+ The direct output from learned diffusion model.
+ timestep (`float`):
+ The current discrete timestep in the diffusion chain.
+ sample (`torch.Tensor`):
+ A current instance of a sample created by the diffusion process.
+ eta (`float`):
+ The weight of noise for added noise in diffusion step.
+ use_clipped_model_output (`bool`, defaults to `False`):
+ If `True`, computes "corrected" `model_output` from the clipped predicted original sample. Necessary
+ because predicted original sample is clipped to [-1, 1] when `self.config.clip_sample` is `True`. If no
+ clipping has happened, "corrected" `model_output` would coincide with the one provided as input and
+ `use_clipped_model_output` has no effect.
+ generator (`torch.Generator`, *optional*):
+ A random number generator.
+ variance_noise (`torch.Tensor`):
+ Alternative to generating noise with `generator` by directly providing the noise for the variance
+ itself. Useful for methods such as [`CycleDiffusion`].
+ return_dict (`bool`, *optional*, defaults to `True`):
+ Whether or not to return a [`~schedulers.scheduling_ddim.DDIMSchedulerOutput`] or `tuple`.
+
+ Returns:
+ [`~schedulers.scheduling_ddim.DDIMSchedulerOutput`] or `tuple`:
+ If return_dict is `True`, [`~schedulers.scheduling_ddim.DDIMSchedulerOutput`] is returned, otherwise a
+ tuple is returned where the first element is the sample tensor.
+
+ """
+ if self.num_inference_steps is None:
+ raise ValueError(
+ "Number of inference steps is 'None', you need to run 'set_timesteps' after creating the scheduler"
+ )
+
+ # See formulas (12) and (16) of DDIM paper https://arxiv.org/pdf/2010.02502.pdf
+ # Ideally, read DDIM paper in-detail understanding
+
+ # Notation ( ->
+ # - pred_noise_t -> e_theta(x_t, t)
+ # - pred_original_sample -> f_theta(x_t, t) or x_0
+ # - std_dev_t -> sigma_t
+ # - eta -> η
+ # - pred_sample_direction -> "direction pointing to x_t"
+ # - pred_prev_sample -> "x_t-1"
+
+ # 1. get previous step value (=t-1)
+ if self.config.timestep_spacing != "matryoshka_style":
+ prev_timestep = timestep - self.config.num_train_timesteps // self.num_inference_steps
+ else:
+ prev_timestep = self.timesteps[torch.nonzero(self.timesteps == timestep).item() + 1]
+
+ # 2. compute alphas, betas
+ alpha_prod_t = self.alphas_cumprod[timestep]
+ alpha_prod_t_prev = self.alphas_cumprod[prev_timestep] if prev_timestep >= 0 else self.final_alpha_cumprod
+
+ if self.config.timestep_spacing == "matryoshka_style" and len(model_output) > 1:
+ alpha_prod_t = torch.tensor([self.get_schedule_shifted(alpha_prod_t, s) for s in self.scales])
+ alpha_prod_t_prev = torch.tensor([self.get_schedule_shifted(alpha_prod_t_prev, s) for s in self.scales])
+
+ beta_prod_t = 1 - alpha_prod_t
+
+ # 3. compute predicted original sample from predicted noise also called
+ # "predicted x_0" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+ if self.config.prediction_type == "epsilon":
+ pred_original_sample = (sample - beta_prod_t ** (0.5) * model_output) / alpha_prod_t ** (0.5)
+ pred_epsilon = model_output
+ elif self.config.prediction_type == "sample":
+ pred_original_sample = model_output
+ pred_epsilon = (sample - alpha_prod_t ** (0.5) * pred_original_sample) / beta_prod_t ** (0.5)
+ elif self.config.prediction_type == "v_prediction":
+ if len(model_output) > 1:
+ pred_original_sample = []
+ pred_epsilon = []
+ for m_o, s, a_p_t, b_p_t in zip(model_output, sample, alpha_prod_t, beta_prod_t):
+ pred_original_sample.append((a_p_t**0.5) * s - (b_p_t**0.5) * m_o)
+ pred_epsilon.append((a_p_t**0.5) * m_o + (b_p_t**0.5) * s)
+ else:
+ pred_original_sample = (alpha_prod_t**0.5) * sample - (beta_prod_t**0.5) * model_output
+ pred_epsilon = (alpha_prod_t**0.5) * model_output + (beta_prod_t**0.5) * sample
+ else:
+ raise ValueError(
+ f"prediction_type given as {self.config.prediction_type} must be one of `epsilon`, `sample`, or"
+ " `v_prediction`"
+ )
+
+ # 4. Clip or threshold "predicted x_0"
+ if self.config.thresholding:
+ if len(model_output) > 1:
+ pred_original_sample = [
+ self._threshold_sample(p_o_s * scale) / scale
+ for p_o_s, scale in zip(pred_original_sample, self.scales)
+ ]
+ else:
+ pred_original_sample = self._threshold_sample(pred_original_sample)
+ elif self.config.clip_sample:
+ if len(model_output) > 1:
+ pred_original_sample = [
+ (p_o_s * scale).clamp(-self.config.clip_sample_range, self.config.clip_sample_range) / scale
+ for p_o_s, scale in zip(pred_original_sample, self.scales)
+ ]
+ else:
+ pred_original_sample = pred_original_sample.clamp(
+ -self.config.clip_sample_range, self.config.clip_sample_range
+ )
+
+ # 5. compute variance: "sigma_t(η)" -> see formula (16)
+ # σ_t = sqrt((1 − α_t−1)/(1 − α_t)) * sqrt(1 − α_t/α_t−1)
+ variance = self._get_variance(timestep, prev_timestep)
+ std_dev_t = eta * variance ** (0.5)
+
+ if use_clipped_model_output:
+ # the pred_epsilon is always re-derived from the clipped x_0 in Glide
+ if len(model_output) > 1:
+ pred_epsilon = []
+ for s, a_p_t, p_o_s, b_p_t in zip(sample, alpha_prod_t, pred_original_sample, beta_prod_t):
+ pred_epsilon.append((s - a_p_t ** (0.5) * p_o_s) / b_p_t ** (0.5))
+ else:
+ pred_epsilon = (sample - alpha_prod_t ** (0.5) * pred_original_sample) / beta_prod_t ** (0.5)
+
+ # 6. compute "direction pointing to x_t" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+ if len(model_output) > 1:
+ pred_sample_direction = []
+ for p_e, a_p_t_p in zip(pred_epsilon, alpha_prod_t_prev):
+ pred_sample_direction.append((1 - a_p_t_p - std_dev_t**2) ** (0.5) * p_e)
+ else:
+ pred_sample_direction = (1 - alpha_prod_t_prev - std_dev_t**2) ** (0.5) * pred_epsilon
+
+ # 7. compute x_t without "random noise" of formula (12) from https://arxiv.org/pdf/2010.02502.pdf
+ if len(model_output) > 1:
+ prev_sample = []
+ for p_o_s, p_s_d, a_p_t_p in zip(pred_original_sample, pred_sample_direction, alpha_prod_t_prev):
+ prev_sample.append(a_p_t_p ** (0.5) * p_o_s + p_s_d)
+ else:
+ prev_sample = alpha_prod_t_prev ** (0.5) * pred_original_sample + pred_sample_direction
+
+ if eta > 0:
+ if variance_noise is not None and generator is not None:
+ raise ValueError(
+ "Cannot pass both generator and variance_noise. Please make sure that either `generator` or"
+ " `variance_noise` stays `None`."
+ )
+
+ if variance_noise is None:
+ if len(model_output) > 1:
+ variance_noise = []
+ for m_o in model_output:
+ variance_noise.append(
+ randn_tensor(m_o.shape, generator=generator, device=m_o.device, dtype=m_o.dtype)
+ )
+ else:
+ variance_noise = randn_tensor(
+ model_output.shape, generator=generator, device=model_output.device, dtype=model_output.dtype
+ )
+ if len(model_output) > 1:
+ prev_sample = [p_s + std_dev_t * v_n for v_n, p_s in zip(variance_noise, prev_sample)]
+ else:
+ variance = std_dev_t * variance_noise
+
+ prev_sample = prev_sample + variance
+
+ if not return_dict:
+ return (prev_sample,)
+
+ return MatryoshkaDDIMSchedulerOutput(prev_sample=prev_sample, pred_original_sample=pred_original_sample)
+
+ # Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler.add_noise
+ def add_noise(
+ self,
+ original_samples: torch.Tensor,
+ noise: torch.Tensor,
+ timesteps: torch.IntTensor,
+ ) -> torch.Tensor:
+ # Make sure alphas_cumprod and timestep have same device and dtype as original_samples
+ # Move the self.alphas_cumprod to device to avoid redundant CPU to GPU data movement
+ # for the subsequent add_noise calls
+ self.alphas_cumprod = self.alphas_cumprod.to(device=original_samples.device)
+ alphas_cumprod = self.alphas_cumprod.to(dtype=original_samples.dtype)
+ timesteps = timesteps.to(original_samples.device)
+
+ sqrt_alpha_prod = alphas_cumprod[timesteps] ** 0.5
+ sqrt_alpha_prod = sqrt_alpha_prod.flatten()
+ while len(sqrt_alpha_prod.shape) < len(original_samples.shape):
+ sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze(-1)
+
+ sqrt_one_minus_alpha_prod = (1 - alphas_cumprod[timesteps]) ** 0.5
+ sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten()
+ while len(sqrt_one_minus_alpha_prod.shape) < len(original_samples.shape):
+ sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze(-1)
+
+ noisy_samples = sqrt_alpha_prod * original_samples + sqrt_one_minus_alpha_prod * noise
+ return noisy_samples
+
+ # Copied from diffusers.schedulers.scheduling_ddpm.DDPMScheduler.get_velocity
+ def get_velocity(self, sample: torch.Tensor, noise: torch.Tensor, timesteps: torch.IntTensor) -> torch.Tensor:
+ # Make sure alphas_cumprod and timestep have same device and dtype as sample
+ self.alphas_cumprod = self.alphas_cumprod.to(device=sample.device)
+ alphas_cumprod = self.alphas_cumprod.to(dtype=sample.dtype)
+ timesteps = timesteps.to(sample.device)
+
+ sqrt_alpha_prod = alphas_cumprod[timesteps] ** 0.5
+ sqrt_alpha_prod = sqrt_alpha_prod.flatten()
+ while len(sqrt_alpha_prod.shape) < len(sample.shape):
+ sqrt_alpha_prod = sqrt_alpha_prod.unsqueeze(-1)
+
+ sqrt_one_minus_alpha_prod = (1 - alphas_cumprod[timesteps]) ** 0.5
+ sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.flatten()
+ while len(sqrt_one_minus_alpha_prod.shape) < len(sample.shape):
+ sqrt_one_minus_alpha_prod = sqrt_one_minus_alpha_prod.unsqueeze(-1)
+
+ velocity = sqrt_alpha_prod * noise - sqrt_one_minus_alpha_prod * sample
+ return velocity
+
+ def __len__(self):
+ return self.config.num_train_timesteps
+
+
+class CrossAttnDownBlock2D(nn.Module):
+ def __init__(
+ self,
+ in_channels: int,
+ out_channels: int,
+ temb_channels: int,
+ dropout: float = 0.0,
+ num_layers: int = 1,
+ transformer_layers_per_block: Union[int, Tuple[int]] = 1,
+ resnet_eps: float = 1e-6,
+ resnet_time_scale_shift: str = "default",
+ resnet_act_fn: str = "swish",
+ resnet_groups: int = 32,
+ resnet_pre_norm: bool = True,
+ norm_type: str = "layer_norm",
+ num_attention_heads: int = 1,
+ cross_attention_dim: int = 1280,
+ cross_attention_norm: Optional[str] = None,
+ output_scale_factor: float = 1.0,
+ downsample_padding: int = 1,
+ add_downsample: bool = True,
+ dual_cross_attention: bool = False,
+ use_linear_projection: bool = False,
+ only_cross_attention: bool = False,
+ upcast_attention: bool = False,
+ attention_type: str = "default",
+ attention_pre_only: bool = False,
+ attention_bias: bool = False,
+ use_attention_ffn: bool = True,
+ ):
+ super().__init__()
+ resnets = []
+ attentions = []
+
+ self.has_cross_attention = True
+ self.num_attention_heads = num_attention_heads
+ if isinstance(transformer_layers_per_block, int):
+ transformer_layers_per_block = [transformer_layers_per_block] * num_layers
+
+ for i in range(num_layers):
+ in_channels = in_channels if i == 0 else out_channels
+ resnets.append(
+ ResnetBlock2D(
+ in_channels=in_channels,
+ out_channels=out_channels,
+ temb_channels=temb_channels,
+ eps=resnet_eps,
+ groups=resnet_groups,
+ dropout=dropout,
+ time_embedding_norm=resnet_time_scale_shift,
+ non_linearity=resnet_act_fn,
+ output_scale_factor=output_scale_factor,
+ pre_norm=resnet_pre_norm,
+ )
+ )
+ attentions.append(
+ MatryoshkaTransformer2DModel(
+ num_attention_heads,
+ out_channels // num_attention_heads,
+ in_channels=out_channels,
+ num_layers=transformer_layers_per_block[i],
+ cross_attention_dim=cross_attention_dim,
+ upcast_attention=upcast_attention,
+ use_attention_ffn=use_attention_ffn,
+ )
+ )
+ self.attentions = nn.ModuleList(attentions)
+ self.resnets = nn.ModuleList(resnets)
+
+ if add_downsample:
+ self.downsamplers = nn.ModuleList(
+ [
+ Downsample2D(
+ out_channels, use_conv=True, out_channels=out_channels, padding=downsample_padding, name="op"
+ )
+ ]
+ )
+ else:
+ self.downsamplers = None
+
+ self.gradient_checkpointing = False
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ temb: Optional[torch.Tensor] = None,
+ encoder_hidden_states: Optional[torch.Tensor] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ additional_residuals: Optional[torch.Tensor] = None,
+ ) -> Tuple[torch.Tensor, Tuple[torch.Tensor, ...]]:
+ if cross_attention_kwargs is not None:
+ if cross_attention_kwargs.get("scale", None) is not None:
+ logger.warning("Passing `scale` to `cross_attention_kwargs` is deprecated. `scale` will be ignored.")
+
+ output_states = ()
+
+ blocks = list(zip(self.resnets, self.attentions))
+
+ for i, (resnet, attn) in enumerate(blocks):
+ if self.training and self.gradient_checkpointing:
+
+ def create_custom_forward(module, return_dict=None):
+ def custom_forward(*inputs):
+ if return_dict is not None:
+ return module(*inputs, return_dict=return_dict)
+ else:
+ return module(*inputs)
+
+ return custom_forward
+
+ ckpt_kwargs: Dict[str, Any] = {"use_reentrant": False} if is_torch_version(">=", "1.11.0") else {}
+ hidden_states = torch.utils.checkpoint.checkpoint(
+ create_custom_forward(resnet),
+ hidden_states,
+ temb,
+ **ckpt_kwargs,
+ )
+ hidden_states = attn(
+ hidden_states,
+ encoder_hidden_states=encoder_hidden_states,
+ cross_attention_kwargs=cross_attention_kwargs,
+ attention_mask=attention_mask,
+ encoder_attention_mask=encoder_attention_mask,
+ return_dict=False,
+ )[0]
+ else:
+ hidden_states = resnet(hidden_states, temb)
+ hidden_states = attn(
+ hidden_states,
+ encoder_hidden_states=encoder_hidden_states,
+ cross_attention_kwargs=cross_attention_kwargs,
+ attention_mask=attention_mask,
+ encoder_attention_mask=encoder_attention_mask,
+ return_dict=False,
+ )[0]
+
+ # apply additional residuals to the output of the last pair of resnet and attention blocks
+ if i == len(blocks) - 1 and additional_residuals is not None:
+ hidden_states = hidden_states + additional_residuals
+
+ output_states = output_states + (hidden_states,)
+
+ if self.downsamplers is not None:
+ for downsampler in self.downsamplers:
+ hidden_states = downsampler(hidden_states)
+
+ output_states = output_states + (hidden_states,)
+
+ return hidden_states, output_states
+
+
+class UNetMidBlock2DCrossAttn(nn.Module):
+ def __init__(
+ self,
+ in_channels: int,
+ temb_channels: int,
+ out_channels: Optional[int] = None,
+ dropout: float = 0.0,
+ num_layers: int = 1,
+ transformer_layers_per_block: Union[int, Tuple[int]] = 1,
+ resnet_eps: float = 1e-6,
+ resnet_time_scale_shift: str = "default",
+ resnet_act_fn: str = "swish",
+ resnet_groups: int = 32,
+ resnet_groups_out: Optional[int] = None,
+ resnet_pre_norm: bool = True,
+ norm_type: str = "layer_norm",
+ num_attention_heads: int = 1,
+ output_scale_factor: float = 1.0,
+ cross_attention_dim: int = 1280,
+ cross_attention_norm: Optional[str] = None,
+ dual_cross_attention: bool = False,
+ use_linear_projection: bool = False,
+ upcast_attention: bool = False,
+ attention_type: str = "default",
+ attention_pre_only: bool = False,
+ attention_bias: bool = False,
+ use_attention_ffn: bool = True,
+ ):
+ super().__init__()
+
+ out_channels = out_channels or in_channels
+ self.in_channels = in_channels
+ self.out_channels = out_channels
+
+ self.has_cross_attention = True
+ self.num_attention_heads = num_attention_heads
+ resnet_groups = resnet_groups if resnet_groups is not None else min(in_channels // 4, 32)
+
+ # support for variable transformer layers per block
+ if isinstance(transformer_layers_per_block, int):
+ transformer_layers_per_block = [transformer_layers_per_block] * num_layers
+
+ resnet_groups_out = resnet_groups_out or resnet_groups
+
+ # there is always at least one resnet
+ resnets = [
+ ResnetBlock2D(
+ in_channels=in_channels,
+ out_channels=out_channels,
+ temb_channels=temb_channels,
+ eps=resnet_eps,
+ groups=resnet_groups,
+ groups_out=resnet_groups_out,
+ dropout=dropout,
+ time_embedding_norm=resnet_time_scale_shift,
+ non_linearity=resnet_act_fn,
+ output_scale_factor=output_scale_factor,
+ pre_norm=resnet_pre_norm,
+ )
+ ]
+ attentions = []
+
+ for i in range(num_layers):
+ attentions.append(
+ MatryoshkaTransformer2DModel(
+ num_attention_heads,
+ out_channels // num_attention_heads,
+ in_channels=out_channels,
+ num_layers=transformer_layers_per_block[i],
+ cross_attention_dim=cross_attention_dim,
+ upcast_attention=upcast_attention,
+ use_attention_ffn=use_attention_ffn,
+ )
+ )
+ resnets.append(
+ ResnetBlock2D(
+ in_channels=out_channels,
+ out_channels=out_channels,
+ temb_channels=temb_channels,
+ eps=resnet_eps,
+ groups=resnet_groups_out,
+ dropout=dropout,
+ time_embedding_norm=resnet_time_scale_shift,
+ non_linearity=resnet_act_fn,
+ output_scale_factor=output_scale_factor,
+ pre_norm=resnet_pre_norm,
+ )
+ )
+
+ self.attentions = nn.ModuleList(attentions)
+ self.resnets = nn.ModuleList(resnets)
+
+ self.gradient_checkpointing = False
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ temb: Optional[torch.Tensor] = None,
+ encoder_hidden_states: Optional[torch.Tensor] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ ) -> torch.Tensor:
+ if cross_attention_kwargs is not None:
+ if cross_attention_kwargs.get("scale", None) is not None:
+ logger.warning("Passing `scale` to `cross_attention_kwargs` is deprecated. `scale` will be ignored.")
+
+ hidden_states = self.resnets[0](hidden_states, temb)
+ for attn, resnet in zip(self.attentions, self.resnets[1:]):
+ if self.training and self.gradient_checkpointing:
+
+ def create_custom_forward(module, return_dict=None):
+ def custom_forward(*inputs):
+ if return_dict is not None:
+ return module(*inputs, return_dict=return_dict)
+ else:
+ return module(*inputs)
+
+ return custom_forward
+
+ ckpt_kwargs: Dict[str, Any] = {"use_reentrant": False} if is_torch_version(">=", "1.11.0") else {}
+ hidden_states = attn(
+ hidden_states,
+ encoder_hidden_states=encoder_hidden_states,
+ cross_attention_kwargs=cross_attention_kwargs,
+ attention_mask=attention_mask,
+ encoder_attention_mask=encoder_attention_mask,
+ return_dict=False,
+ )[0]
+ hidden_states = torch.utils.checkpoint.checkpoint(
+ create_custom_forward(resnet),
+ hidden_states,
+ temb,
+ **ckpt_kwargs,
+ )
+ else:
+ hidden_states = attn(
+ hidden_states,
+ encoder_hidden_states=encoder_hidden_states,
+ cross_attention_kwargs=cross_attention_kwargs,
+ attention_mask=attention_mask,
+ encoder_attention_mask=encoder_attention_mask,
+ return_dict=False,
+ )[0]
+ hidden_states = resnet(hidden_states, temb)
+
+ return hidden_states
+
+
+class CrossAttnUpBlock2D(nn.Module):
+ def __init__(
+ self,
+ in_channels: int,
+ out_channels: int,
+ prev_output_channel: int,
+ temb_channels: int,
+ resolution_idx: Optional[int] = None,
+ dropout: float = 0.0,
+ num_layers: int = 1,
+ transformer_layers_per_block: Union[int, Tuple[int]] = 1,
+ resnet_eps: float = 1e-6,
+ resnet_time_scale_shift: str = "default",
+ resnet_act_fn: str = "swish",
+ resnet_groups: int = 32,
+ resnet_pre_norm: bool = True,
+ norm_type: str = "layer_norm",
+ num_attention_heads: int = 1,
+ cross_attention_dim: int = 1280,
+ cross_attention_norm: Optional[str] = None,
+ output_scale_factor: float = 1.0,
+ add_upsample: bool = True,
+ dual_cross_attention: bool = False,
+ use_linear_projection: bool = False,
+ only_cross_attention: bool = False,
+ upcast_attention: bool = False,
+ attention_type: str = "default",
+ attention_pre_only: bool = False,
+ attention_bias: bool = False,
+ use_attention_ffn: bool = True,
+ ):
+ super().__init__()
+ resnets = []
+ attentions = []
+
+ self.has_cross_attention = True
+ self.num_attention_heads = num_attention_heads
+
+ if isinstance(transformer_layers_per_block, int):
+ transformer_layers_per_block = [transformer_layers_per_block] * num_layers
+
+ for i in range(num_layers):
+ res_skip_channels = in_channels if (i == num_layers - 1) else out_channels
+ resnet_in_channels = prev_output_channel if i == 0 else out_channels
+
+ resnets.append(
+ ResnetBlock2D(
+ in_channels=resnet_in_channels + res_skip_channels,
+ out_channels=out_channels,
+ temb_channels=temb_channels,
+ eps=resnet_eps,
+ groups=resnet_groups,
+ dropout=dropout,
+ time_embedding_norm=resnet_time_scale_shift,
+ non_linearity=resnet_act_fn,
+ output_scale_factor=output_scale_factor,
+ pre_norm=resnet_pre_norm,
+ )
+ )
+ attentions.append(
+ MatryoshkaTransformer2DModel(
+ num_attention_heads,
+ out_channels // num_attention_heads,
+ in_channels=out_channels,
+ num_layers=transformer_layers_per_block[i],
+ cross_attention_dim=cross_attention_dim,
+ upcast_attention=upcast_attention,
+ use_attention_ffn=use_attention_ffn,
+ )
+ )
+ self.attentions = nn.ModuleList(attentions)
+ self.resnets = nn.ModuleList(resnets)
+
+ if add_upsample:
+ self.upsamplers = nn.ModuleList([Upsample2D(out_channels, use_conv=True, out_channels=out_channels)])
+ else:
+ self.upsamplers = None
+
+ self.gradient_checkpointing = False
+ self.resolution_idx = resolution_idx
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ res_hidden_states_tuple: Tuple[torch.Tensor, ...],
+ temb: Optional[torch.Tensor] = None,
+ encoder_hidden_states: Optional[torch.Tensor] = None,
+ cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+ upsample_size: Optional[int] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ ) -> torch.Tensor:
+ if cross_attention_kwargs is not None:
+ if cross_attention_kwargs.get("scale", None) is not None:
+ logger.warning("Passing `scale` to `cross_attention_kwargs` is deprecated. `scale` will be ignored.")
+
+ is_freeu_enabled = (
+ getattr(self, "s1", None)
+ and getattr(self, "s2", None)
+ and getattr(self, "b1", None)
+ and getattr(self, "b2", None)
+ )
+
+ for resnet, attn in zip(self.resnets, self.attentions):
+ # pop res hidden states
+ res_hidden_states = res_hidden_states_tuple[-1]
+ res_hidden_states_tuple = res_hidden_states_tuple[:-1]
+
+ # FreeU: Only operate on the first two stages
+ if is_freeu_enabled:
+ hidden_states, res_hidden_states = apply_freeu(
+ self.resolution_idx,
+ hidden_states,
+ res_hidden_states,
+ s1=self.s1,
+ s2=self.s2,
+ b1=self.b1,
+ b2=self.b2,
+ )
+
+ hidden_states = torch.cat([hidden_states, res_hidden_states], dim=1)
+
+ if self.training and self.gradient_checkpointing:
+
+ def create_custom_forward(module, return_dict=None):
+ def custom_forward(*inputs):
+ if return_dict is not None:
+ return module(*inputs, return_dict=return_dict)
+ else:
+ return module(*inputs)
+
+ return custom_forward
+
+ ckpt_kwargs: Dict[str, Any] = {"use_reentrant": False} if is_torch_version(">=", "1.11.0") else {}
+ hidden_states = torch.utils.checkpoint.checkpoint(
+ create_custom_forward(resnet),
+ hidden_states,
+ temb,
+ **ckpt_kwargs,
+ )
+ hidden_states = attn(
+ hidden_states,
+ encoder_hidden_states=encoder_hidden_states,
+ cross_attention_kwargs=cross_attention_kwargs,
+ attention_mask=attention_mask,
+ encoder_attention_mask=encoder_attention_mask,
+ return_dict=False,
+ )[0]
+ else:
+ hidden_states = resnet(hidden_states, temb)
+ hidden_states = attn(
+ hidden_states,
+ encoder_hidden_states=encoder_hidden_states,
+ cross_attention_kwargs=cross_attention_kwargs,
+ attention_mask=attention_mask,
+ encoder_attention_mask=encoder_attention_mask,
+ return_dict=False,
+ )[0]
+
+ if self.upsamplers is not None:
+ for upsampler in self.upsamplers:
+ hidden_states = upsampler(hidden_states, upsample_size)
+
+ return hidden_states
+
+
+@dataclass
+class MatryoshkaTransformer2DModelOutput(BaseOutput):
+ """
+ The output of [`MatryoshkaTransformer2DModel`].
+
+ Args:
+ sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)` or `(batch size, num_vector_embeds - 1, num_latent_pixels)` if [`MatryoshkaTransformer2DModel`] is discrete):
+ The hidden states output conditioned on the `encoder_hidden_states` input. If discrete, returns probability
+ distributions for the unnoised latent pixels.
+ """
+
+ sample: "torch.Tensor" # noqa: F821
+
+
+class MatryoshkaTransformer2DModel(LegacyModelMixin, LegacyConfigMixin):
+ _supports_gradient_checkpointing = True
+ _no_split_modules = ["MatryoshkaTransformerBlock"]
+
+ @register_to_config
+ def __init__(
+ self,
+ num_attention_heads: int = 16,
+ attention_head_dim: int = 88,
+ in_channels: Optional[int] = None,
+ num_layers: int = 1,
+ cross_attention_dim: Optional[int] = None,
+ upcast_attention: bool = False,
+ use_attention_ffn: bool = True,
+ ):
+ super().__init__()
+ self.in_channels = self.config.num_attention_heads * self.config.attention_head_dim
+ self.gradient_checkpointing = False
+
+ self.transformer_blocks = nn.ModuleList(
+ [
+ MatryoshkaTransformerBlock(
+ self.in_channels,
+ self.config.num_attention_heads,
+ self.config.attention_head_dim,
+ cross_attention_dim=self.config.cross_attention_dim,
+ upcast_attention=self.config.upcast_attention,
+ use_attention_ffn=self.config.use_attention_ffn,
+ )
+ for _ in range(self.config.num_layers)
+ ]
+ )
+
+ def _set_gradient_checkpointing(self, module, value=False):
+ if hasattr(module, "gradient_checkpointing"):
+ module.gradient_checkpointing = value
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ encoder_hidden_states: Optional[torch.Tensor] = None,
+ timestep: Optional[torch.LongTensor] = None,
+ added_cond_kwargs: Dict[str, torch.Tensor] = None,
+ class_labels: Optional[torch.LongTensor] = None,
+ cross_attention_kwargs: Dict[str, Any] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ return_dict: bool = True,
+ ):
+ """
+ The [`MatryoshkaTransformer2DModel`] forward method.
+
+ Args:
+ hidden_states (`torch.LongTensor` of shape `(batch size, num latent pixels)` if discrete, `torch.Tensor` of shape `(batch size, channel, height, width)` if continuous):
+ Input `hidden_states`.
+ encoder_hidden_states ( `torch.Tensor` of shape `(batch size, sequence len, embed dims)`, *optional*):
+ Conditional embeddings for cross attention layer. If not given, cross-attention defaults to
+ self-attention.
+ timestep ( `torch.LongTensor`, *optional*):
+ Used to indicate denoising step. Optional timestep to be applied as an embedding in `AdaLayerNorm`.
+ class_labels ( `torch.LongTensor` of shape `(batch size, num classes)`, *optional*):
+ Used to indicate class labels conditioning. Optional class labels to be applied as an embedding in
+ `AdaLayerZeroNorm`.
+ cross_attention_kwargs ( `Dict[str, Any]`, *optional*):
+ A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
+ `self.processor` in
+ [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+ attention_mask ( `torch.Tensor`, *optional*):
+ An attention mask of shape `(batch, key_tokens)` is applied to `encoder_hidden_states`. If `1` the mask
+ is kept, otherwise if `0` it is discarded. Mask will be converted into a bias, which adds large
+ negative values to the attention scores corresponding to "discard" tokens.
+ encoder_attention_mask ( `torch.Tensor`, *optional*):
+ Cross-attention mask applied to `encoder_hidden_states`. Two formats supported:
+
+ * Mask `(batch, sequence_length)` True = keep, False = discard.
+ * Bias `(batch, 1, sequence_length)` 0 = keep, -10000 = discard.
+
+ If `ndim == 2`: will be interpreted as a mask, then converted into a bias consistent with the format
+ above. This bias will be added to the cross-attention scores.
+ return_dict (`bool`, *optional*, defaults to `True`):
+ Whether or not to return a [`~NestedUNet2DConditionOutput`] instead of a plain
+ tuple.
+
+ Returns:
+ If `return_dict` is True, an [`~MatryoshkaTransformer2DModelOutput`] is returned,
+ otherwise a `tuple` where the first element is the sample tensor.
+ """
+ if cross_attention_kwargs is not None:
+ if cross_attention_kwargs.get("scale", None) is not None:
+ logger.warning("Passing `scale` to `cross_attention_kwargs` is deprecated. `scale` will be ignored.")
+ # ensure attention_mask is a bias, and give it a singleton query_tokens dimension.
+ # we may have done this conversion already, e.g. if we came here via UNet2DConditionModel#forward.
+ # we can tell by counting dims; if ndim == 2: it's a mask rather than a bias.
+ # expects mask of shape:
+ # [batch, key_tokens]
+ # adds singleton query_tokens dimension:
+ # [batch, 1, key_tokens]
+ # this helps to broadcast it as a bias over attention scores, which will be in one of the following shapes:
+ # [batch, heads, query_tokens, key_tokens] (e.g. torch sdp attn)
+ # [batch * heads, query_tokens, key_tokens] (e.g. xformers or classic attn)
+ if attention_mask is not None and attention_mask.ndim == 2:
+ # assume that mask is expressed as:
+ # (1 = keep, 0 = discard)
+ # convert mask into a bias that can be added to attention scores:
+ # (keep = +0, discard = -10000.0)
+ attention_mask = (1 - attention_mask.to(hidden_states.dtype)) * -10000.0
+ attention_mask = attention_mask.unsqueeze(1)
+
+ # convert encoder_attention_mask to a bias the same way we do for attention_mask
+ if encoder_attention_mask is not None and encoder_attention_mask.ndim == 2:
+ encoder_attention_mask = (1 - encoder_attention_mask.to(hidden_states.dtype)) * -10000.0
+ encoder_attention_mask = encoder_attention_mask.unsqueeze(1)
+
+ # Blocks
+ for block in self.transformer_blocks:
+ if self.training and self.gradient_checkpointing:
+
+ def create_custom_forward(module, return_dict=None):
+ def custom_forward(*inputs):
+ if return_dict is not None:
+ return module(*inputs, return_dict=return_dict)
+ else:
+ return module(*inputs)
+
+ return custom_forward
+
+ ckpt_kwargs: Dict[str, Any] = {"use_reentrant": False} if is_torch_version(">=", "1.11.0") else {}
+ hidden_states = torch.utils.checkpoint.checkpoint(
+ create_custom_forward(block),
+ hidden_states,
+ attention_mask,
+ encoder_hidden_states,
+ encoder_attention_mask,
+ timestep,
+ cross_attention_kwargs,
+ class_labels,
+ **ckpt_kwargs,
+ )
+ else:
+ hidden_states = block(
+ hidden_states,
+ attention_mask=attention_mask,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=encoder_attention_mask,
+ timestep=timestep,
+ cross_attention_kwargs=cross_attention_kwargs,
+ class_labels=class_labels,
+ )
+
+ # Output
+ output = hidden_states
+
+ if not return_dict:
+ return (output,)
+
+ return MatryoshkaTransformer2DModelOutput(sample=output)
+
+
+class MatryoshkaTransformerBlock(nn.Module):
+ r"""
+ Matryoshka Transformer block.
+
+ Parameters:
+ """
+
+ def __init__(
+ self,
+ dim: int,
+ num_attention_heads: int,
+ attention_head_dim: int,
+ cross_attention_dim: Optional[int] = None,
+ upcast_attention: bool = False,
+ use_attention_ffn: bool = True,
+ ):
+ super().__init__()
+ self.dim = dim
+ self.num_attention_heads = num_attention_heads
+ self.attention_head_dim = attention_head_dim
+ self.cross_attention_dim = cross_attention_dim
+
+ # Define 3 blocks.
+ # 1. Self-Attn
+ self.attn1 = Attention(
+ query_dim=dim,
+ cross_attention_dim=None,
+ heads=num_attention_heads,
+ dim_head=attention_head_dim,
+ norm_num_groups=32,
+ bias=True,
+ upcast_attention=upcast_attention,
+ pre_only=True,
+ processor=MatryoshkaFusedAttnProcessor2_0(),
+ )
+ self.attn1.fuse_projections()
+ del self.attn1.to_q
+ del self.attn1.to_k
+ del self.attn1.to_v
+
+ # 2. Cross-Attn
+ if cross_attention_dim is not None and cross_attention_dim > 0:
+ self.attn2 = Attention(
+ query_dim=dim,
+ cross_attention_dim=cross_attention_dim,
+ cross_attention_norm="layer_norm",
+ heads=num_attention_heads,
+ dim_head=attention_head_dim,
+ bias=True,
+ upcast_attention=upcast_attention,
+ pre_only=True,
+ processor=MatryoshkaFusedAttnProcessor2_0(),
+ )
+ self.attn2.fuse_projections()
+ del self.attn2.to_q
+ del self.attn2.to_k
+ del self.attn2.to_v
+
+ self.proj_out = nn.Linear(dim, dim)
+
+ if use_attention_ffn:
+ # 3. Feed-forward
+ self.ff = MatryoshkaFeedForward(dim)
+ else:
+ self.ff = None
+
+ # let chunk size default to None
+ self._chunk_size = None
+ self._chunk_dim = 0
+
+ # Copied from diffusers.models.attention.BasicTransformerBlock.set_chunk_feed_forward
+ def set_chunk_feed_forward(self, chunk_size: Optional[int], dim: int = 0):
+ # Sets chunk feed-forward
+ self._chunk_size = chunk_size
+ self._chunk_dim = dim
+
+ def forward(
+ self,
+ hidden_states: torch.Tensor,
+ attention_mask: Optional[torch.Tensor] = None,
+ encoder_hidden_states: Optional[torch.Tensor] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ timestep: Optional[torch.LongTensor] = None,
+ cross_attention_kwargs: Dict[str, Any] = None,
+ class_labels: Optional[torch.LongTensor] = None,
+ added_cond_kwargs: Optional[Dict[str, torch.Tensor]] = None,
+ ) -> torch.Tensor:
+ if cross_attention_kwargs is not None:
+ if cross_attention_kwargs.get("scale", None) is not None:
+ logger.warning("Passing `scale` to `cross_attention_kwargs` is deprecated. `scale` will be ignored.")
+
+ # 1. Self-Attention
+ batch_size, channels, *spatial_dims = hidden_states.shape
+
+ attn_output, query = self.attn1(
+ hidden_states,
+ # **cross_attention_kwargs,
+ )
+
+ # 2. Cross-Attention
+ if self.cross_attention_dim is not None and self.cross_attention_dim > 0:
+ attn_output_cond = self.attn2(
+ hidden_states,
+ encoder_hidden_states=encoder_hidden_states,
+ attention_mask=encoder_attention_mask,
+ self_attention_output=attn_output,
+ self_attention_query=query,
+ # **cross_attention_kwargs,
+ )
+
+ attn_output_cond = self.proj_out(attn_output_cond)
+ attn_output_cond = attn_output_cond.permute(0, 2, 1).reshape(batch_size, channels, *spatial_dims)
+ hidden_states = hidden_states + attn_output_cond
+
+ if self.ff is not None:
+ # 3. Feed-forward
+ if self._chunk_size is not None:
+ # "feed_forward_chunk_size" can be used to save memory
+ ff_output = _chunked_feed_forward(self.ff, hidden_states, self._chunk_dim, self._chunk_size)
+ else:
+ ff_output = self.ff(hidden_states)
+
+ hidden_states = ff_output + hidden_states
+
+ return hidden_states
+
+
+class MatryoshkaFusedAttnProcessor2_0:
+ r"""
+ Processor for implementing scaled dot-product attention (enabled by default if you're using PyTorch 2.0). It uses
+ fused projection layers. For self-attention modules, all projection matrices (i.e., query, key, value) are fused.
+ For cross-attention modules, key and value projection matrices are fused.
+
+
+
+ This API is currently 🧪 experimental in nature and can change in future.
+
+
+ """
+
+ def __init__(self):
+ if not hasattr(F, "scaled_dot_product_attention"):
+ raise ImportError(
+ "MatryoshkaFusedAttnProcessor2_0 requires PyTorch 2.x, to use it. Please upgrade PyTorch to > 2.x."
+ )
+
+ def __call__(
+ self,
+ attn: Attention,
+ hidden_states: torch.Tensor,
+ encoder_hidden_states: Optional[torch.Tensor] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ temb: Optional[torch.Tensor] = None,
+ self_attention_query: Optional[torch.Tensor] = None,
+ self_attention_output: Optional[torch.Tensor] = None,
+ *args,
+ **kwargs,
+ ) -> torch.Tensor:
+ if len(args) > 0 or kwargs.get("scale", None) is not None:
+ deprecation_message = "The `scale` argument is deprecated and will be ignored. Please remove it, as passing it will raise an error in the future. `scale` should directly be passed while calling the underlying pipeline component i.e., via `cross_attention_kwargs`."
+ deprecate("scale", "1.0.0", deprecation_message)
+
+ residual = hidden_states
+ if attn.spatial_norm is not None:
+ hidden_states = attn.spatial_norm(hidden_states, temb)
+
+ input_ndim = hidden_states.ndim
+
+ if attn.group_norm is not None:
+ hidden_states = attn.group_norm(hidden_states)
+
+ if input_ndim == 4:
+ batch_size, channel, height, width = hidden_states.shape
+ hidden_states = hidden_states.view(batch_size, channel, height * width).transpose(1, 2).contiguous()
+
+ if encoder_hidden_states is None:
+ qkv = attn.to_qkv(hidden_states)
+ split_size = qkv.shape[-1] // 3
+ query, key, value = torch.split(qkv, split_size, dim=-1)
+ else:
+ if attn.norm_cross:
+ encoder_hidden_states = attn.norm_encoder_hidden_states(encoder_hidden_states)
+ if self_attention_query is not None:
+ query = self_attention_query
+ else:
+ query = attn.to_q(hidden_states)
+
+ kv = attn.to_kv(encoder_hidden_states)
+ split_size = kv.shape[-1] // 2
+ key, value = torch.split(kv, split_size, dim=-1)
+
+ if attn.norm_q is not None:
+ query = attn.norm_q(query)
+ if attn.norm_k is not None:
+ key = attn.norm_k(key)
+
+ inner_dim = key.shape[-1]
+ head_dim = inner_dim // attn.heads
+
+ if self_attention_output is None:
+ query = query.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
+ key = key.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
+ value = value.view(batch_size, -1, attn.heads, head_dim).transpose(1, 2)
+
+ if attn.norm_q is not None:
+ query = attn.norm_q(query)
+ if attn.norm_k is not None:
+ key = attn.norm_k(key)
+
+ # the output of sdp = (batch, num_heads, seq_len, head_dim)
+ # TODO: add support for attn.scale when we move to Torch 2.1
+ hidden_states = F.scaled_dot_product_attention(
+ query, key, value, attn_mask=attention_mask, dropout_p=0.0, is_causal=False
+ )
+
+ hidden_states = hidden_states.to(query.dtype)
+
+ if self_attention_output is not None:
+ hidden_states = hidden_states + self_attention_output
+ hidden_states = hidden_states.transpose(1, 2).reshape(batch_size, -1, attn.heads * head_dim)
+
+ if attn.residual_connection:
+ hidden_states = hidden_states + residual
+
+ hidden_states = hidden_states / attn.rescale_output_factor
+
+ return hidden_states if self_attention_output is not None else (hidden_states, query)
+
+
+class MatryoshkaFeedForward(nn.Module):
+ r"""
+ A feed-forward layer for the Matryoshka models.
+
+ Parameters:"""
+
+ def __init__(
+ self,
+ dim: int,
+ ):
+ super().__init__()
+
+ self.group_norm = nn.GroupNorm(32, dim)
+ self.linear_gelu = GELU(dim, dim * 4)
+ self.linear_out = nn.Linear(dim * 4, dim)
+
+ def forward(self, x):
+ batch_size, channels, *spatial_dims = x.shape
+ x = self.group_norm(x)
+ x = x.view(batch_size, channels, -1).permute(0, 2, 1)
+ x = self.linear_out(self.linear_gelu(x))
+ x = x.permute(0, 2, 1).view(batch_size, channels, *spatial_dims)
+ return x
+
+
+def get_down_block(
+ down_block_type: str,
+ num_layers: int,
+ in_channels: int,
+ out_channels: int,
+ temb_channels: int,
+ add_downsample: bool,
+ resnet_eps: float,
+ resnet_act_fn: str,
+ norm_type: str = "layer_norm",
+ transformer_layers_per_block: int = 1,
+ num_attention_heads: Optional[int] = None,
+ resnet_groups: Optional[int] = None,
+ cross_attention_dim: Optional[int] = None,
+ downsample_padding: Optional[int] = None,
+ dual_cross_attention: bool = False,
+ use_linear_projection: bool = False,
+ only_cross_attention: bool = False,
+ upcast_attention: bool = False,
+ resnet_time_scale_shift: str = "default",
+ attention_type: str = "default",
+ attention_pre_only: bool = False,
+ resnet_skip_time_act: bool = False,
+ resnet_out_scale_factor: float = 1.0,
+ cross_attention_norm: Optional[str] = None,
+ attention_head_dim: Optional[int] = None,
+ use_attention_ffn: bool = True,
+ downsample_type: Optional[str] = None,
+ dropout: float = 0.0,
+):
+ # If attn head dim is not defined, we default it to the number of heads
+ if attention_head_dim is None:
+ logger.warning(
+ f"It is recommended to provide `attention_head_dim` when calling `get_down_block`. Defaulting `attention_head_dim` to {num_attention_heads}."
+ )
+ attention_head_dim = num_attention_heads
+
+ down_block_type = down_block_type[7:] if down_block_type.startswith("UNetRes") else down_block_type
+ if down_block_type == "DownBlock2D":
+ return DownBlock2D(
+ num_layers=num_layers,
+ in_channels=in_channels,
+ out_channels=out_channels,
+ temb_channels=temb_channels,
+ dropout=dropout,
+ add_downsample=add_downsample,
+ resnet_eps=resnet_eps,
+ resnet_act_fn=resnet_act_fn,
+ resnet_groups=resnet_groups,
+ downsample_padding=downsample_padding,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ )
+ elif down_block_type == "CrossAttnDownBlock2D":
+ if cross_attention_dim is None:
+ raise ValueError("cross_attention_dim must be specified for CrossAttnDownBlock2D")
+ return CrossAttnDownBlock2D(
+ num_layers=num_layers,
+ transformer_layers_per_block=transformer_layers_per_block,
+ in_channels=in_channels,
+ out_channels=out_channels,
+ temb_channels=temb_channels,
+ dropout=dropout,
+ add_downsample=add_downsample,
+ resnet_eps=resnet_eps,
+ resnet_act_fn=resnet_act_fn,
+ norm_type=norm_type,
+ resnet_groups=resnet_groups,
+ downsample_padding=downsample_padding,
+ cross_attention_dim=cross_attention_dim,
+ cross_attention_norm=cross_attention_norm,
+ num_attention_heads=num_attention_heads,
+ dual_cross_attention=dual_cross_attention,
+ use_linear_projection=use_linear_projection,
+ only_cross_attention=only_cross_attention,
+ upcast_attention=upcast_attention,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ attention_type=attention_type,
+ attention_pre_only=attention_pre_only,
+ use_attention_ffn=use_attention_ffn,
+ )
+
+
+def get_mid_block(
+ mid_block_type: str,
+ temb_channels: int,
+ in_channels: int,
+ resnet_eps: float,
+ resnet_act_fn: str,
+ resnet_groups: int,
+ norm_type: str = "layer_norm",
+ output_scale_factor: float = 1.0,
+ transformer_layers_per_block: int = 1,
+ num_attention_heads: Optional[int] = None,
+ cross_attention_dim: Optional[int] = None,
+ dual_cross_attention: bool = False,
+ use_linear_projection: bool = False,
+ mid_block_only_cross_attention: bool = False,
+ upcast_attention: bool = False,
+ resnet_time_scale_shift: str = "default",
+ attention_type: str = "default",
+ attention_pre_only: bool = False,
+ resnet_skip_time_act: bool = False,
+ cross_attention_norm: Optional[str] = None,
+ attention_head_dim: Optional[int] = 1,
+ dropout: float = 0.0,
+):
+ if mid_block_type == "UNetMidBlock2DCrossAttn":
+ return UNetMidBlock2DCrossAttn(
+ transformer_layers_per_block=transformer_layers_per_block,
+ in_channels=in_channels,
+ temb_channels=temb_channels,
+ dropout=dropout,
+ resnet_eps=resnet_eps,
+ resnet_act_fn=resnet_act_fn,
+ norm_type=norm_type,
+ output_scale_factor=output_scale_factor,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ cross_attention_dim=cross_attention_dim,
+ cross_attention_norm=cross_attention_norm,
+ num_attention_heads=num_attention_heads,
+ resnet_groups=resnet_groups,
+ dual_cross_attention=dual_cross_attention,
+ use_linear_projection=use_linear_projection,
+ upcast_attention=upcast_attention,
+ attention_type=attention_type,
+ attention_pre_only=attention_pre_only,
+ )
+
+
+def get_up_block(
+ up_block_type: str,
+ num_layers: int,
+ in_channels: int,
+ out_channels: int,
+ prev_output_channel: int,
+ temb_channels: int,
+ add_upsample: bool,
+ resnet_eps: float,
+ resnet_act_fn: str,
+ norm_type: str = "layer_norm",
+ resolution_idx: Optional[int] = None,
+ transformer_layers_per_block: int = 1,
+ num_attention_heads: Optional[int] = None,
+ resnet_groups: Optional[int] = None,
+ cross_attention_dim: Optional[int] = None,
+ dual_cross_attention: bool = False,
+ use_linear_projection: bool = False,
+ only_cross_attention: bool = False,
+ upcast_attention: bool = False,
+ resnet_time_scale_shift: str = "default",
+ attention_type: str = "default",
+ attention_pre_only: bool = False,
+ resnet_skip_time_act: bool = False,
+ resnet_out_scale_factor: float = 1.0,
+ cross_attention_norm: Optional[str] = None,
+ attention_head_dim: Optional[int] = None,
+ use_attention_ffn: bool = True,
+ upsample_type: Optional[str] = None,
+ dropout: float = 0.0,
+) -> nn.Module:
+ # If attn head dim is not defined, we default it to the number of heads
+ if attention_head_dim is None:
+ logger.warning(
+ f"It is recommended to provide `attention_head_dim` when calling `get_up_block`. Defaulting `attention_head_dim` to {num_attention_heads}."
+ )
+ attention_head_dim = num_attention_heads
+
+ up_block_type = up_block_type[7:] if up_block_type.startswith("UNetRes") else up_block_type
+ if up_block_type == "UpBlock2D":
+ return UpBlock2D(
+ num_layers=num_layers,
+ in_channels=in_channels,
+ out_channels=out_channels,
+ prev_output_channel=prev_output_channel,
+ temb_channels=temb_channels,
+ resolution_idx=resolution_idx,
+ dropout=dropout,
+ add_upsample=add_upsample,
+ resnet_eps=resnet_eps,
+ resnet_act_fn=resnet_act_fn,
+ resnet_groups=resnet_groups,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ )
+ elif up_block_type == "CrossAttnUpBlock2D":
+ if cross_attention_dim is None:
+ raise ValueError("cross_attention_dim must be specified for CrossAttnUpBlock2D")
+ return CrossAttnUpBlock2D(
+ num_layers=num_layers,
+ transformer_layers_per_block=transformer_layers_per_block,
+ in_channels=in_channels,
+ out_channels=out_channels,
+ prev_output_channel=prev_output_channel,
+ temb_channels=temb_channels,
+ resolution_idx=resolution_idx,
+ dropout=dropout,
+ add_upsample=add_upsample,
+ resnet_eps=resnet_eps,
+ resnet_act_fn=resnet_act_fn,
+ norm_type=norm_type,
+ resnet_groups=resnet_groups,
+ cross_attention_dim=cross_attention_dim,
+ cross_attention_norm=cross_attention_norm,
+ num_attention_heads=num_attention_heads,
+ dual_cross_attention=dual_cross_attention,
+ use_linear_projection=use_linear_projection,
+ only_cross_attention=only_cross_attention,
+ upcast_attention=upcast_attention,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ attention_type=attention_type,
+ attention_pre_only=attention_pre_only,
+ use_attention_ffn=use_attention_ffn,
+ )
+
+
+class MatryoshkaCombinedTimestepTextEmbedding(nn.Module):
+ def __init__(self, addition_time_embed_dim, cross_attention_dim, time_embed_dim, type):
+ super().__init__()
+ if type == "unet":
+ self.cond_emb = nn.Linear(cross_attention_dim, time_embed_dim, bias=False)
+ elif type == "nested_unet":
+ self.cond_emb = None
+ self.add_time_proj = Timesteps(addition_time_embed_dim, flip_sin_to_cos=False, downscale_freq_shift=0)
+ self.add_timestep_embedder = TimestepEmbedding(addition_time_embed_dim, time_embed_dim)
+
+ def forward(self, emb, encoder_hidden_states, added_cond_kwargs):
+ conditioning_mask = added_cond_kwargs.get("conditioning_mask", None)
+ masked_cross_attention = added_cond_kwargs.get("masked_cross_attention", False)
+ if self.cond_emb is not None and not added_cond_kwargs.get("from_nested", False):
+ if conditioning_mask is None:
+ y = encoder_hidden_states.mean(dim=1)
+ else:
+ y = (conditioning_mask.unsqueeze(-1) * encoder_hidden_states).sum(dim=1) / conditioning_mask.sum(
+ dim=1, keepdim=True
+ )
+ cond_emb = self.cond_emb(y)
+ else:
+ cond_emb = None
+
+ if not masked_cross_attention:
+ conditioning_mask = None
+
+ micro = added_cond_kwargs.get("micro_conditioning_scale", None)
+ if micro is not None:
+ temb = self.add_time_proj(torch.tensor([micro], device=emb.device, dtype=emb.dtype))
+ temb_micro_conditioning = self.add_timestep_embedder(temb.to(emb.dtype))
+ # if self.cond_emb is not None and not added_cond_kwargs.get("from_nested", False):
+ return temb_micro_conditioning, conditioning_mask, cond_emb
+
+ return None, conditioning_mask, cond_emb
+
+
+@dataclass
+class MatryoshkaUNet2DConditionOutput(BaseOutput):
+ """
+ The output of [`MatryoshkaUNet2DConditionOutput`].
+
+ Args:
+ sample (`torch.Tensor` of shape `(batch_size, num_channels, height, width)`):
+ The hidden states output conditioned on `encoder_hidden_states` input. Output of last layer of model.
+ """
+
+ sample: torch.Tensor = None
+ sample_inner: torch.Tensor = None
+
+
+class MatryoshkaUNet2DConditionModel(
+ ModelMixin, ConfigMixin, FromOriginalModelMixin, UNet2DConditionLoadersMixin, PeftAdapterMixin
+):
+ r"""
+ A conditional 2D UNet model that takes a noisy sample, conditional state, and a timestep and returns a sample
+ shaped output.
+
+ This model inherits from [`ModelMixin`]. Check the superclass documentation for it's generic methods implemented
+ for all models (such as downloading or saving).
+
+ Parameters:
+ sample_size (`int` or `Tuple[int, int]`, *optional*, defaults to `None`):
+ Height and width of input/output sample.
+ in_channels (`int`, *optional*, defaults to 4): Number of channels in the input sample.
+ out_channels (`int`, *optional*, defaults to 4): Number of channels in the output.
+ center_input_sample (`bool`, *optional*, defaults to `False`): Whether to center the input sample.
+ flip_sin_to_cos (`bool`, *optional*, defaults to `True`):
+ Whether to flip the sin to cos in the time embedding.
+ freq_shift (`int`, *optional*, defaults to 0): The frequency shift to apply to the time embedding.
+ down_block_types (`Tuple[str]`, *optional*, defaults to `("CrossAttnDownBlock2D", "CrossAttnDownBlock2D", "CrossAttnDownBlock2D", "DownBlock2D")`):
+ The tuple of downsample blocks to use.
+ mid_block_type (`str`, *optional*, defaults to `"UNetMidBlock2DCrossAttn"`):
+ Block type for middle of UNet, it can be one of `UNetMidBlock2DCrossAttn`, `UNetMidBlock2D`, or
+ `UNetMidBlock2DSimpleCrossAttn`. If `None`, the mid block layer is skipped.
+ up_block_types (`Tuple[str]`, *optional*, defaults to `("UpBlock2D", "CrossAttnUpBlock2D", "CrossAttnUpBlock2D", "CrossAttnUpBlock2D")`):
+ The tuple of upsample blocks to use.
+ only_cross_attention(`bool` or `Tuple[bool]`, *optional*, default to `False`):
+ Whether to include self-attention in the basic transformer blocks, see
+ [`~models.attention.BasicTransformerBlock`].
+ block_out_channels (`Tuple[int]`, *optional*, defaults to `(320, 640, 1280, 1280)`):
+ The tuple of output channels for each block.
+ layers_per_block (`int`, *optional*, defaults to 2): The number of layers per block.
+ downsample_padding (`int`, *optional*, defaults to 1): The padding to use for the downsampling convolution.
+ mid_block_scale_factor (`float`, *optional*, defaults to 1.0): The scale factor to use for the mid block.
+ dropout (`float`, *optional*, defaults to 0.0): The dropout probability to use.
+ act_fn (`str`, *optional*, defaults to `"silu"`): The activation function to use.
+ norm_num_groups (`int`, *optional*, defaults to 32): The number of groups to use for the normalization.
+ If `None`, normalization and activation layers is skipped in post-processing.
+ norm_eps (`float`, *optional*, defaults to 1e-5): The epsilon to use for the normalization.
+ cross_attention_dim (`int` or `Tuple[int]`, *optional*, defaults to 1280):
+ The dimension of the cross attention features.
+ transformer_layers_per_block (`int`, `Tuple[int]`, or `Tuple[Tuple]` , *optional*, defaults to 1):
+ The number of transformer blocks of type [`~models.attention.BasicTransformerBlock`]. Only relevant for
+ [`~models.unets.unet_2d_blocks.CrossAttnDownBlock2D`], [`~models.unets.unet_2d_blocks.CrossAttnUpBlock2D`],
+ [`~models.unets.unet_2d_blocks.UNetMidBlock2DCrossAttn`].
+ reverse_transformer_layers_per_block : (`Tuple[Tuple]`, *optional*, defaults to None):
+ The number of transformer blocks of type [`~models.attention.BasicTransformerBlock`], in the upsampling
+ blocks of the U-Net. Only relevant if `transformer_layers_per_block` is of type `Tuple[Tuple]` and for
+ [`~models.unets.unet_2d_blocks.CrossAttnDownBlock2D`], [`~models.unets.unet_2d_blocks.CrossAttnUpBlock2D`],
+ [`~models.unets.unet_2d_blocks.UNetMidBlock2DCrossAttn`].
+ encoder_hid_dim (`int`, *optional*, defaults to None):
+ If `encoder_hid_dim_type` is defined, `encoder_hidden_states` will be projected from `encoder_hid_dim`
+ dimension to `cross_attention_dim`.
+ encoder_hid_dim_type (`str`, *optional*, defaults to `None`):
+ If given, the `encoder_hidden_states` and potentially other embeddings are down-projected to text
+ embeddings of dimension `cross_attention` according to `encoder_hid_dim_type`.
+ attention_head_dim (`int`, *optional*, defaults to 8): The dimension of the attention heads.
+ num_attention_heads (`int`, *optional*):
+ The number of attention heads. If not defined, defaults to `attention_head_dim`
+ resnet_time_scale_shift (`str`, *optional*, defaults to `"default"`): Time scale shift config
+ for ResNet blocks (see [`~models.resnet.ResnetBlock2D`]). Choose from `default` or `scale_shift`.
+ class_embed_type (`str`, *optional*, defaults to `None`):
+ The type of class embedding to use which is ultimately summed with the time embeddings. Choose from `None`,
+ `"timestep"`, `"identity"`, `"projection"`, or `"simple_projection"`.
+ addition_embed_type (`str`, *optional*, defaults to `None`):
+ Configures an optional embedding which will be summed with the time embeddings. Choose from `None` or
+ "text". "text" will use the `TextTimeEmbedding` layer.
+ addition_time_embed_dim: (`int`, *optional*, defaults to `None`):
+ Dimension for the timestep embeddings.
+ num_class_embeds (`int`, *optional*, defaults to `None`):
+ Input dimension of the learnable embedding matrix to be projected to `time_embed_dim`, when performing
+ class conditioning with `class_embed_type` equal to `None`.
+ time_embedding_type (`str`, *optional*, defaults to `positional`):
+ The type of position embedding to use for timesteps. Choose from `positional` or `fourier`.
+ time_embedding_dim (`int`, *optional*, defaults to `None`):
+ An optional override for the dimension of the projected time embedding.
+ time_embedding_act_fn (`str`, *optional*, defaults to `None`):
+ Optional activation function to use only once on the time embeddings before they are passed to the rest of
+ the UNet. Choose from `silu`, `mish`, `gelu`, and `swish`.
+ timestep_post_act (`str`, *optional*, defaults to `None`):
+ The second activation function to use in timestep embedding. Choose from `silu`, `mish` and `gelu`.
+ time_cond_proj_dim (`int`, *optional*, defaults to `None`):
+ The dimension of `cond_proj` layer in the timestep embedding.
+ conv_in_kernel (`int`, *optional*, default to `3`): The kernel size of `conv_in` layer.
+ conv_out_kernel (`int`, *optional*, default to `3`): The kernel size of `conv_out` layer.
+ projection_class_embeddings_input_dim (`int`, *optional*): The dimension of the `class_labels` input when
+ `class_embed_type="projection"`. Required when `class_embed_type="projection"`.
+ class_embeddings_concat (`bool`, *optional*, defaults to `False`): Whether to concatenate the time
+ embeddings with the class embeddings.
+ mid_block_only_cross_attention (`bool`, *optional*, defaults to `None`):
+ Whether to use cross attention with the mid block when using the `UNetMidBlock2DSimpleCrossAttn`. If
+ `only_cross_attention` is given as a single boolean and `mid_block_only_cross_attention` is `None`, the
+ `only_cross_attention` value is used as the value for `mid_block_only_cross_attention`. Default to `False`
+ otherwise.
+ """
+
+ _supports_gradient_checkpointing = True
+ _no_split_modules = ["MatryoshkaTransformerBlock", "ResnetBlock2D", "CrossAttnUpBlock2D"]
+
+ @register_to_config
+ def __init__(
+ self,
+ sample_size: Optional[int] = None,
+ in_channels: int = 3,
+ out_channels: int = 3,
+ center_input_sample: bool = False,
+ flip_sin_to_cos: bool = True,
+ freq_shift: int = 0,
+ down_block_types: Tuple[str] = (
+ "CrossAttnDownBlock2D",
+ "CrossAttnDownBlock2D",
+ "CrossAttnDownBlock2D",
+ "DownBlock2D",
+ ),
+ mid_block_type: Optional[str] = "UNetMidBlock2DCrossAttn",
+ up_block_types: Tuple[str] = ("UpBlock2D", "CrossAttnUpBlock2D", "CrossAttnUpBlock2D", "CrossAttnUpBlock2D"),
+ only_cross_attention: Union[bool, Tuple[bool]] = False,
+ block_out_channels: Tuple[int] = (320, 640, 1280, 1280),
+ layers_per_block: Union[int, Tuple[int]] = 2,
+ downsample_padding: int = 1,
+ mid_block_scale_factor: float = 1,
+ dropout: float = 0.0,
+ act_fn: str = "silu",
+ norm_type: str = "layer_norm",
+ norm_num_groups: Optional[int] = 32,
+ norm_eps: float = 1e-5,
+ cross_attention_dim: Union[int, Tuple[int]] = 1280,
+ transformer_layers_per_block: Union[int, Tuple[int], Tuple[Tuple]] = 1,
+ reverse_transformer_layers_per_block: Optional[Tuple[Tuple[int]]] = None,
+ encoder_hid_dim: Optional[int] = None,
+ encoder_hid_dim_type: Optional[str] = None,
+ attention_head_dim: Union[int, Tuple[int]] = 8,
+ num_attention_heads: Optional[Union[int, Tuple[int]]] = None,
+ dual_cross_attention: bool = False,
+ use_attention_ffn: bool = True,
+ use_linear_projection: bool = False,
+ class_embed_type: Optional[str] = None,
+ addition_embed_type: Optional[str] = None,
+ addition_time_embed_dim: Optional[int] = None,
+ num_class_embeds: Optional[int] = None,
+ upcast_attention: bool = False,
+ resnet_time_scale_shift: str = "default",
+ resnet_skip_time_act: bool = False,
+ resnet_out_scale_factor: float = 1.0,
+ time_embedding_type: str = "positional",
+ time_embedding_dim: Optional[int] = None,
+ time_embedding_act_fn: Optional[str] = None,
+ timestep_post_act: Optional[str] = None,
+ time_cond_proj_dim: Optional[int] = None,
+ conv_in_kernel: int = 3,
+ conv_out_kernel: int = 3,
+ projection_class_embeddings_input_dim: Optional[int] = None,
+ attention_type: str = "default",
+ attention_pre_only: bool = False,
+ masked_cross_attention: bool = False,
+ micro_conditioning_scale: int = None,
+ class_embeddings_concat: bool = False,
+ mid_block_only_cross_attention: Optional[bool] = None,
+ cross_attention_norm: Optional[str] = None,
+ addition_embed_type_num_heads: int = 64,
+ temporal_mode: bool = False,
+ temporal_spatial_ds: bool = False,
+ skip_cond_emb: bool = False,
+ nesting: Optional[int] = False,
+ ):
+ super().__init__()
+
+ self.sample_size = sample_size
+
+ if num_attention_heads is not None:
+ raise ValueError(
+ "At the moment it is not possible to define the number of attention heads via `num_attention_heads` because of a naming issue as described in https://github.com/huggingface/diffusers/issues/2011#issuecomment-1547958131. Passing `num_attention_heads` will only be supported in diffusers v0.19."
+ )
+
+ # If `num_attention_heads` is not defined (which is the case for most models)
+ # it will default to `attention_head_dim`. This looks weird upon first reading it and it is.
+ # The reason for this behavior is to correct for incorrectly named variables that were introduced
+ # when this library was created. The incorrect naming was only discovered much later in https://github.com/huggingface/diffusers/issues/2011#issuecomment-1547958131
+ # Changing `attention_head_dim` to `num_attention_heads` for 40,000+ configurations is too backwards breaking
+ # which is why we correct for the naming here.
+ num_attention_heads = num_attention_heads or attention_head_dim
+
+ # Check inputs
+ self._check_config(
+ down_block_types=down_block_types,
+ up_block_types=up_block_types,
+ only_cross_attention=only_cross_attention,
+ block_out_channels=block_out_channels,
+ layers_per_block=layers_per_block,
+ cross_attention_dim=cross_attention_dim,
+ transformer_layers_per_block=transformer_layers_per_block,
+ reverse_transformer_layers_per_block=reverse_transformer_layers_per_block,
+ attention_head_dim=attention_head_dim,
+ num_attention_heads=num_attention_heads,
+ )
+
+ # input
+ conv_in_padding = (conv_in_kernel - 1) // 2
+ self.conv_in = nn.Conv2d(
+ in_channels, block_out_channels[0], kernel_size=conv_in_kernel, padding=conv_in_padding
+ )
+
+ # time
+ time_embed_dim, timestep_input_dim = self._set_time_proj(
+ time_embedding_type,
+ block_out_channels=block_out_channels,
+ flip_sin_to_cos=flip_sin_to_cos,
+ freq_shift=freq_shift,
+ time_embedding_dim=time_embedding_dim,
+ )
+
+ self.time_embedding = TimestepEmbedding(
+ time_embedding_dim // 4 if time_embedding_dim is not None else timestep_input_dim,
+ time_embed_dim,
+ act_fn=act_fn,
+ post_act_fn=timestep_post_act,
+ cond_proj_dim=time_cond_proj_dim,
+ )
+
+ self._set_encoder_hid_proj(
+ encoder_hid_dim_type,
+ cross_attention_dim=cross_attention_dim,
+ encoder_hid_dim=encoder_hid_dim,
+ )
+
+ # class embedding
+ self._set_class_embedding(
+ class_embed_type,
+ act_fn=act_fn,
+ num_class_embeds=num_class_embeds,
+ projection_class_embeddings_input_dim=projection_class_embeddings_input_dim,
+ time_embed_dim=time_embed_dim,
+ timestep_input_dim=timestep_input_dim,
+ )
+
+ self._set_add_embedding(
+ addition_embed_type,
+ addition_embed_type_num_heads=addition_embed_type_num_heads,
+ addition_time_embed_dim=timestep_input_dim,
+ cross_attention_dim=cross_attention_dim,
+ encoder_hid_dim=encoder_hid_dim,
+ flip_sin_to_cos=flip_sin_to_cos,
+ freq_shift=freq_shift,
+ projection_class_embeddings_input_dim=projection_class_embeddings_input_dim,
+ time_embed_dim=time_embed_dim,
+ )
+
+ if time_embedding_act_fn is None:
+ self.time_embed_act = None
+ else:
+ self.time_embed_act = get_activation(time_embedding_act_fn)
+
+ self.down_blocks = nn.ModuleList([])
+ self.up_blocks = nn.ModuleList([])
+
+ if isinstance(only_cross_attention, bool):
+ if mid_block_only_cross_attention is None:
+ mid_block_only_cross_attention = only_cross_attention
+
+ only_cross_attention = [only_cross_attention] * len(down_block_types)
+
+ if mid_block_only_cross_attention is None:
+ mid_block_only_cross_attention = False
+
+ if isinstance(num_attention_heads, int):
+ num_attention_heads = (num_attention_heads,) * len(down_block_types)
+
+ if isinstance(attention_head_dim, int):
+ attention_head_dim = (attention_head_dim,) * len(down_block_types)
+
+ if isinstance(cross_attention_dim, int):
+ cross_attention_dim = (cross_attention_dim,) * len(down_block_types)
+
+ if isinstance(layers_per_block, int):
+ layers_per_block = [layers_per_block] * len(down_block_types)
+
+ if isinstance(transformer_layers_per_block, int):
+ transformer_layers_per_block = [transformer_layers_per_block] * len(down_block_types)
+
+ if class_embeddings_concat:
+ # The time embeddings are concatenated with the class embeddings. The dimension of the
+ # time embeddings passed to the down, middle, and up blocks is twice the dimension of the
+ # regular time embeddings
+ blocks_time_embed_dim = time_embed_dim * 2
+ else:
+ blocks_time_embed_dim = time_embed_dim
+
+ # down
+ output_channel = block_out_channels[0]
+ for i, down_block_type in enumerate(down_block_types):
+ input_channel = output_channel
+ output_channel = block_out_channels[i]
+ is_final_block = i == len(block_out_channels) - 1
+
+ down_block = get_down_block(
+ down_block_type,
+ num_layers=layers_per_block[i],
+ transformer_layers_per_block=transformer_layers_per_block[i],
+ in_channels=input_channel,
+ out_channels=output_channel,
+ temb_channels=blocks_time_embed_dim,
+ add_downsample=not is_final_block,
+ resnet_eps=norm_eps,
+ resnet_act_fn=act_fn,
+ norm_type=norm_type,
+ resnet_groups=norm_num_groups,
+ cross_attention_dim=cross_attention_dim[i],
+ num_attention_heads=num_attention_heads[i],
+ downsample_padding=downsample_padding,
+ dual_cross_attention=dual_cross_attention,
+ use_linear_projection=use_linear_projection,
+ only_cross_attention=only_cross_attention[i],
+ upcast_attention=upcast_attention,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ attention_type=attention_type,
+ attention_pre_only=attention_pre_only,
+ resnet_skip_time_act=resnet_skip_time_act,
+ resnet_out_scale_factor=resnet_out_scale_factor,
+ cross_attention_norm=cross_attention_norm,
+ use_attention_ffn=use_attention_ffn,
+ attention_head_dim=attention_head_dim[i] if attention_head_dim[i] is not None else output_channel,
+ dropout=dropout,
+ )
+ self.down_blocks.append(down_block)
+
+ # mid
+ self.mid_block = get_mid_block(
+ mid_block_type,
+ temb_channels=blocks_time_embed_dim,
+ in_channels=block_out_channels[-1],
+ resnet_eps=norm_eps,
+ resnet_act_fn=act_fn,
+ norm_type=norm_type,
+ resnet_groups=norm_num_groups,
+ output_scale_factor=mid_block_scale_factor,
+ transformer_layers_per_block=1,
+ num_attention_heads=num_attention_heads[-1],
+ cross_attention_dim=cross_attention_dim[-1],
+ dual_cross_attention=dual_cross_attention,
+ use_linear_projection=use_linear_projection,
+ mid_block_only_cross_attention=mid_block_only_cross_attention,
+ upcast_attention=upcast_attention,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ attention_type=attention_type,
+ attention_pre_only=attention_pre_only,
+ resnet_skip_time_act=resnet_skip_time_act,
+ cross_attention_norm=cross_attention_norm,
+ attention_head_dim=attention_head_dim[-1],
+ dropout=dropout,
+ )
+
+ # count how many layers upsample the images
+ self.num_upsamplers = 0
+
+ # up
+ reversed_block_out_channels = list(reversed(block_out_channels))
+ reversed_num_attention_heads = list(reversed(num_attention_heads))
+ reversed_layers_per_block = list(reversed(layers_per_block))
+ reversed_cross_attention_dim = list(reversed(cross_attention_dim))
+ reversed_transformer_layers_per_block = (
+ list(reversed(transformer_layers_per_block))
+ if reverse_transformer_layers_per_block is None
+ else reverse_transformer_layers_per_block
+ )
+ only_cross_attention = list(reversed(only_cross_attention))
+
+ output_channel = reversed_block_out_channels[0]
+ for i, up_block_type in enumerate(up_block_types):
+ is_final_block = i == len(block_out_channels) - 1
+
+ prev_output_channel = output_channel
+ output_channel = reversed_block_out_channels[i]
+ input_channel = reversed_block_out_channels[min(i + 1, len(block_out_channels) - 1)]
+
+ # add upsample block for all BUT final layer
+ if not is_final_block:
+ add_upsample = True
+ self.num_upsamplers += 1
+ else:
+ add_upsample = False
+
+ up_block = get_up_block(
+ up_block_type,
+ num_layers=reversed_layers_per_block[i] + 1,
+ transformer_layers_per_block=reversed_transformer_layers_per_block[i],
+ in_channels=input_channel,
+ out_channels=output_channel,
+ prev_output_channel=prev_output_channel,
+ temb_channels=blocks_time_embed_dim,
+ add_upsample=add_upsample,
+ resnet_eps=norm_eps,
+ resnet_act_fn=act_fn,
+ norm_type=norm_type,
+ resolution_idx=i,
+ resnet_groups=norm_num_groups,
+ cross_attention_dim=reversed_cross_attention_dim[i],
+ num_attention_heads=reversed_num_attention_heads[i],
+ dual_cross_attention=dual_cross_attention,
+ use_linear_projection=use_linear_projection,
+ only_cross_attention=only_cross_attention[i],
+ upcast_attention=upcast_attention,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ attention_type=attention_type,
+ attention_pre_only=attention_pre_only,
+ resnet_skip_time_act=resnet_skip_time_act,
+ resnet_out_scale_factor=resnet_out_scale_factor,
+ cross_attention_norm=cross_attention_norm,
+ use_attention_ffn=use_attention_ffn,
+ attention_head_dim=attention_head_dim[i] if attention_head_dim[i] is not None else output_channel,
+ dropout=dropout,
+ )
+ self.up_blocks.append(up_block)
+
+ # out
+ if norm_num_groups is not None:
+ self.conv_norm_out = nn.GroupNorm(
+ num_channels=block_out_channels[0], num_groups=norm_num_groups, eps=norm_eps
+ )
+
+ self.conv_act = get_activation(act_fn)
+
+ else:
+ self.conv_norm_out = None
+ self.conv_act = None
+
+ conv_out_padding = (conv_out_kernel - 1) // 2
+ self.conv_out = nn.Conv2d(
+ block_out_channels[0], out_channels, kernel_size=conv_out_kernel, padding=conv_out_padding
+ )
+
+ self._set_pos_net_if_use_gligen(attention_type=attention_type, cross_attention_dim=cross_attention_dim)
+
+ self.is_temporal = []
+
+ def _check_config(
+ self,
+ down_block_types: Tuple[str],
+ up_block_types: Tuple[str],
+ only_cross_attention: Union[bool, Tuple[bool]],
+ block_out_channels: Tuple[int],
+ layers_per_block: Union[int, Tuple[int]],
+ cross_attention_dim: Union[int, Tuple[int]],
+ transformer_layers_per_block: Union[int, Tuple[int], Tuple[Tuple[int]]],
+ reverse_transformer_layers_per_block: bool,
+ attention_head_dim: int,
+ num_attention_heads: Optional[Union[int, Tuple[int]]],
+ ):
+ if len(down_block_types) != len(up_block_types):
+ raise ValueError(
+ f"Must provide the same number of `down_block_types` as `up_block_types`. `down_block_types`: {down_block_types}. `up_block_types`: {up_block_types}."
+ )
+
+ if len(block_out_channels) != len(down_block_types):
+ raise ValueError(
+ f"Must provide the same number of `block_out_channels` as `down_block_types`. `block_out_channels`: {block_out_channels}. `down_block_types`: {down_block_types}."
+ )
+
+ if not isinstance(only_cross_attention, bool) and len(only_cross_attention) != len(down_block_types):
+ raise ValueError(
+ f"Must provide the same number of `only_cross_attention` as `down_block_types`. `only_cross_attention`: {only_cross_attention}. `down_block_types`: {down_block_types}."
+ )
+
+ if not isinstance(num_attention_heads, int) and len(num_attention_heads) != len(down_block_types):
+ raise ValueError(
+ f"Must provide the same number of `num_attention_heads` as `down_block_types`. `num_attention_heads`: {num_attention_heads}. `down_block_types`: {down_block_types}."
+ )
+
+ if not isinstance(attention_head_dim, int) and len(attention_head_dim) != len(down_block_types):
+ raise ValueError(
+ f"Must provide the same number of `attention_head_dim` as `down_block_types`. `attention_head_dim`: {attention_head_dim}. `down_block_types`: {down_block_types}."
+ )
+
+ if isinstance(cross_attention_dim, list) and len(cross_attention_dim) != len(down_block_types):
+ raise ValueError(
+ f"Must provide the same number of `cross_attention_dim` as `down_block_types`. `cross_attention_dim`: {cross_attention_dim}. `down_block_types`: {down_block_types}."
+ )
+
+ if not isinstance(layers_per_block, int) and len(layers_per_block) != len(down_block_types):
+ raise ValueError(
+ f"Must provide the same number of `layers_per_block` as `down_block_types`. `layers_per_block`: {layers_per_block}. `down_block_types`: {down_block_types}."
+ )
+ if isinstance(transformer_layers_per_block, list) and reverse_transformer_layers_per_block is None:
+ for layer_number_per_block in transformer_layers_per_block:
+ if isinstance(layer_number_per_block, list):
+ raise ValueError("Must provide 'reverse_transformer_layers_per_block` if using asymmetrical UNet.")
+
+ def _set_time_proj(
+ self,
+ time_embedding_type: str,
+ block_out_channels: int,
+ flip_sin_to_cos: bool,
+ freq_shift: float,
+ time_embedding_dim: int,
+ ) -> Tuple[int, int]:
+ if time_embedding_type == "fourier":
+ time_embed_dim = time_embedding_dim or block_out_channels[0] * 2
+ if time_embed_dim % 2 != 0:
+ raise ValueError(f"`time_embed_dim` should be divisible by 2, but is {time_embed_dim}.")
+ self.time_proj = GaussianFourierProjection(
+ time_embed_dim // 2, set_W_to_weight=False, log=False, flip_sin_to_cos=flip_sin_to_cos
+ )
+ timestep_input_dim = time_embed_dim
+ elif time_embedding_type == "positional":
+ time_embed_dim = time_embedding_dim or block_out_channels[0] * 4
+
+ if self.model_type == "unet":
+ self.time_proj = Timesteps(block_out_channels[0], flip_sin_to_cos, freq_shift)
+ elif self.model_type == "nested_unet" and self.config.micro_conditioning_scale == 256:
+ self.time_proj = Timesteps(block_out_channels[0] * 4, flip_sin_to_cos, freq_shift)
+ elif self.model_type == "nested_unet" and self.config.micro_conditioning_scale == 1024:
+ self.time_proj = Timesteps(block_out_channels[0] * 4 * 2, flip_sin_to_cos, freq_shift)
+ timestep_input_dim = block_out_channels[0]
+ else:
+ raise ValueError(
+ f"{time_embedding_type} does not exist. Please make sure to use one of `fourier` or `positional`."
+ )
+
+ return time_embed_dim, timestep_input_dim
+
+ def _set_encoder_hid_proj(
+ self,
+ encoder_hid_dim_type: Optional[str],
+ cross_attention_dim: Union[int, Tuple[int]],
+ encoder_hid_dim: Optional[int],
+ ):
+ if encoder_hid_dim_type is None and encoder_hid_dim is not None:
+ encoder_hid_dim_type = "text_proj"
+ self.register_to_config(encoder_hid_dim_type=encoder_hid_dim_type)
+ logger.info("encoder_hid_dim_type defaults to 'text_proj' as `encoder_hid_dim` is defined.")
+
+ if encoder_hid_dim is None and encoder_hid_dim_type is not None:
+ raise ValueError(
+ f"`encoder_hid_dim` has to be defined when `encoder_hid_dim_type` is set to {encoder_hid_dim_type}."
+ )
+
+ if encoder_hid_dim_type == "text_proj":
+ self.encoder_hid_proj = nn.Linear(encoder_hid_dim, cross_attention_dim)
+ elif encoder_hid_dim_type == "text_image_proj":
+ # image_embed_dim DOESN'T have to be `cross_attention_dim`. To not clutter the __init__ too much
+ # they are set to `cross_attention_dim` here as this is exactly the required dimension for the currently only use
+ # case when `addition_embed_type == "text_image_proj"` (Kandinsky 2.1)`
+ self.encoder_hid_proj = TextImageProjection(
+ text_embed_dim=encoder_hid_dim,
+ image_embed_dim=cross_attention_dim,
+ cross_attention_dim=cross_attention_dim,
+ )
+ elif encoder_hid_dim_type == "image_proj":
+ # Kandinsky 2.2
+ self.encoder_hid_proj = ImageProjection(
+ image_embed_dim=encoder_hid_dim,
+ cross_attention_dim=cross_attention_dim,
+ )
+ elif encoder_hid_dim_type is not None:
+ raise ValueError(
+ f"`encoder_hid_dim_type`: {encoder_hid_dim_type} must be None, 'text_proj', 'text_image_proj', or 'image_proj'."
+ )
+ else:
+ self.encoder_hid_proj = None
+
+ def _set_class_embedding(
+ self,
+ class_embed_type: Optional[str],
+ act_fn: str,
+ num_class_embeds: Optional[int],
+ projection_class_embeddings_input_dim: Optional[int],
+ time_embed_dim: int,
+ timestep_input_dim: int,
+ ):
+ if class_embed_type is None and num_class_embeds is not None:
+ self.class_embedding = nn.Embedding(num_class_embeds, time_embed_dim)
+ elif class_embed_type == "timestep":
+ self.class_embedding = TimestepEmbedding(timestep_input_dim, time_embed_dim, act_fn=act_fn)
+ elif class_embed_type == "identity":
+ self.class_embedding = nn.Identity(time_embed_dim, time_embed_dim)
+ elif class_embed_type == "projection":
+ if projection_class_embeddings_input_dim is None:
+ raise ValueError(
+ "`class_embed_type`: 'projection' requires `projection_class_embeddings_input_dim` be set"
+ )
+ # The projection `class_embed_type` is the same as the timestep `class_embed_type` except
+ # 1. the `class_labels` inputs are not first converted to sinusoidal embeddings
+ # 2. it projects from an arbitrary input dimension.
+ #
+ # Note that `TimestepEmbedding` is quite general, being mainly linear layers and activations.
+ # When used for embedding actual timesteps, the timesteps are first converted to sinusoidal embeddings.
+ # As a result, `TimestepEmbedding` can be passed arbitrary vectors.
+ self.class_embedding = TimestepEmbedding(projection_class_embeddings_input_dim, time_embed_dim)
+ elif class_embed_type == "simple_projection":
+ if projection_class_embeddings_input_dim is None:
+ raise ValueError(
+ "`class_embed_type`: 'simple_projection' requires `projection_class_embeddings_input_dim` be set"
+ )
+ self.class_embedding = nn.Linear(projection_class_embeddings_input_dim, time_embed_dim)
+ else:
+ self.class_embedding = None
+
+ def _set_add_embedding(
+ self,
+ addition_embed_type: str,
+ addition_embed_type_num_heads: int,
+ addition_time_embed_dim: Optional[int],
+ flip_sin_to_cos: bool,
+ freq_shift: float,
+ cross_attention_dim: Optional[int],
+ encoder_hid_dim: Optional[int],
+ projection_class_embeddings_input_dim: Optional[int],
+ time_embed_dim: int,
+ ):
+ if addition_embed_type == "text":
+ if encoder_hid_dim is not None:
+ text_time_embedding_from_dim = encoder_hid_dim
+ else:
+ text_time_embedding_from_dim = cross_attention_dim
+
+ self.add_embedding = TextTimeEmbedding(
+ text_time_embedding_from_dim, time_embed_dim, num_heads=addition_embed_type_num_heads
+ )
+ elif addition_embed_type == "matryoshka":
+ self.add_embedding = MatryoshkaCombinedTimestepTextEmbedding(
+ self.config.time_embedding_dim // 4
+ if self.config.time_embedding_dim is not None
+ else addition_time_embed_dim,
+ cross_attention_dim,
+ time_embed_dim,
+ self.model_type, # if not self.config.nesting else "inner_" + self.model_type,
+ )
+ elif addition_embed_type == "text_image":
+ # text_embed_dim and image_embed_dim DON'T have to be `cross_attention_dim`. To not clutter the __init__ too much
+ # they are set to `cross_attention_dim` here as this is exactly the required dimension for the currently only use
+ # case when `addition_embed_type == "text_image"` (Kandinsky 2.1)`
+ self.add_embedding = TextImageTimeEmbedding(
+ text_embed_dim=cross_attention_dim, image_embed_dim=cross_attention_dim, time_embed_dim=time_embed_dim
+ )
+ elif addition_embed_type == "text_time":
+ self.add_time_proj = Timesteps(addition_time_embed_dim, flip_sin_to_cos, freq_shift)
+ self.add_embedding = TimestepEmbedding(projection_class_embeddings_input_dim, time_embed_dim)
+ elif addition_embed_type == "image":
+ # Kandinsky 2.2
+ self.add_embedding = ImageTimeEmbedding(image_embed_dim=encoder_hid_dim, time_embed_dim=time_embed_dim)
+ elif addition_embed_type == "image_hint":
+ # Kandinsky 2.2 ControlNet
+ self.add_embedding = ImageHintTimeEmbedding(image_embed_dim=encoder_hid_dim, time_embed_dim=time_embed_dim)
+ elif addition_embed_type is not None:
+ raise ValueError(
+ f"`addition_embed_type`: {addition_embed_type} must be None, 'text', 'text_image', 'text_time', 'image', or 'image_hint'."
+ )
+
+ def _set_pos_net_if_use_gligen(self, attention_type: str, cross_attention_dim: int):
+ if attention_type in ["gated", "gated-text-image"]:
+ positive_len = 768
+ if isinstance(cross_attention_dim, int):
+ positive_len = cross_attention_dim
+ elif isinstance(cross_attention_dim, (list, tuple)):
+ positive_len = cross_attention_dim[0]
+
+ feature_type = "text-only" if attention_type == "gated" else "text-image"
+ self.position_net = GLIGENTextBoundingboxProjection(
+ positive_len=positive_len, out_dim=cross_attention_dim, feature_type=feature_type
+ )
+
+ @property
+ def attn_processors(self) -> Dict[str, AttentionProcessor]:
+ r"""
+ Returns:
+ `dict` of attention processors: A dictionary containing all attention processors used in the model with
+ indexed by its weight name.
+ """
+ # set recursively
+ processors = {}
+
+ def fn_recursive_add_processors(name: str, module: torch.nn.Module, processors: Dict[str, AttentionProcessor]):
+ if hasattr(module, "get_processor"):
+ processors[f"{name}.processor"] = module.get_processor()
+
+ for sub_name, child in module.named_children():
+ fn_recursive_add_processors(f"{name}.{sub_name}", child, processors)
+
+ return processors
+
+ for name, module in self.named_children():
+ fn_recursive_add_processors(name, module, processors)
+
+ return processors
+
+ def set_attn_processor(self, processor: Union[AttentionProcessor, Dict[str, AttentionProcessor]]):
+ r"""
+ Sets the attention processor to use to compute attention.
+
+ Parameters:
+ processor (`dict` of `AttentionProcessor` or only `AttentionProcessor`):
+ The instantiated processor class or a dictionary of processor classes that will be set as the processor
+ for **all** `Attention` layers.
+
+ If `processor` is a dict, the key needs to define the path to the corresponding cross attention
+ processor. This is strongly recommended when setting trainable attention processors.
+
+ """
+ count = len(self.attn_processors.keys())
+
+ if isinstance(processor, dict) and len(processor) != count:
+ raise ValueError(
+ f"A dict of processors was passed, but the number of processors {len(processor)} does not match the"
+ f" number of attention layers: {count}. Please make sure to pass {count} processor classes."
+ )
+
+ def fn_recursive_attn_processor(name: str, module: torch.nn.Module, processor):
+ if hasattr(module, "set_processor"):
+ if not isinstance(processor, dict):
+ module.set_processor(processor)
+ else:
+ module.set_processor(processor.pop(f"{name}.processor"))
+
+ for sub_name, child in module.named_children():
+ fn_recursive_attn_processor(f"{name}.{sub_name}", child, processor)
+
+ for name, module in self.named_children():
+ fn_recursive_attn_processor(name, module, processor)
+
+ def set_default_attn_processor(self):
+ """
+ Disables custom attention processors and sets the default attention implementation.
+ """
+ if all(proc.__class__ in ADDED_KV_ATTENTION_PROCESSORS for proc in self.attn_processors.values()):
+ processor = AttnAddedKVProcessor()
+ elif all(proc.__class__ in CROSS_ATTENTION_PROCESSORS for proc in self.attn_processors.values()):
+ processor = AttnProcessor()
+ else:
+ raise ValueError(
+ f"Cannot call `set_default_attn_processor` when attention processors are of type {next(iter(self.attn_processors.values()))}"
+ )
+
+ self.set_attn_processor(processor)
+
+ def set_attention_slice(self, slice_size: Union[str, int, List[int]] = "auto"):
+ r"""
+ Enable sliced attention computation.
+
+ When this option is enabled, the attention module splits the input tensor in slices to compute attention in
+ several steps. This is useful for saving some memory in exchange for a small decrease in speed.
+
+ Args:
+ slice_size (`str` or `int` or `list(int)`, *optional*, defaults to `"auto"`):
+ When `"auto"`, input to the attention heads is halved, so attention is computed in two steps. If
+ `"max"`, maximum amount of memory is saved by running only one slice at a time. If a number is
+ provided, uses as many slices as `attention_head_dim // slice_size`. In this case, `attention_head_dim`
+ must be a multiple of `slice_size`.
+ """
+ sliceable_head_dims = []
+
+ def fn_recursive_retrieve_sliceable_dims(module: torch.nn.Module):
+ if hasattr(module, "set_attention_slice"):
+ sliceable_head_dims.append(module.sliceable_head_dim)
+
+ for child in module.children():
+ fn_recursive_retrieve_sliceable_dims(child)
+
+ # retrieve number of attention layers
+ for module in self.children():
+ fn_recursive_retrieve_sliceable_dims(module)
+
+ num_sliceable_layers = len(sliceable_head_dims)
+
+ if slice_size == "auto":
+ # half the attention head size is usually a good trade-off between
+ # speed and memory
+ slice_size = [dim // 2 for dim in sliceable_head_dims]
+ elif slice_size == "max":
+ # make smallest slice possible
+ slice_size = num_sliceable_layers * [1]
+
+ slice_size = num_sliceable_layers * [slice_size] if not isinstance(slice_size, list) else slice_size
+
+ if len(slice_size) != len(sliceable_head_dims):
+ raise ValueError(
+ f"You have provided {len(slice_size)}, but {self.config} has {len(sliceable_head_dims)} different"
+ f" attention layers. Make sure to match `len(slice_size)` to be {len(sliceable_head_dims)}."
+ )
+
+ for i in range(len(slice_size)):
+ size = slice_size[i]
+ dim = sliceable_head_dims[i]
+ if size is not None and size > dim:
+ raise ValueError(f"size {size} has to be smaller or equal to {dim}.")
+
+ # Recursively walk through all the children.
+ # Any children which exposes the set_attention_slice method
+ # gets the message
+ def fn_recursive_set_attention_slice(module: torch.nn.Module, slice_size: List[int]):
+ if hasattr(module, "set_attention_slice"):
+ module.set_attention_slice(slice_size.pop())
+
+ for child in module.children():
+ fn_recursive_set_attention_slice(child, slice_size)
+
+ reversed_slice_size = list(reversed(slice_size))
+ for module in self.children():
+ fn_recursive_set_attention_slice(module, reversed_slice_size)
+
+ def _set_gradient_checkpointing(self, module, value=False):
+ if hasattr(module, "gradient_checkpointing"):
+ module.gradient_checkpointing = value
+
+ def enable_freeu(self, s1: float, s2: float, b1: float, b2: float):
+ r"""Enables the FreeU mechanism from https://arxiv.org/abs/2309.11497.
+
+ The suffixes after the scaling factors represent the stage blocks where they are being applied.
+
+ Please refer to the [official repository](https://github.com/ChenyangSi/FreeU) for combinations of values that
+ are known to work well for different pipelines such as Stable Diffusion v1, v2, and Stable Diffusion XL.
+
+ Args:
+ s1 (`float`):
+ Scaling factor for stage 1 to attenuate the contributions of the skip features. This is done to
+ mitigate the "oversmoothing effect" in the enhanced denoising process.
+ s2 (`float`):
+ Scaling factor for stage 2 to attenuate the contributions of the skip features. This is done to
+ mitigate the "oversmoothing effect" in the enhanced denoising process.
+ b1 (`float`): Scaling factor for stage 1 to amplify the contributions of backbone features.
+ b2 (`float`): Scaling factor for stage 2 to amplify the contributions of backbone features.
+ """
+ for i, upsample_block in enumerate(self.up_blocks):
+ setattr(upsample_block, "s1", s1)
+ setattr(upsample_block, "s2", s2)
+ setattr(upsample_block, "b1", b1)
+ setattr(upsample_block, "b2", b2)
+
+ def disable_freeu(self):
+ """Disables the FreeU mechanism."""
+ freeu_keys = {"s1", "s2", "b1", "b2"}
+ for i, upsample_block in enumerate(self.up_blocks):
+ for k in freeu_keys:
+ if hasattr(upsample_block, k) or getattr(upsample_block, k, None) is not None:
+ setattr(upsample_block, k, None)
+
+ def fuse_qkv_projections(self):
+ """
+ Enables fused QKV projections. For self-attention modules, all projection matrices (i.e., query, key, value)
+ are fused. For cross-attention modules, key and value projection matrices are fused.
+
+
+
+ This API is 🧪 experimental.
+
+
+ """
+ self.original_attn_processors = None
+
+ for _, attn_processor in self.attn_processors.items():
+ if "Added" in str(attn_processor.__class__.__name__):
+ raise ValueError("`fuse_qkv_projections()` is not supported for models having added KV projections.")
+
+ self.original_attn_processors = self.attn_processors
+
+ for module in self.modules():
+ if isinstance(module, Attention):
+ module.fuse_projections(fuse=True)
+
+ self.set_attn_processor(FusedAttnProcessor2_0())
+
+ def unfuse_qkv_projections(self):
+ """Disables the fused QKV projection if enabled.
+
+
+
+ This API is 🧪 experimental.
+
+
+
+ """
+ if self.original_attn_processors is not None:
+ self.set_attn_processor(self.original_attn_processors)
+
+ def get_time_embed(
+ self, sample: torch.Tensor, timestep: Union[torch.Tensor, float, int]
+ ) -> Optional[torch.Tensor]:
+ timesteps = timestep
+ if not torch.is_tensor(timesteps):
+ # TODO: this requires sync between CPU and GPU. So try to pass timesteps as tensors if you can
+ # This would be a good case for the `match` statement (Python 3.10+)
+ is_mps = sample.device.type == "mps"
+ if isinstance(timestep, float):
+ dtype = torch.float32 if is_mps else torch.float64
+ else:
+ dtype = torch.int32 if is_mps else torch.int64
+ timesteps = torch.tensor([timesteps], dtype=dtype, device=sample.device)
+ elif len(timesteps.shape) == 0:
+ timesteps = timesteps[None].to(sample.device)
+
+ # broadcast to batch dimension in a way that's compatible with ONNX/Core ML
+ timesteps = timesteps.expand(sample.shape[0])
+
+ t_emb = self.time_proj(timesteps)
+ # `Timesteps` does not contain any weights and will always return f32 tensors
+ # but time_embedding might actually be running in fp16. so we need to cast here.
+ # there might be better ways to encapsulate this.
+ t_emb = t_emb.to(dtype=sample.dtype)
+ return t_emb
+
+ def get_class_embed(self, sample: torch.Tensor, class_labels: Optional[torch.Tensor]) -> Optional[torch.Tensor]:
+ class_emb = None
+ if self.class_embedding is not None:
+ if class_labels is None:
+ raise ValueError("class_labels should be provided when num_class_embeds > 0")
+
+ if self.config.class_embed_type == "timestep":
+ class_labels = self.time_proj(class_labels)
+
+ # `Timesteps` does not contain any weights and will always return f32 tensors
+ # there might be better ways to encapsulate this.
+ class_labels = class_labels.to(dtype=sample.dtype)
+
+ class_emb = self.class_embedding(class_labels).to(dtype=sample.dtype)
+ return class_emb
+
+ def get_aug_embed(
+ self, emb: torch.Tensor, encoder_hidden_states: torch.Tensor, added_cond_kwargs: Dict[str, Any]
+ ) -> Optional[torch.Tensor]:
+ aug_emb = None
+ if self.config.addition_embed_type == "text":
+ aug_emb = self.add_embedding(encoder_hidden_states)
+ elif self.config.addition_embed_type == "matryoshka":
+ aug_emb = self.add_embedding(emb, encoder_hidden_states, added_cond_kwargs)
+ elif self.config.addition_embed_type == "text_image":
+ # Kandinsky 2.1 - style
+ if "image_embeds" not in added_cond_kwargs:
+ raise ValueError(
+ f"{self.__class__} has the config param `addition_embed_type` set to 'text_image' which requires the keyword argument `image_embeds` to be passed in `added_cond_kwargs`"
+ )
+
+ image_embs = added_cond_kwargs.get("image_embeds")
+ text_embs = added_cond_kwargs.get("text_embeds", encoder_hidden_states)
+ aug_emb = self.add_embedding(text_embs, image_embs)
+ elif self.config.addition_embed_type == "text_time":
+ # SDXL - style
+ if "text_embeds" not in added_cond_kwargs:
+ raise ValueError(
+ f"{self.__class__} has the config param `addition_embed_type` set to 'text_time' which requires the keyword argument `text_embeds` to be passed in `added_cond_kwargs`"
+ )
+ text_embeds = added_cond_kwargs.get("text_embeds")
+ if "time_ids" not in added_cond_kwargs:
+ raise ValueError(
+ f"{self.__class__} has the config param `addition_embed_type` set to 'text_time' which requires the keyword argument `time_ids` to be passed in `added_cond_kwargs`"
+ )
+ time_ids = added_cond_kwargs.get("time_ids")
+ time_embeds = self.add_time_proj(time_ids.flatten())
+ time_embeds = time_embeds.reshape((text_embeds.shape[0], -1))
+ add_embeds = torch.concat([text_embeds, time_embeds], dim=-1)
+ add_embeds = add_embeds.to(emb.dtype)
+ aug_emb = self.add_embedding(add_embeds)
+ elif self.config.addition_embed_type == "image":
+ # Kandinsky 2.2 - style
+ if "image_embeds" not in added_cond_kwargs:
+ raise ValueError(
+ f"{self.__class__} has the config param `addition_embed_type` set to 'image' which requires the keyword argument `image_embeds` to be passed in `added_cond_kwargs`"
+ )
+ image_embs = added_cond_kwargs.get("image_embeds")
+ aug_emb = self.add_embedding(image_embs)
+ elif self.config.addition_embed_type == "image_hint":
+ # Kandinsky 2.2 ControlNet - style
+ if "image_embeds" not in added_cond_kwargs or "hint" not in added_cond_kwargs:
+ raise ValueError(
+ f"{self.__class__} has the config param `addition_embed_type` set to 'image_hint' which requires the keyword arguments `image_embeds` and `hint` to be passed in `added_cond_kwargs`"
+ )
+ image_embs = added_cond_kwargs.get("image_embeds")
+ hint = added_cond_kwargs.get("hint")
+ aug_emb = self.add_embedding(image_embs, hint)
+ return aug_emb
+
+ def process_encoder_hidden_states(
+ self, encoder_hidden_states: torch.Tensor, added_cond_kwargs: Dict[str, Any]
+ ) -> torch.Tensor:
+ if self.encoder_hid_proj is not None and self.config.encoder_hid_dim_type == "text_proj":
+ encoder_hidden_states = self.encoder_hid_proj(encoder_hidden_states)
+ elif self.encoder_hid_proj is not None and self.config.encoder_hid_dim_type == "text_image_proj":
+ # Kandinsky 2.1 - style
+ if "image_embeds" not in added_cond_kwargs:
+ raise ValueError(
+ f"{self.__class__} has the config param `encoder_hid_dim_type` set to 'text_image_proj' which requires the keyword argument `image_embeds` to be passed in `added_cond_kwargs`"
+ )
+
+ image_embeds = added_cond_kwargs.get("image_embeds")
+ encoder_hidden_states = self.encoder_hid_proj(encoder_hidden_states, image_embeds)
+ elif self.encoder_hid_proj is not None and self.config.encoder_hid_dim_type == "image_proj":
+ # Kandinsky 2.2 - style
+ if "image_embeds" not in added_cond_kwargs:
+ raise ValueError(
+ f"{self.__class__} has the config param `encoder_hid_dim_type` set to 'image_proj' which requires the keyword argument `image_embeds` to be passed in `added_cond_kwargs`"
+ )
+ image_embeds = added_cond_kwargs.get("image_embeds")
+ encoder_hidden_states = self.encoder_hid_proj(image_embeds)
+ elif self.encoder_hid_proj is not None and self.config.encoder_hid_dim_type == "ip_image_proj":
+ if "image_embeds" not in added_cond_kwargs:
+ raise ValueError(
+ f"{self.__class__} has the config param `encoder_hid_dim_type` set to 'ip_image_proj' which requires the keyword argument `image_embeds` to be passed in `added_cond_kwargs`"
+ )
+
+ if hasattr(self, "text_encoder_hid_proj") and self.text_encoder_hid_proj is not None:
+ encoder_hidden_states = self.text_encoder_hid_proj(encoder_hidden_states)
+
+ image_embeds = added_cond_kwargs.get("image_embeds")
+ image_embeds = self.encoder_hid_proj(image_embeds)
+ encoder_hidden_states = (encoder_hidden_states, image_embeds)
+ return encoder_hidden_states
+
+ @property
+ def model_type(self) -> str:
+ return "unet"
+
+ def forward(
+ self,
+ sample: torch.Tensor,
+ timestep: Union[torch.Tensor, float, int],
+ encoder_hidden_states: torch.Tensor,
+ cond_emb: Optional[torch.Tensor] = None,
+ class_labels: Optional[torch.Tensor] = None,
+ timestep_cond: Optional[torch.Tensor] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+ added_cond_kwargs: Optional[Dict[str, torch.Tensor]] = None,
+ down_block_additional_residuals: Optional[Tuple[torch.Tensor]] = None,
+ mid_block_additional_residual: Optional[torch.Tensor] = None,
+ down_intrablock_additional_residuals: Optional[Tuple[torch.Tensor]] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ return_dict: bool = True,
+ from_nested: bool = False,
+ ) -> Union[MatryoshkaUNet2DConditionOutput, Tuple]:
+ r"""
+ The [`NestedUNet2DConditionModel`] forward method.
+
+ Args:
+ sample (`torch.Tensor`):
+ The noisy input tensor with the following shape `(batch, channel, height, width)`.
+ timestep (`torch.Tensor` or `float` or `int`): The number of timesteps to denoise an input.
+ encoder_hidden_states (`torch.Tensor`):
+ The encoder hidden states with shape `(batch, sequence_length, feature_dim)`.
+ class_labels (`torch.Tensor`, *optional*, defaults to `None`):
+ Optional class labels for conditioning. Their embeddings will be summed with the timestep embeddings.
+ timestep_cond: (`torch.Tensor`, *optional*, defaults to `None`):
+ Conditional embeddings for timestep. If provided, the embeddings will be summed with the samples passed
+ through the `self.time_embedding` layer to obtain the timestep embeddings.
+ attention_mask (`torch.Tensor`, *optional*, defaults to `None`):
+ An attention mask of shape `(batch, key_tokens)` is applied to `encoder_hidden_states`. If `1` the mask
+ is kept, otherwise if `0` it is discarded. Mask will be converted into a bias, which adds large
+ negative values to the attention scores corresponding to "discard" tokens.
+ cross_attention_kwargs (`dict`, *optional*):
+ A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
+ `self.processor` in
+ [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+ added_cond_kwargs: (`dict`, *optional*):
+ A kwargs dictionary containing additional embeddings that if specified are added to the embeddings that
+ are passed along to the UNet blocks.
+ down_block_additional_residuals: (`tuple` of `torch.Tensor`, *optional*):
+ A tuple of tensors that if specified are added to the residuals of down unet blocks.
+ mid_block_additional_residual: (`torch.Tensor`, *optional*):
+ A tensor that if specified is added to the residual of the middle unet block.
+ down_intrablock_additional_residuals (`tuple` of `torch.Tensor`, *optional*):
+ additional residuals to be added within UNet down blocks, for example from T2I-Adapter side model(s)
+ encoder_attention_mask (`torch.Tensor`):
+ A cross-attention mask of shape `(batch, sequence_length)` is applied to `encoder_hidden_states`. If
+ `True` the mask is kept, otherwise if `False` it is discarded. Mask will be converted into a bias,
+ which adds large negative values to the attention scores corresponding to "discard" tokens.
+ return_dict (`bool`, *optional*, defaults to `True`):
+ Whether or not to return a [`~NestedUNet2DConditionOutput`] instead of a plain
+ tuple.
+
+ Returns:
+ [`~NestedUNet2DConditionOutput`] or `tuple`:
+ If `return_dict` is True, an [`~NestedUNet2DConditionOutput`] is returned,
+ otherwise a `tuple` is returned where the first element is the sample tensor.
+ """
+ # By default samples have to be AT least a multiple of the overall upsampling factor.
+ # The overall upsampling factor is equal to 2 ** (# num of upsampling layers).
+ # However, the upsampling interpolation output size can be forced to fit any upsampling size
+ # on the fly if necessary.
+ default_overall_up_factor = 2**self.num_upsamplers
+
+ # upsample size should be forwarded when sample is not a multiple of `default_overall_up_factor`
+ forward_upsample_size = False
+ upsample_size = None
+
+ if self.config.nesting:
+ sample, sample_feat = sample
+ if isinstance(sample, list) and len(sample) == 1:
+ sample = sample[0]
+
+ for dim in sample.shape[-2:]:
+ if dim % default_overall_up_factor != 0:
+ # Forward upsample size to force interpolation output size.
+ forward_upsample_size = True
+ break
+
+ # ensure attention_mask is a bias, and give it a singleton query_tokens dimension
+ # expects mask of shape:
+ # [batch, key_tokens]
+ # adds singleton query_tokens dimension:
+ # [batch, 1, key_tokens]
+ # this helps to broadcast it as a bias over attention scores, which will be in one of the following shapes:
+ # [batch, heads, query_tokens, key_tokens] (e.g. torch sdp attn)
+ # [batch * heads, query_tokens, key_tokens] (e.g. xformers or classic attn)
+ if attention_mask is not None:
+ # assume that mask is expressed as:
+ # (1 = keep, 0 = discard)
+ # convert mask into a bias that can be added to attention scores:
+ # (keep = +0, discard = -10000.0)
+ attention_mask = (1 - attention_mask.to(sample.dtype)) * -10000.0
+ attention_mask = attention_mask.unsqueeze(1)
+
+ # 0. center input if necessary
+ if self.config.center_input_sample:
+ sample = 2 * sample - 1.0
+
+ # 1. time
+ t_emb = self.get_time_embed(sample=sample, timestep=timestep)
+ emb = self.time_embedding(t_emb, timestep_cond)
+
+ class_emb = self.get_class_embed(sample=sample, class_labels=class_labels)
+ if class_emb is not None:
+ if self.config.class_embeddings_concat:
+ emb = torch.cat([emb, class_emb], dim=-1)
+ else:
+ emb = emb + class_emb
+
+ added_cond_kwargs = added_cond_kwargs or {}
+ added_cond_kwargs["masked_cross_attention"] = self.config.masked_cross_attention
+ added_cond_kwargs["micro_conditioning_scale"] = self.config.micro_conditioning_scale
+ added_cond_kwargs["from_nested"] = from_nested
+ added_cond_kwargs["conditioning_mask"] = encoder_attention_mask
+
+ if not from_nested:
+ encoder_hidden_states = self.process_encoder_hidden_states(
+ encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+
+ aug_emb, encoder_attention_mask, cond_emb = self.get_aug_embed(
+ emb=emb, encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+ else:
+ aug_emb, encoder_attention_mask, _ = self.get_aug_embed(
+ emb=emb, encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+
+ # convert encoder_attention_mask to a bias the same way we do for attention_mask
+ if encoder_attention_mask is not None:
+ encoder_attention_mask = (1 - encoder_attention_mask.to(sample[0][0].dtype)) * -10000.0
+ encoder_attention_mask = encoder_attention_mask.unsqueeze(1)
+
+ if self.config.addition_embed_type == "image_hint":
+ aug_emb, hint = aug_emb
+ sample = torch.cat([sample, hint], dim=1)
+
+ emb = emb + aug_emb + cond_emb if aug_emb is not None else emb
+
+ if self.time_embed_act is not None:
+ emb = self.time_embed_act(emb)
+
+ # 2. pre-process
+ sample = self.conv_in(sample)
+ if self.config.nesting:
+ sample = sample + sample_feat
+
+ # 2.5 GLIGEN position net
+ if cross_attention_kwargs is not None and cross_attention_kwargs.get("gligen", None) is not None:
+ cross_attention_kwargs = cross_attention_kwargs.copy()
+ gligen_args = cross_attention_kwargs.pop("gligen")
+ cross_attention_kwargs["gligen"] = {"objs": self.position_net(**gligen_args)}
+
+ # 3. down
+ # we're popping the `scale` instead of getting it because otherwise `scale` will be propagated
+ # to the internal blocks and will raise deprecation warnings. this will be confusing for our users.
+ if cross_attention_kwargs is not None:
+ cross_attention_kwargs = cross_attention_kwargs.copy()
+ lora_scale = cross_attention_kwargs.pop("scale", 1.0)
+ else:
+ lora_scale = 1.0
+
+ if USE_PEFT_BACKEND:
+ # weight the lora layers by setting `lora_scale` for each PEFT layer
+ scale_lora_layers(self, lora_scale)
+
+ is_controlnet = mid_block_additional_residual is not None and down_block_additional_residuals is not None
+ # using new arg down_intrablock_additional_residuals for T2I-Adapters, to distinguish from controlnets
+ is_adapter = down_intrablock_additional_residuals is not None
+ # maintain backward compatibility for legacy usage, where
+ # T2I-Adapter and ControlNet both use down_block_additional_residuals arg
+ # but can only use one or the other
+ if not is_adapter and mid_block_additional_residual is None and down_block_additional_residuals is not None:
+ deprecate(
+ "T2I should not use down_block_additional_residuals",
+ "1.3.0",
+ "Passing intrablock residual connections with `down_block_additional_residuals` is deprecated \
+ and will be removed in diffusers 1.3.0. `down_block_additional_residuals` should only be used \
+ for ControlNet. Please make sure use `down_intrablock_additional_residuals` instead. ",
+ standard_warn=False,
+ )
+ down_intrablock_additional_residuals = down_block_additional_residuals
+ is_adapter = True
+
+ down_block_res_samples = (sample,)
+ for downsample_block in self.down_blocks:
+ if hasattr(downsample_block, "has_cross_attention") and downsample_block.has_cross_attention:
+ # For t2i-adapter CrossAttnDownBlock2D
+ additional_residuals = {}
+ if is_adapter and len(down_intrablock_additional_residuals) > 0:
+ additional_residuals["additional_residuals"] = down_intrablock_additional_residuals.pop(0)
+
+ sample, res_samples = downsample_block(
+ hidden_states=sample,
+ temb=emb,
+ encoder_hidden_states=encoder_hidden_states,
+ attention_mask=attention_mask,
+ cross_attention_kwargs=cross_attention_kwargs,
+ encoder_attention_mask=encoder_attention_mask,
+ **additional_residuals,
+ )
+ else:
+ sample, res_samples = downsample_block(hidden_states=sample, temb=emb)
+ if is_adapter and len(down_intrablock_additional_residuals) > 0:
+ sample += down_intrablock_additional_residuals.pop(0)
+
+ down_block_res_samples += res_samples
+
+ if is_controlnet:
+ new_down_block_res_samples = ()
+
+ for down_block_res_sample, down_block_additional_residual in zip(
+ down_block_res_samples, down_block_additional_residuals
+ ):
+ down_block_res_sample = down_block_res_sample + down_block_additional_residual
+ new_down_block_res_samples = new_down_block_res_samples + (down_block_res_sample,)
+
+ down_block_res_samples = new_down_block_res_samples
+
+ # 4. mid
+ if self.mid_block is not None:
+ if hasattr(self.mid_block, "has_cross_attention") and self.mid_block.has_cross_attention:
+ sample = self.mid_block(
+ sample,
+ emb,
+ encoder_hidden_states=encoder_hidden_states,
+ attention_mask=attention_mask,
+ cross_attention_kwargs=cross_attention_kwargs,
+ encoder_attention_mask=encoder_attention_mask,
+ )
+ else:
+ sample = self.mid_block(sample, emb)
+
+ # To support T2I-Adapter-XL
+ if (
+ is_adapter
+ and len(down_intrablock_additional_residuals) > 0
+ and sample.shape == down_intrablock_additional_residuals[0].shape
+ ):
+ sample += down_intrablock_additional_residuals.pop(0)
+
+ if is_controlnet:
+ sample = sample + mid_block_additional_residual
+
+ # 5. up
+ for i, upsample_block in enumerate(self.up_blocks):
+ is_final_block = i == len(self.up_blocks) - 1
+
+ res_samples = down_block_res_samples[-len(upsample_block.resnets) :]
+ down_block_res_samples = down_block_res_samples[: -len(upsample_block.resnets)]
+
+ # if we have not reached the final block and need to forward the
+ # upsample size, we do it here
+ if not is_final_block and forward_upsample_size:
+ upsample_size = down_block_res_samples[-1].shape[2:]
+
+ if hasattr(upsample_block, "has_cross_attention") and upsample_block.has_cross_attention:
+ sample = upsample_block(
+ hidden_states=sample,
+ temb=emb,
+ res_hidden_states_tuple=res_samples,
+ encoder_hidden_states=encoder_hidden_states,
+ cross_attention_kwargs=cross_attention_kwargs,
+ upsample_size=upsample_size,
+ attention_mask=attention_mask,
+ encoder_attention_mask=encoder_attention_mask,
+ )
+ else:
+ sample = upsample_block(
+ hidden_states=sample,
+ temb=emb,
+ res_hidden_states_tuple=res_samples,
+ upsample_size=upsample_size,
+ )
+
+ sample_inner = sample
+
+ # 6. post-process
+ if self.conv_norm_out:
+ sample = self.conv_norm_out(sample_inner)
+ sample = self.conv_act(sample)
+ sample = self.conv_out(sample)
+
+ if USE_PEFT_BACKEND:
+ # remove `lora_scale` from each PEFT layer
+ unscale_lora_layers(self, lora_scale)
+
+ if not return_dict:
+ return (sample,)
+
+ if self.config.nesting:
+ return MatryoshkaUNet2DConditionOutput(sample=sample, sample_inner=sample_inner)
+
+ return MatryoshkaUNet2DConditionOutput(sample=sample)
+
+
+class NestedUNet2DConditionOutput(BaseOutput):
+ """
+ Output type for the [`NestedUNet2DConditionModel`] model.
+ """
+
+ sample: list = None
+ sample_inner: torch.Tensor = None
+
+
+class NestedUNet2DConditionModel(MatryoshkaUNet2DConditionModel):
+ """
+ Nested UNet model with condition for image denoising.
+ """
+
+ @register_to_config
+ def __init__(
+ self,
+ in_channels=3,
+ out_channels=3,
+ block_out_channels=(64, 128, 256),
+ cross_attention_dim=2048,
+ resnet_time_scale_shift="scale_shift",
+ down_block_types=("DownBlock2D", "DownBlock2D", "DownBlock2D"),
+ up_block_types=("UpBlock2D", "UpBlock2D", "UpBlock2D"),
+ mid_block_type=None,
+ nesting=False,
+ flip_sin_to_cos=False,
+ transformer_layers_per_block=[0, 0, 0],
+ layers_per_block=[2, 2, 1],
+ masked_cross_attention=True,
+ micro_conditioning_scale=256,
+ addition_embed_type="matryoshka",
+ skip_normalization=True,
+ time_embedding_dim=1024,
+ skip_inner_unet_input=False,
+ temporal_mode=False,
+ temporal_spatial_ds=False,
+ initialize_inner_with_pretrained=None,
+ use_attention_ffn=False,
+ act_fn="silu",
+ addition_embed_type_num_heads=64,
+ addition_time_embed_dim=None,
+ attention_head_dim=8,
+ attention_pre_only=False,
+ attention_type="default",
+ center_input_sample=False,
+ class_embed_type=None,
+ class_embeddings_concat=False,
+ conv_in_kernel=3,
+ conv_out_kernel=3,
+ cross_attention_norm=None,
+ downsample_padding=1,
+ dropout=0.0,
+ dual_cross_attention=False,
+ encoder_hid_dim=None,
+ encoder_hid_dim_type=None,
+ freq_shift=0,
+ mid_block_only_cross_attention=None,
+ mid_block_scale_factor=1,
+ norm_eps=1e-05,
+ norm_num_groups=32,
+ norm_type="layer_norm",
+ num_attention_heads=None,
+ num_class_embeds=None,
+ only_cross_attention=False,
+ projection_class_embeddings_input_dim=None,
+ resnet_out_scale_factor=1.0,
+ resnet_skip_time_act=False,
+ reverse_transformer_layers_per_block=None,
+ sample_size=None,
+ skip_cond_emb=False,
+ time_cond_proj_dim=None,
+ time_embedding_act_fn=None,
+ time_embedding_type="positional",
+ timestep_post_act=None,
+ upcast_attention=False,
+ use_linear_projection=False,
+ is_temporal=None,
+ inner_config={},
+ ):
+ super().__init__(
+ in_channels=in_channels,
+ out_channels=out_channels,
+ block_out_channels=block_out_channels,
+ cross_attention_dim=cross_attention_dim,
+ resnet_time_scale_shift=resnet_time_scale_shift,
+ down_block_types=down_block_types,
+ up_block_types=up_block_types,
+ mid_block_type=mid_block_type,
+ nesting=nesting,
+ flip_sin_to_cos=flip_sin_to_cos,
+ transformer_layers_per_block=transformer_layers_per_block,
+ layers_per_block=layers_per_block,
+ masked_cross_attention=masked_cross_attention,
+ micro_conditioning_scale=micro_conditioning_scale,
+ addition_embed_type=addition_embed_type,
+ time_embedding_dim=time_embedding_dim,
+ temporal_mode=temporal_mode,
+ temporal_spatial_ds=temporal_spatial_ds,
+ use_attention_ffn=use_attention_ffn,
+ sample_size=sample_size,
+ )
+ # self.config.inner_config.conditioning_feature_dim = self.config.conditioning_feature_dim
+
+ if "inner_config" not in self.config.inner_config:
+ self.inner_unet = MatryoshkaUNet2DConditionModel(**self.config.inner_config)
+ else:
+ self.inner_unet = NestedUNet2DConditionModel(**self.config.inner_config)
+
+ if not self.config.skip_inner_unet_input:
+ self.in_adapter = nn.Conv2d(
+ self.config.block_out_channels[-1],
+ self.config.inner_config["block_out_channels"][0],
+ kernel_size=3,
+ padding=1,
+ )
+ else:
+ self.in_adapter = None
+ self.out_adapter = nn.Conv2d(
+ self.config.inner_config["block_out_channels"][0],
+ self.config.block_out_channels[-1],
+ kernel_size=3,
+ padding=1,
+ )
+
+ self.is_temporal = [self.config.temporal_mode and (not self.config.temporal_spatial_ds)]
+ if hasattr(self.inner_unet, "is_temporal"):
+ self.is_temporal = self.is_temporal + self.inner_unet.is_temporal
+
+ nest_ratio = int(2 ** (len(self.config.block_out_channels) - 1))
+ if self.is_temporal[0]:
+ nest_ratio = int(np.sqrt(nest_ratio))
+ if self.inner_unet.config.nesting and self.inner_unet.model_type == "nested_unet":
+ self.nest_ratio = [nest_ratio * self.inner_unet.nest_ratio[0]] + self.inner_unet.nest_ratio
+ else:
+ self.nest_ratio = [nest_ratio]
+
+ # self.register_modules(inner_unet=self.inner_unet)
+
+ @property
+ def model_type(self):
+ return "nested_unet"
+
+ def forward(
+ self,
+ sample: torch.Tensor,
+ timestep: Union[torch.Tensor, float, int],
+ encoder_hidden_states: torch.Tensor,
+ cond_emb: Optional[torch.Tensor] = None,
+ from_nested: bool = False,
+ class_labels: Optional[torch.Tensor] = None,
+ timestep_cond: Optional[torch.Tensor] = None,
+ attention_mask: Optional[torch.Tensor] = None,
+ cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+ added_cond_kwargs: Optional[Dict[str, torch.Tensor]] = None,
+ down_block_additional_residuals: Optional[Tuple[torch.Tensor]] = None,
+ mid_block_additional_residual: Optional[torch.Tensor] = None,
+ down_intrablock_additional_residuals: Optional[Tuple[torch.Tensor]] = None,
+ encoder_attention_mask: Optional[torch.Tensor] = None,
+ return_dict: bool = True,
+ ) -> Union[MatryoshkaUNet2DConditionOutput, Tuple]:
+ r"""
+ The [`NestedUNet2DConditionModel`] forward method.
+
+ Args:
+ sample (`torch.Tensor`):
+ The noisy input tensor with the following shape `(batch, channel, height, width)`.
+ timestep (`torch.Tensor` or `float` or `int`): The number of timesteps to denoise an input.
+ encoder_hidden_states (`torch.Tensor`):
+ The encoder hidden states with shape `(batch, sequence_length, feature_dim)`.
+ class_labels (`torch.Tensor`, *optional*, defaults to `None`):
+ Optional class labels for conditioning. Their embeddings will be summed with the timestep embeddings.
+ timestep_cond: (`torch.Tensor`, *optional*, defaults to `None`):
+ Conditional embeddings for timestep. If provided, the embeddings will be summed with the samples passed
+ through the `self.time_embedding` layer to obtain the timestep embeddings.
+ attention_mask (`torch.Tensor`, *optional*, defaults to `None`):
+ An attention mask of shape `(batch, key_tokens)` is applied to `encoder_hidden_states`. If `1` the mask
+ is kept, otherwise if `0` it is discarded. Mask will be converted into a bias, which adds large
+ negative values to the attention scores corresponding to "discard" tokens.
+ cross_attention_kwargs (`dict`, *optional*):
+ A kwargs dictionary that if specified is passed along to the `AttentionProcessor` as defined under
+ `self.processor` in
+ [diffusers.models.attention_processor](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+ added_cond_kwargs: (`dict`, *optional*):
+ A kwargs dictionary containing additional embeddings that if specified are added to the embeddings that
+ are passed along to the UNet blocks.
+ down_block_additional_residuals: (`tuple` of `torch.Tensor`, *optional*):
+ A tuple of tensors that if specified are added to the residuals of down unet blocks.
+ mid_block_additional_residual: (`torch.Tensor`, *optional*):
+ A tensor that if specified is added to the residual of the middle unet block.
+ down_intrablock_additional_residuals (`tuple` of `torch.Tensor`, *optional*):
+ additional residuals to be added within UNet down blocks, for example from T2I-Adapter side model(s)
+ encoder_attention_mask (`torch.Tensor`):
+ A cross-attention mask of shape `(batch, sequence_length)` is applied to `encoder_hidden_states`. If
+ `True` the mask is kept, otherwise if `False` it is discarded. Mask will be converted into a bias,
+ which adds large negative values to the attention scores corresponding to "discard" tokens.
+ return_dict (`bool`, *optional*, defaults to `True`):
+ Whether or not to return a [`~NestedUNet2DConditionOutput`] instead of a plain
+ tuple.
+
+ Returns:
+ [`~NestedUNet2DConditionOutput`] or `tuple`:
+ If `return_dict` is True, an [`~NestedUNet2DConditionOutput`] is returned,
+ otherwise a `tuple` is returned where the first element is the sample tensor.
+ """
+ # By default samples have to be AT least a multiple of the overall upsampling factor.
+ # The overall upsampling factor is equal to 2 ** (# num of upsampling layers).
+ # However, the upsampling interpolation output size can be forced to fit any upsampling size
+ # on the fly if necessary.
+ default_overall_up_factor = 2**self.num_upsamplers
+
+ # upsample size should be forwarded when sample is not a multiple of `default_overall_up_factor`
+ forward_upsample_size = False
+ upsample_size = None
+
+ if self.config.nesting:
+ sample, sample_feat = sample
+ if isinstance(sample, list) and len(sample) == 1:
+ sample = sample[0]
+
+ # 2. input layer (normalize the input)
+ bsz = [x.size(0) for x in sample]
+ bh, bl = bsz[0], bsz[1]
+ x_t_low, sample = sample[1:], sample[0]
+
+ for dim in sample.shape[-2:]:
+ if dim % default_overall_up_factor != 0:
+ # Forward upsample size to force interpolation output size.
+ forward_upsample_size = True
+ break
+
+ # ensure attention_mask is a bias, and give it a singleton query_tokens dimension
+ # expects mask of shape:
+ # [batch, key_tokens]
+ # adds singleton query_tokens dimension:
+ # [batch, 1, key_tokens]
+ # this helps to broadcast it as a bias over attention scores, which will be in one of the following shapes:
+ # [batch, heads, query_tokens, key_tokens] (e.g. torch sdp attn)
+ # [batch * heads, query_tokens, key_tokens] (e.g. xformers or classic attn)
+ if attention_mask is not None:
+ # assume that mask is expressed as:
+ # (1 = keep, 0 = discard)
+ # convert mask into a bias that can be added to attention scores:
+ # (keep = +0, discard = -10000.0)
+ attention_mask = (1 - attention_mask.to(sample.dtype)) * -10000.0
+ attention_mask = attention_mask.unsqueeze(1)
+
+ # 0. center input if necessary
+ if self.config.center_input_sample:
+ sample = 2 * sample - 1.0
+
+ # 1. time
+ t_emb = self.get_time_embed(sample=sample, timestep=timestep)
+ emb = self.time_embedding(t_emb, timestep_cond)
+
+ class_emb = self.get_class_embed(sample=sample, class_labels=class_labels)
+ if class_emb is not None:
+ if self.config.class_embeddings_concat:
+ emb = torch.cat([emb, class_emb], dim=-1)
+ else:
+ emb = emb + class_emb
+
+ if self.inner_unet.model_type == "unet":
+ added_cond_kwargs = added_cond_kwargs or {}
+ added_cond_kwargs["masked_cross_attention"] = self.inner_unet.config.masked_cross_attention
+ added_cond_kwargs["micro_conditioning_scale"] = self.config.micro_conditioning_scale
+ added_cond_kwargs["conditioning_mask"] = encoder_attention_mask
+
+ if not self.config.nesting:
+ encoder_hidden_states = self.inner_unet.process_encoder_hidden_states(
+ encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+
+ aug_emb_inner_unet, cond_mask, cond_emb = self.inner_unet.get_aug_embed(
+ emb=emb, encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+ added_cond_kwargs["masked_cross_attention"] = self.config.masked_cross_attention
+ aug_emb, __, _ = self.get_aug_embed(
+ emb=emb, encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+ else:
+ aug_emb, cond_mask, _ = self.get_aug_embed(
+ emb=emb, encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+
+ elif self.inner_unet.model_type == "nested_unet":
+ added_cond_kwargs = added_cond_kwargs or {}
+ added_cond_kwargs["masked_cross_attention"] = self.inner_unet.inner_unet.config.masked_cross_attention
+ added_cond_kwargs["micro_conditioning_scale"] = self.config.micro_conditioning_scale
+ added_cond_kwargs["conditioning_mask"] = encoder_attention_mask
+
+ encoder_hidden_states = self.inner_unet.inner_unet.process_encoder_hidden_states(
+ encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+
+ aug_emb_inner_unet, cond_mask, cond_emb = self.inner_unet.inner_unet.get_aug_embed(
+ emb=emb, encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+
+ aug_emb, __, _ = self.get_aug_embed(
+ emb=emb, encoder_hidden_states=encoder_hidden_states, added_cond_kwargs=added_cond_kwargs
+ )
+
+ # convert encoder_attention_mask to a bias the same way we do for attention_mask
+ if encoder_attention_mask is not None:
+ encoder_attention_mask = (1 - encoder_attention_mask.to(sample.dtype)) * -10000.0
+ encoder_attention_mask = encoder_attention_mask.unsqueeze(1)
+
+ if self.config.addition_embed_type == "image_hint":
+ aug_emb, hint = aug_emb
+ sample = torch.cat([sample, hint], dim=1)
+
+ emb = emb + aug_emb + cond_emb if aug_emb is not None else emb
+
+ if self.time_embed_act is not None:
+ emb = self.time_embed_act(emb)
+
+ if not self.config.skip_normalization:
+ sample = sample / sample.std((1, 2, 3), keepdims=True)
+ if isinstance(sample, list) and len(sample) == 1:
+ sample = sample[0]
+ sample = self.conv_in(sample)
+ if self.config.nesting:
+ sample = sample + sample_feat
+
+ # we're popping the `scale` instead of getting it because otherwise `scale` will be propagated
+ # to the internal blocks and will raise deprecation warnings. this will be confusing for our users.
+ if cross_attention_kwargs is not None:
+ cross_attention_kwargs = cross_attention_kwargs.copy()
+ lora_scale = cross_attention_kwargs.pop("scale", 1.0)
+ else:
+ lora_scale = 1.0
+
+ if USE_PEFT_BACKEND:
+ # weight the lora layers by setting `lora_scale` for each PEFT layer
+ scale_lora_layers(self, lora_scale)
+
+ # using new arg down_intrablock_additional_residuals for T2I-Adapters, to distinguish from controlnets
+ is_adapter = down_intrablock_additional_residuals is not None
+ # maintain backward compatibility for legacy usage, where
+ # T2I-Adapter and ControlNet both use down_block_additional_residuals arg
+ # but can only use one or the other
+ if not is_adapter and mid_block_additional_residual is None and down_block_additional_residuals is not None:
+ deprecate(
+ "T2I should not use down_block_additional_residuals",
+ "1.3.0",
+ "Passing intrablock residual connections with `down_block_additional_residuals` is deprecated \
+ and will be removed in diffusers 1.3.0. `down_block_additional_residuals` should only be used \
+ for ControlNet. Please make sure use `down_intrablock_additional_residuals` instead. ",
+ standard_warn=False,
+ )
+ down_intrablock_additional_residuals = down_block_additional_residuals
+ is_adapter = True
+
+ # 3. downsample blocks in the outer layers
+ down_block_res_samples = (sample,)
+ for downsample_block in self.down_blocks:
+ if hasattr(downsample_block, "has_cross_attention") and downsample_block.has_cross_attention:
+ # For t2i-adapter CrossAttnDownBlock2D
+ additional_residuals = {}
+ if is_adapter and len(down_intrablock_additional_residuals) > 0:
+ additional_residuals["additional_residuals"] = down_intrablock_additional_residuals.pop(0)
+
+ sample, res_samples = downsample_block(
+ hidden_states=sample,
+ temb=emb[:bh],
+ encoder_hidden_states=encoder_hidden_states[:bh],
+ attention_mask=attention_mask,
+ cross_attention_kwargs=cross_attention_kwargs,
+ encoder_attention_mask=cond_mask[:bh] if cond_mask is not None else cond_mask,
+ **additional_residuals,
+ )
+ else:
+ sample, res_samples = downsample_block(hidden_states=sample, temb=emb)
+ if is_adapter and len(down_intrablock_additional_residuals) > 0:
+ sample += down_intrablock_additional_residuals.pop(0)
+
+ down_block_res_samples += res_samples
+
+ # 4. run inner unet
+ x_inner = self.in_adapter(sample) if self.in_adapter is not None else None
+ x_inner = (
+ torch.cat([x_inner, x_inner.new_zeros(bl - bh, *x_inner.size()[1:])], 0) if bh < bl else x_inner
+ ) # pad zeros for low-resolutions
+ inner_unet_output = self.inner_unet(
+ (x_t_low, x_inner),
+ timestep,
+ cond_emb=cond_emb,
+ encoder_hidden_states=encoder_hidden_states,
+ encoder_attention_mask=cond_mask,
+ from_nested=True,
+ )
+ x_low, x_inner = inner_unet_output.sample, inner_unet_output.sample_inner
+ x_inner = self.out_adapter(x_inner)
+ sample = sample + x_inner[:bh] if bh < bl else sample + x_inner
+
+ # 5. upsample blocks in the outer layers
+ for i, upsample_block in enumerate(self.up_blocks):
+ is_final_block = i == len(self.up_blocks) - 1
+
+ res_samples = down_block_res_samples[-len(upsample_block.resnets) :]
+ down_block_res_samples = down_block_res_samples[: -len(upsample_block.resnets)]
+
+ # if we have not reached the final block and need to forward the
+ # upsample size, we do it here
+ if not is_final_block and forward_upsample_size:
+ upsample_size = down_block_res_samples[-1].shape[2:]
+
+ if hasattr(upsample_block, "has_cross_attention") and upsample_block.has_cross_attention:
+ sample = upsample_block(
+ hidden_states=sample,
+ temb=emb[:bh],
+ res_hidden_states_tuple=res_samples,
+ encoder_hidden_states=encoder_hidden_states[:bh],
+ cross_attention_kwargs=cross_attention_kwargs,
+ upsample_size=upsample_size,
+ attention_mask=attention_mask,
+ encoder_attention_mask=cond_mask[:bh] if cond_mask is not None else cond_mask,
+ )
+ else:
+ sample = upsample_block(
+ hidden_states=sample,
+ temb=emb,
+ res_hidden_states_tuple=res_samples,
+ upsample_size=upsample_size,
+ )
+
+ # 6. post-process
+ if self.conv_norm_out:
+ sample_out = self.conv_norm_out(sample)
+ sample_out = self.conv_act(sample_out)
+ sample_out = self.conv_out(sample_out)
+
+ if USE_PEFT_BACKEND:
+ # remove `lora_scale` from each PEFT layer
+ unscale_lora_layers(self, lora_scale)
+
+ # 7. output both low and high-res output
+ if isinstance(x_low, list):
+ out = [sample_out] + x_low
+ else:
+ out = [sample_out, x_low]
+ if self.config.nesting:
+ return NestedUNet2DConditionOutput(sample=out, sample_inner=sample)
+ if not return_dict:
+ return (out,)
+ else:
+ return NestedUNet2DConditionOutput(sample=out)
+
+
+@dataclass
+class MatryoshkaPipelineOutput(BaseOutput):
+ """
+ Output class for Matryoshka pipelines.
+
+ Args:
+ images (`List[PIL.Image.Image]` or `np.ndarray`)
+ List of denoised PIL images of length `batch_size` or numpy array of shape `(batch_size, height, width,
+ num_channels)`. PIL images or numpy array present the denoised images of the diffusion pipeline.
+ """
+
+ images: Union[List[Image.Image], List[List[Image.Image]], np.ndarray, List[np.ndarray]]
+
+
+class MatryoshkaPipeline(
+ DiffusionPipeline,
+ StableDiffusionMixin,
+ TextualInversionLoaderMixin,
+ StableDiffusionLoraLoaderMixin,
+ IPAdapterMixin,
+ FromSingleFileMixin,
+):
+ r"""
+ Pipeline for text-to-image generation using Matryoshka Diffusion Models.
+
+ This model inherits from [`DiffusionPipeline`]. Check the superclass documentation for the generic methods
+ implemented for all pipelines (downloading, saving, running on a particular device, etc.).
+
+ The pipeline also inherits the following loading methods:
+ - [`~loaders.TextualInversionLoaderMixin.load_textual_inversion`] for loading textual inversion embeddings
+ - [`~loaders.StableDiffusionLoraLoaderMixin.load_lora_weights`] for loading LoRA weights
+ - [`~loaders.StableDiffusionLoraLoaderMixin.save_lora_weights`] for saving LoRA weights
+ - [`~loaders.FromSingleFileMixin.from_single_file`] for loading `.ckpt` files
+ - [`~loaders.IPAdapterMixin.load_ip_adapter`] for loading IP Adapters
+
+ Args:
+ text_encoder ([`~transformers.T5EncoderModel`]):
+ Frozen text-encoder ([flan-t5-xl](https://huggingface.co/google/flan-t5-xl)).
+ tokenizer ([`~transformers.T5Tokenizer`]):
+ A `T5Tokenizer` to tokenize text.
+ unet ([`MatryoshkaUNet2DConditionModel`]):
+ A `MatryoshkaUNet2DConditionModel` to denoise the encoded image latents.
+ scheduler ([`SchedulerMixin`]):
+ A scheduler to be used in combination with `unet` to denoise the encoded image latents. Can be one of
+ [`MatryoshkaDDIMScheduler`] and other schedulers with proper modifications, see an example usage in README.md.
+ feature_extractor ([`~transformers.`]):
+ A `AnImageProcessor` to extract features from generated images; used as inputs to the `safety_checker`.
+ """
+
+ model_cpu_offload_seq = "text_encoder->image_encoder->unet"
+ _optional_components = ["unet", "feature_extractor", "image_encoder"]
+ _callback_tensor_inputs = ["latents", "prompt_embeds", "negative_prompt_embeds"]
+
+ def __init__(
+ self,
+ text_encoder: T5EncoderModel,
+ tokenizer: T5TokenizerFast,
+ scheduler: MatryoshkaDDIMScheduler,
+ unet: MatryoshkaUNet2DConditionModel = None,
+ feature_extractor: CLIPImageProcessor = None,
+ image_encoder: CLIPVisionModelWithProjection = None,
+ trust_remote_code: bool = False,
+ nesting_level: int = 0,
+ ):
+ super().__init__()
+
+ if nesting_level == 0:
+ unet = MatryoshkaUNet2DConditionModel.from_pretrained(
+ "tolgacangoz/matryoshka-diffusion-models", subfolder="unet/nesting_level_0"
+ )
+ elif nesting_level == 1:
+ unet = NestedUNet2DConditionModel.from_pretrained(
+ "tolgacangoz/matryoshka-diffusion-models", subfolder="unet/nesting_level_1"
+ )
+ elif nesting_level == 2:
+ unet = NestedUNet2DConditionModel.from_pretrained(
+ "tolgacangoz/matryoshka-diffusion-models", subfolder="unet/nesting_level_2"
+ )
+ else:
+ raise ValueError("Currently, nesting levels 0, 1, and 2 are supported.")
+
+ if hasattr(scheduler.config, "steps_offset") and scheduler.config.steps_offset != 1:
+ deprecation_message = (
+ f"The configuration file of this scheduler: {scheduler} is outdated. `steps_offset`"
+ f" should be set to 1 instead of {scheduler.config.steps_offset}. Please make sure "
+ "to update the config accordingly as leaving `steps_offset` might led to incorrect results"
+ " in future versions. If you have downloaded this checkpoint from the Hugging Face Hub,"
+ " it would be very nice if you could open a Pull request for the `scheduler/scheduler_config.json`"
+ " file"
+ )
+ deprecate("steps_offset!=1", "1.0.0", deprecation_message, standard_warn=False)
+ new_config = dict(scheduler.config)
+ new_config["steps_offset"] = 1
+ scheduler._internal_dict = FrozenDict(new_config)
+
+ # if hasattr(scheduler.config, "clip_sample") and scheduler.config.clip_sample is True:
+ # deprecation_message = (
+ # f"The configuration file of this scheduler: {scheduler} has not set the configuration `clip_sample`."
+ # " `clip_sample` should be set to False in the configuration file. Please make sure to update the"
+ # " config accordingly as not setting `clip_sample` in the config might lead to incorrect results in"
+ # " future versions. If you have downloaded this checkpoint from the Hugging Face Hub, it would be very"
+ # " nice if you could open a Pull request for the `scheduler/scheduler_config.json` file"
+ # )
+ # deprecate("clip_sample not set", "1.0.0", deprecation_message, standard_warn=False)
+ # new_config = dict(scheduler.config)
+ # new_config["clip_sample"] = False
+ # scheduler._internal_dict = FrozenDict(new_config)
+
+ is_unet_version_less_0_9_0 = hasattr(unet.config, "_diffusers_version") and version.parse(
+ version.parse(unet.config._diffusers_version).base_version
+ ) < version.parse("0.9.0.dev0")
+ is_unet_sample_size_less_64 = hasattr(unet.config, "sample_size") and unet.config.sample_size < 64
+ if is_unet_version_less_0_9_0 and is_unet_sample_size_less_64:
+ deprecation_message = (
+ "The configuration file of the unet has set the default `sample_size` to smaller than"
+ " 64 which seems highly unlikely. If your checkpoint is a fine-tuned version of any of the"
+ " following: \n- CompVis/stable-diffusion-v1-4 \n- CompVis/stable-diffusion-v1-3 \n-"
+ " CompVis/stable-diffusion-v1-2 \n- CompVis/stable-diffusion-v1-1 \n- runwayml/stable-diffusion-v1-5"
+ " \n- runwayml/stable-diffusion-inpainting \n you should change 'sample_size' to 64 in the"
+ " configuration file. Please make sure to update the config accordingly as leaving `sample_size=32`"
+ " in the config might lead to incorrect results in future versions. If you have downloaded this"
+ " checkpoint from the Hugging Face Hub, it would be very nice if you could open a Pull request for"
+ " the `unet/config.json` file"
+ )
+ deprecate("sample_size<64", "1.0.0", deprecation_message, standard_warn=False)
+ new_config = dict(unet.config)
+ new_config["sample_size"] = 64
+ unet._internal_dict = FrozenDict(new_config)
+
+ if hasattr(unet, "nest_ratio"):
+ scheduler.scales = unet.nest_ratio + [1]
+
+ self.register_modules(
+ text_encoder=text_encoder,
+ tokenizer=tokenizer,
+ unet=unet,
+ scheduler=scheduler,
+ feature_extractor=feature_extractor,
+ image_encoder=image_encoder,
+ )
+ self.register_to_config(nesting_level=nesting_level)
+ self.image_processor = VaeImageProcessor(do_resize=False)
+
+ def change_nesting_level(self, nesting_level: int):
+ if nesting_level == 0:
+ if hasattr(self.unet, "nest_ratio"):
+ self.scheduler.scales = None
+ self.unet = MatryoshkaUNet2DConditionModel.from_pretrained(
+ "tolgacangoz/matryoshka-diffusion-models", subfolder="unet/nesting_level_0"
+ ).to(self.device)
+ self.config.nesting_level = 0
+ elif nesting_level == 1:
+ self.unet = NestedUNet2DConditionModel.from_pretrained(
+ "tolgacangoz/matryoshka-diffusion-models", subfolder="unet/nesting_level_1"
+ ).to(self.device)
+ self.config.nesting_level = 1
+ self.scheduler.scales = self.unet.nest_ratio + [1]
+ elif nesting_level == 2:
+ self.unet = NestedUNet2DConditionModel.from_pretrained(
+ "tolgacangoz/matryoshka-diffusion-models", subfolder="unet/nesting_level_2"
+ ).to(self.device)
+ self.config.nesting_level = 2
+ self.scheduler.scales = self.unet.nest_ratio + [1]
+ else:
+ raise ValueError("Currently, nesting levels 0, 1, and 2 are supported.")
+
+ gc.collect()
+ torch.cuda.empty_cache()
+
+ def encode_prompt(
+ self,
+ prompt,
+ device,
+ num_images_per_prompt,
+ do_classifier_free_guidance,
+ negative_prompt=None,
+ prompt_embeds: Optional[torch.Tensor] = None,
+ negative_prompt_embeds: Optional[torch.Tensor] = None,
+ lora_scale: Optional[float] = None,
+ clip_skip: Optional[int] = None,
+ ):
+ r"""
+ Encodes the prompt into text encoder hidden states.
+
+ Args:
+ prompt (`str` or `List[str]`, *optional*):
+ prompt to be encoded
+ device: (`torch.device`):
+ torch device
+ num_images_per_prompt (`int`):
+ number of images that should be generated per prompt
+ do_classifier_free_guidance (`bool`):
+ whether to use classifier free guidance or not
+ negative_prompt (`str` or `List[str]`, *optional*):
+ The prompt or prompts not to guide the image generation. If not defined, one has to pass
+ `negative_prompt_embeds` instead. Ignored when not using guidance (i.e., ignored if `guidance_scale` is
+ less than `1`).
+ prompt_embeds (`torch.Tensor`, *optional*):
+ Pre-generated text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt weighting. If not
+ provided, text embeddings will be generated from `prompt` input argument.
+ negative_prompt_embeds (`torch.Tensor`, *optional*):
+ Pre-generated negative text embeddings. Can be used to easily tweak text inputs, *e.g.* prompt
+ weighting. If not provided, negative_prompt_embeds will be generated from `negative_prompt` input
+ argument.
+ lora_scale (`float`, *optional*):
+ A LoRA scale that will be applied to all LoRA layers of the text encoder if LoRA layers are loaded.
+ clip_skip (`int`, *optional*):
+ Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
+ the output of the pre-final layer will be used for computing the prompt embeddings.
+ """
+ # set lora scale so that monkey patched LoRA
+ # function of text encoder can correctly access it
+ if lora_scale is not None and isinstance(self, StableDiffusionLoraLoaderMixin):
+ self._lora_scale = lora_scale
+
+ # dynamically adjust the LoRA scale
+ if not USE_PEFT_BACKEND:
+ adjust_lora_scale_text_encoder(self.text_encoder, lora_scale)
+ else:
+ scale_lora_layers(self.text_encoder, lora_scale)
+
+ if prompt is not None and isinstance(prompt, str):
+ batch_size = 1
+ elif prompt is not None and isinstance(prompt, list):
+ batch_size = len(prompt)
+ else:
+ batch_size = prompt_embeds.shape[0]
+
+ if prompt_embeds is None:
+ # textual inversion: process multi-vector tokens if necessary
+ if isinstance(self, TextualInversionLoaderMixin):
+ prompt = self.maybe_convert_prompt(prompt, self.tokenizer)
+
+ text_inputs = self.tokenizer(
+ prompt,
+ return_tensors="pt",
+ )
+ text_input_ids = text_inputs.input_ids
+ untruncated_ids = self.tokenizer(prompt, padding="longest", return_tensors="pt").input_ids
+
+ if untruncated_ids.shape[-1] >= text_input_ids.shape[-1] and not torch.equal(
+ text_input_ids, untruncated_ids
+ ):
+ removed_text = self.tokenizer.batch_decode(
+ untruncated_ids[:, self.tokenizer.model_max_length - 1 : -1]
+ )
+ logger.warning(
+ "The following part of your input was truncated because FLAN-T5-XL for this pipeline can only handle sequences up to"
+ f" {self.tokenizer.model_max_length} tokens: {removed_text}"
+ )
+
+ if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+ prompt_attention_mask = text_inputs.attention_mask.to(device)
+ else:
+ prompt_attention_mask = None
+
+ if self.text_encoder is not None:
+ prompt_embeds_dtype = self.text_encoder.dtype
+ elif self.unet is not None:
+ prompt_embeds_dtype = self.unet.dtype
+ else:
+ prompt_embeds_dtype = prompt_embeds.dtype
+
+ # get unconditional embeddings for classifier free guidance
+ if do_classifier_free_guidance and negative_prompt_embeds is None:
+ uncond_tokens: List[str]
+ if negative_prompt is None:
+ uncond_tokens = [""] * batch_size
+ elif prompt is not None and type(prompt) is not type(negative_prompt):
+ raise TypeError(
+ f"`negative_prompt` should be the same type to `prompt`, but got {type(negative_prompt)} !="
+ f" {type(prompt)}."
+ )
+ elif isinstance(negative_prompt, str):
+ uncond_tokens = [negative_prompt]
+ elif batch_size != len(negative_prompt):
+ raise ValueError(
+ f"`negative_prompt`: {negative_prompt} has batch size {len(negative_prompt)}, but `prompt`:"
+ f" {prompt} has batch size {batch_size}. Please make sure that passed `negative_prompt` matches"
+ " the batch size of `prompt`."
+ )
+ else:
+ uncond_tokens = negative_prompt
+
+ # textual inversion: process multi-vector tokens if necessary
+ if isinstance(self, TextualInversionLoaderMixin):
+ uncond_tokens = self.maybe_convert_prompt(uncond_tokens, self.tokenizer)
+
+ uncond_input = self.tokenizer(
+ uncond_tokens,
+ return_tensors="pt",
+ )
+ uncond_input_ids = uncond_input.input_ids
+
+ if hasattr(self.text_encoder.config, "use_attention_mask") and self.text_encoder.config.use_attention_mask:
+ negative_prompt_attention_mask = uncond_input.attention_mask.to(device)
+ else:
+ negative_prompt_attention_mask = None
+
+ if not do_classifier_free_guidance:
+ if clip_skip is None:
+ prompt_embeds = self.text_encoder(text_input_ids.to(device), attention_mask=prompt_attention_mask)
+ prompt_embeds = prompt_embeds[0]
+ else:
+ prompt_embeds = self.text_encoder(
+ text_input_ids.to(device), attention_mask=prompt_attention_mask, output_hidden_states=True
+ )
+ # Access the `hidden_states` first, that contains a tuple of
+ # all the hidden states from the encoder layers. Then index into
+ # the tuple to access the hidden states from the desired layer.
+ prompt_embeds = prompt_embeds[-1][-(clip_skip + 1)]
+ # We also need to apply the final LayerNorm here to not mess with the
+ # representations. The `last_hidden_states` that we typically use for
+ # obtaining the final prompt representations passes through the LayerNorm
+ # layer.
+ prompt_embeds = self.text_encoder.text_model.final_layer_norm(prompt_embeds)
+ else:
+ max_len = max(len(text_input_ids[0]), len(uncond_input_ids[0]))
+ if len(text_input_ids[0]) < max_len:
+ text_input_ids = torch.cat(
+ [text_input_ids, torch.zeros(batch_size, max_len - len(text_input_ids[0]), dtype=torch.long)],
+ dim=1,
+ )
+ prompt_attention_mask = torch.cat(
+ [
+ prompt_attention_mask,
+ torch.zeros(
+ batch_size, max_len - len(prompt_attention_mask[0]), dtype=torch.long, device=device
+ ),
+ ],
+ dim=1,
+ )
+ elif len(uncond_input_ids[0]) < max_len:
+ uncond_input_ids = torch.cat(
+ [uncond_input_ids, torch.zeros(batch_size, max_len - len(uncond_input_ids[0]), dtype=torch.long)],
+ dim=1,
+ )
+ negative_prompt_attention_mask = torch.cat(
+ [
+ negative_prompt_attention_mask,
+ torch.zeros(
+ batch_size,
+ max_len - len(negative_prompt_attention_mask[0]),
+ dtype=torch.long,
+ device=device,
+ ),
+ ],
+ dim=1,
+ )
+ cfg_input_ids = torch.cat([uncond_input_ids, text_input_ids], dim=0)
+ cfg_attention_mask = torch.cat([negative_prompt_attention_mask, prompt_attention_mask], dim=0)
+ prompt_embeds = self.text_encoder(
+ cfg_input_ids.to(device),
+ attention_mask=cfg_attention_mask,
+ )
+ prompt_embeds = prompt_embeds[0]
+
+ prompt_embeds = prompt_embeds.to(dtype=prompt_embeds_dtype, device=device)
+
+ if self.text_encoder is not None:
+ if isinstance(self, StableDiffusionLoraLoaderMixin) and USE_PEFT_BACKEND:
+ # Retrieve the original scale by scaling back the LoRA layers
+ unscale_lora_layers(self.text_encoder, lora_scale)
+
+ if not do_classifier_free_guidance:
+ return prompt_embeds, None, prompt_attention_mask, None
+ return prompt_embeds[1], prompt_embeds[0], prompt_attention_mask, negative_prompt_attention_mask
+
+ def encode_image(self, image, device, num_images_per_prompt, output_hidden_states=None):
+ dtype = next(self.image_encoder.parameters()).dtype
+
+ if not isinstance(image, torch.Tensor):
+ image = self.feature_extractor(image, return_tensors="pt").pixel_values
+
+ image = image.to(device=device, dtype=dtype)
+ if output_hidden_states:
+ image_enc_hidden_states = self.image_encoder(image, output_hidden_states=True).hidden_states[-2]
+ image_enc_hidden_states = image_enc_hidden_states.repeat_interleave(num_images_per_prompt, dim=0)
+ uncond_image_enc_hidden_states = self.image_encoder(
+ torch.zeros_like(image), output_hidden_states=True
+ ).hidden_states[-2]
+ uncond_image_enc_hidden_states = uncond_image_enc_hidden_states.repeat_interleave(
+ num_images_per_prompt, dim=0
+ )
+ return image_enc_hidden_states, uncond_image_enc_hidden_states
+ else:
+ image_embeds = self.image_encoder(image).image_embeds
+ image_embeds = image_embeds.repeat_interleave(num_images_per_prompt, dim=0)
+ uncond_image_embeds = torch.zeros_like(image_embeds)
+
+ return image_embeds, uncond_image_embeds
+
+ def prepare_ip_adapter_image_embeds(
+ self, ip_adapter_image, ip_adapter_image_embeds, device, num_images_per_prompt, do_classifier_free_guidance
+ ):
+ image_embeds = []
+ if do_classifier_free_guidance:
+ negative_image_embeds = []
+ if ip_adapter_image_embeds is None:
+ if not isinstance(ip_adapter_image, list):
+ ip_adapter_image = [ip_adapter_image]
+
+ if len(ip_adapter_image) != len(self.unet.encoder_hid_proj.image_projection_layers):
+ raise ValueError(
+ f"`ip_adapter_image` must have same length as the number of IP Adapters. Got {len(ip_adapter_image)} images and {len(self.unet.encoder_hid_proj.image_projection_layers)} IP Adapters."
+ )
+
+ for single_ip_adapter_image, image_proj_layer in zip(
+ ip_adapter_image, self.unet.encoder_hid_proj.image_projection_layers
+ ):
+ output_hidden_state = not isinstance(image_proj_layer, ImageProjection)
+ single_image_embeds, single_negative_image_embeds = self.encode_image(
+ single_ip_adapter_image, device, 1, output_hidden_state
+ )
+
+ image_embeds.append(single_image_embeds[None, :])
+ if do_classifier_free_guidance:
+ negative_image_embeds.append(single_negative_image_embeds[None, :])
+ else:
+ for single_image_embeds in ip_adapter_image_embeds:
+ if do_classifier_free_guidance:
+ single_negative_image_embeds, single_image_embeds = single_image_embeds.chunk(2)
+ negative_image_embeds.append(single_negative_image_embeds)
+ image_embeds.append(single_image_embeds)
+
+ ip_adapter_image_embeds = []
+ for i, single_image_embeds in enumerate(image_embeds):
+ single_image_embeds = torch.cat([single_image_embeds] * num_images_per_prompt, dim=0)
+ if do_classifier_free_guidance:
+ single_negative_image_embeds = torch.cat([negative_image_embeds[i]] * num_images_per_prompt, dim=0)
+ single_image_embeds = torch.cat([single_negative_image_embeds, single_image_embeds], dim=0)
+
+ single_image_embeds = single_image_embeds.to(device=device)
+ ip_adapter_image_embeds.append(single_image_embeds)
+
+ return ip_adapter_image_embeds
+
+ def prepare_extra_step_kwargs(self, generator, eta):
+ # prepare extra kwargs for the scheduler step, since not all schedulers have the same signature
+ # eta (η) is only used with the DDIMScheduler, it will be ignored for other schedulers.
+ # eta corresponds to η in DDIM paper: https://arxiv.org/abs/2010.02502
+ # and should be between [0, 1]
+
+ accepts_eta = "eta" in set(inspect.signature(self.scheduler.step).parameters.keys())
+ extra_step_kwargs = {}
+ if accepts_eta:
+ extra_step_kwargs["eta"] = eta
+
+ # check if the scheduler accepts generator
+ accepts_generator = "generator" in set(inspect.signature(self.scheduler.step).parameters.keys())
+ if accepts_generator:
+ extra_step_kwargs["generator"] = generator
+ return extra_step_kwargs
+
+ def check_inputs(
+ self,
+ prompt,
+ height,
+ width,
+ callback_steps,
+ negative_prompt=None,
+ prompt_embeds=None,
+ negative_prompt_embeds=None,
+ ip_adapter_image=None,
+ ip_adapter_image_embeds=None,
+ callback_on_step_end_tensor_inputs=None,
+ ):
+ if height % 8 != 0 or width % 8 != 0:
+ raise ValueError(f"`height` and `width` have to be divisible by 8 but are {height} and {width}.")
+
+ if callback_steps is not None and (not isinstance(callback_steps, int) or callback_steps <= 0):
+ raise ValueError(
+ f"`callback_steps` has to be a positive integer but is {callback_steps} of type"
+ f" {type(callback_steps)}."
+ )
+ if callback_on_step_end_tensor_inputs is not None and not all(
+ k in self._callback_tensor_inputs for k in callback_on_step_end_tensor_inputs
+ ):
+ raise ValueError(
+ f"`callback_on_step_end_tensor_inputs` has to be in {self._callback_tensor_inputs}, but found {[k for k in callback_on_step_end_tensor_inputs if k not in self._callback_tensor_inputs]}"
+ )
+
+ if prompt is not None and prompt_embeds is not None:
+ raise ValueError(
+ f"Cannot forward both `prompt`: {prompt} and `prompt_embeds`: {prompt_embeds}. Please make sure to"
+ " only forward one of the two."
+ )
+ elif prompt is None and prompt_embeds is None:
+ raise ValueError(
+ "Provide either `prompt` or `prompt_embeds`. Cannot leave both `prompt` and `prompt_embeds` undefined."
+ )
+ elif prompt is not None and (not isinstance(prompt, str) and not isinstance(prompt, list)):
+ raise ValueError(f"`prompt` has to be of type `str` or `list` but is {type(prompt)}")
+
+ if negative_prompt is not None and negative_prompt_embeds is not None:
+ raise ValueError(
+ f"Cannot forward both `negative_prompt`: {negative_prompt} and `negative_prompt_embeds`:"
+ f" {negative_prompt_embeds}. Please make sure to only forward one of the two."
+ )
+
+ if prompt_embeds is not None and negative_prompt_embeds is not None:
+ if prompt_embeds.shape != negative_prompt_embeds.shape:
+ raise ValueError(
+ "`prompt_embeds` and `negative_prompt_embeds` must have the same shape when passed directly, but"
+ f" got: `prompt_embeds` {prompt_embeds.shape} != `negative_prompt_embeds`"
+ f" {negative_prompt_embeds.shape}."
+ )
+
+ if ip_adapter_image is not None and ip_adapter_image_embeds is not None:
+ raise ValueError(
+ "Provide either `ip_adapter_image` or `ip_adapter_image_embeds`. Cannot leave both `ip_adapter_image` and `ip_adapter_image_embeds` defined."
+ )
+
+ if ip_adapter_image_embeds is not None:
+ if not isinstance(ip_adapter_image_embeds, list):
+ raise ValueError(
+ f"`ip_adapter_image_embeds` has to be of type `list` but is {type(ip_adapter_image_embeds)}"
+ )
+ elif ip_adapter_image_embeds[0].ndim not in [3, 4]:
+ raise ValueError(
+ f"`ip_adapter_image_embeds` has to be a list of 3D or 4D tensors but is {ip_adapter_image_embeds[0].ndim}D"
+ )
+
+ def prepare_latents(
+ self, batch_size, num_channels_latents, height, width, dtype, device, generator, scales, latents=None
+ ):
+ shape = (
+ batch_size,
+ num_channels_latents,
+ int(height),
+ int(width),
+ )
+ if isinstance(generator, list) and len(generator) != batch_size:
+ raise ValueError(
+ f"You have passed a list of generators of length {len(generator)}, but requested an effective batch"
+ f" size of {batch_size}. Make sure the batch size matches the length of the generators."
+ )
+
+ if latents is None:
+ latents = randn_tensor(shape, generator=generator, device=device, dtype=dtype)
+ if scales is not None:
+ out = [latents]
+ for s in scales[1:]:
+ ratio = scales[0] // s
+ sample_low = F.avg_pool2d(latents, ratio) * ratio
+ sample_low = sample_low.normal_(generator=generator)
+ out += [sample_low]
+ latents = out
+ else:
+ if scales is not None:
+ latents = [latent.to(device=device) for latent in latents]
+ else:
+ latents = latents.to(device)
+
+ # scale the initial noise by the standard deviation required by the scheduler
+ if scales is not None:
+ latents = [latent * self.scheduler.init_noise_sigma for latent in latents]
+ else:
+ latents = latents * self.scheduler.init_noise_sigma
+ return latents
+
+ # Copied from diffusers.pipelines.latent_consistency_models.pipeline_latent_consistency_text2img.LatentConsistencyModelPipeline.get_guidance_scale_embedding
+ def get_guidance_scale_embedding(
+ self, w: torch.Tensor, embedding_dim: int = 512, dtype: torch.dtype = torch.float32
+ ) -> torch.Tensor:
+ """
+ See https://github.com/google-research/vdm/blob/dc27b98a554f65cdc654b800da5aa1846545d41b/model_vdm.py#L298
+
+ Args:
+ w (`torch.Tensor`):
+ Generate embedding vectors with a specified guidance scale to subsequently enrich timestep embeddings.
+ embedding_dim (`int`, *optional*, defaults to 512):
+ Dimension of the embeddings to generate.
+ dtype (`torch.dtype`, *optional*, defaults to `torch.float32`):
+ Data type of the generated embeddings.
+
+ Returns:
+ `torch.Tensor`: Embedding vectors with shape `(len(w), embedding_dim)`.
+ """
+ assert len(w.shape) == 1
+ w = w * 1000.0
+
+ half_dim = embedding_dim // 2
+ emb = torch.log(torch.tensor(10000.0)) / (half_dim - 1)
+ emb = torch.exp(torch.arange(half_dim, dtype=dtype) * -emb)
+ emb = w.to(dtype)[:, None] * emb[None, :]
+ emb = torch.cat([torch.sin(emb), torch.cos(emb)], dim=1)
+ if embedding_dim % 2 == 1: # zero pad
+ emb = torch.nn.functional.pad(emb, (0, 1))
+ assert emb.shape == (w.shape[0], embedding_dim)
+ return emb
+
+ @property
+ def guidance_scale(self):
+ return self._guidance_scale
+
+ @property
+ def guidance_rescale(self):
+ return self._guidance_rescale
+
+ @property
+ def clip_skip(self):
+ return self._clip_skip
+
+ # here `guidance_scale` is defined analog to the guidance weight `w` of equation (2)
+ # of the Imagen paper: https://arxiv.org/pdf/2205.11487.pdf . `guidance_scale = 1`
+ # corresponds to doing no classifier free guidance.
+ @property
+ def do_classifier_free_guidance(self):
+ return self._guidance_scale > 1 and self.unet.config.time_cond_proj_dim is None
+
+ @property
+ def cross_attention_kwargs(self):
+ return self._cross_attention_kwargs
+
+ @property
+ def num_timesteps(self):
+ return self._num_timesteps
+
+ @property
+ def interrupt(self):
+ return self._interrupt
+
+ @torch.no_grad()
+ @replace_example_docstring(EXAMPLE_DOC_STRING)
+ def __call__(
+ self,
+ prompt: Union[str, List[str]] = None,
+ height: Optional[int] = None,
+ width: Optional[int] = None,
+ num_inference_steps: int = 50,
+ timesteps: List[int] = None,
+ sigmas: List[float] = None,
+ guidance_scale: float = 7.5,
+ negative_prompt: Optional[Union[str, List[str]]] = None,
+ num_images_per_prompt: Optional[int] = 1,
+ eta: float = 0.0,
+ generator: Optional[Union[torch.Generator, List[torch.Generator]]] = None,
+ latents: Optional[torch.Tensor] = None,
+ prompt_embeds: Optional[torch.Tensor] = None,
+ negative_prompt_embeds: Optional[torch.Tensor] = None,
+ ip_adapter_image: Optional[PipelineImageInput] = None,
+ ip_adapter_image_embeds: Optional[List[torch.Tensor]] = None,
+ output_type: Optional[str] = "pil",
+ return_dict: bool = True,
+ cross_attention_kwargs: Optional[Dict[str, Any]] = None,
+ guidance_rescale: float = 0.0,
+ clip_skip: Optional[int] = None,
+ callback_on_step_end: Optional[
+ Union[Callable[[int, int, Dict], None], PipelineCallback, MultiPipelineCallbacks]
+ ] = None,
+ callback_on_step_end_tensor_inputs: List[str] = ["latents"],
+ **kwargs,
+ ):
+ r"""
+ The call function to the pipeline for generation.
+
+ Args:
+ prompt (`str` or `List[str]`, *optional*):
+ The prompt or prompts to guide image generation. If not defined, you need to pass `prompt_embeds`.
+ height (`int`, *optional*, defaults to `self.unet.config.sample_size`):
+ The height in pixels of the generated image.
+ width (`int`, *optional*, defaults to `self.unet.config.sample_size`):
+ The width in pixels of the generated image.
+ num_inference_steps (`int`, *optional*, defaults to 50):
+ The number of denoising steps. More denoising steps usually lead to a higher quality image at the
+ expense of slower inference.
+ timesteps (`List[int]`, *optional*):
+ Custom timesteps to use for the denoising process with schedulers which support a `timesteps` argument
+ in their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is
+ passed will be used. Must be in descending order.
+ sigmas (`List[float]`, *optional*):
+ Custom sigmas to use for the denoising process with schedulers which support a `sigmas` argument in
+ their `set_timesteps` method. If not defined, the default behavior when `num_inference_steps` is passed
+ will be used.
+ guidance_scale (`float`, *optional*, defaults to 7.5):
+ A higher guidance scale value encourages the model to generate images closely linked to the text
+ `prompt` at the expense of lower image quality. Guidance scale is enabled when `guidance_scale > 1`.
+ negative_prompt (`str` or `List[str]`, *optional*):
+ The prompt or prompts to guide what to not include in image generation. If not defined, you need to
+ pass `negative_prompt_embeds` instead. Ignored when not using guidance (`guidance_scale < 1`).
+ num_images_per_prompt (`int`, *optional*, defaults to 1):
+ The number of images to generate per prompt.
+ eta (`float`, *optional*, defaults to 0.0):
+ Corresponds to parameter eta (η) from the [DDIM](https://arxiv.org/abs/2010.02502) paper. Only applies
+ to the [`~schedulers.DDIMScheduler`], and is ignored in other schedulers.
+ generator (`torch.Generator` or `List[torch.Generator]`, *optional*):
+ A [`torch.Generator`](https://pytorch.org/docs/stable/generated/torch.Generator.html) to make
+ generation deterministic.
+ latents (`torch.Tensor`, *optional*):
+ Pre-generated noisy latents sampled from a Gaussian distribution, to be used as inputs for image
+ generation. Can be used to tweak the same generation with different prompts. If not provided, a latents
+ tensor is generated by sampling using the supplied random `generator`.
+ prompt_embeds (`torch.Tensor`, *optional*):
+ Pre-generated text embeddings. Can be used to easily tweak text inputs (prompt weighting). If not
+ provided, text embeddings are generated from the `prompt` input argument.
+ negative_prompt_embeds (`torch.Tensor`, *optional*):
+ Pre-generated negative text embeddings. Can be used to easily tweak text inputs (prompt weighting). If
+ not provided, `negative_prompt_embeds` are generated from the `negative_prompt` input argument.
+ ip_adapter_image: (`PipelineImageInput`, *optional*): Optional image input to work with IP Adapters.
+ ip_adapter_image_embeds (`List[torch.Tensor]`, *optional*):
+ Pre-generated image embeddings for IP-Adapter. It should be a list of length same as number of
+ IP-adapters. Each element should be a tensor of shape `(batch_size, num_images, emb_dim)`. It should
+ contain the negative image embedding if `do_classifier_free_guidance` is set to `True`. If not
+ provided, embeddings are computed from the `ip_adapter_image` input argument.
+ output_type (`str`, *optional*, defaults to `"pil"`):
+ The output format of the generated image. Choose between `PIL.Image` or `np.array`.
+ return_dict (`bool`, *optional*, defaults to `True`):
+ Whether or not to return a [`~pipelines.stable_diffusion.StableDiffusionPipelineOutput`] instead of a
+ plain tuple.
+ cross_attention_kwargs (`dict`, *optional*):
+ A kwargs dictionary that if specified is passed along to the [`AttentionProcessor`] as defined in
+ [`self.processor`](https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention_processor.py).
+ guidance_rescale (`float`, *optional*, defaults to 0.0):
+ Guidance rescale factor from [Common Diffusion Noise Schedules and Sample Steps are
+ Flawed](https://arxiv.org/pdf/2305.08891.pdf). Guidance rescale factor should fix overexposure when
+ using zero terminal SNR.
+ clip_skip (`int`, *optional*):
+ Number of layers to be skipped from CLIP while computing the prompt embeddings. A value of 1 means that
+ the output of the pre-final layer will be used for computing the prompt embeddings.
+ callback_on_step_end (`Callable`, `PipelineCallback`, `MultiPipelineCallbacks`, *optional*):
+ A function or a subclass of `PipelineCallback` or `MultiPipelineCallbacks` that is called at the end of
+ each denoising step during the inference. with the following arguments: `callback_on_step_end(self:
+ DiffusionPipeline, step: int, timestep: int, callback_kwargs: Dict)`. `callback_kwargs` will include a
+ list of all tensors as specified by `callback_on_step_end_tensor_inputs`.
+ callback_on_step_end_tensor_inputs (`List`, *optional*):
+ The list of tensor inputs for the `callback_on_step_end` function. The tensors specified in the list
+ will be passed as `callback_kwargs` argument. You will only be able to include variables listed in the
+ `._callback_tensor_inputs` attribute of your pipeline class.
+
+ Examples:
+
+ Returns:
+ [`~MatryoshkaPipelineOutput`] or `tuple`:
+ If `return_dict` is `True`, [`~MatryoshkaPipelineOutput`] is returned,
+ otherwise a `tuple` is returned where the first element is a list with the generated images and the
+ second element is a list of `bool`s indicating whether the corresponding generated image contains
+ "not-safe-for-work" (nsfw) content.
+ """
+
+ callback = kwargs.pop("callback", None)
+ callback_steps = kwargs.pop("callback_steps", None)
+
+ if callback is not None:
+ deprecate(
+ "callback",
+ "1.0.0",
+ "Passing `callback` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`",
+ )
+ if callback_steps is not None:
+ deprecate(
+ "callback_steps",
+ "1.0.0",
+ "Passing `callback_steps` as an input argument to `__call__` is deprecated, consider using `callback_on_step_end`",
+ )
+
+ if isinstance(callback_on_step_end, (PipelineCallback, MultiPipelineCallbacks)):
+ callback_on_step_end_tensor_inputs = callback_on_step_end.tensor_inputs
+
+ # 0. Default height and width to unet
+ height = height or self.unet.config.sample_size
+ width = width or self.unet.config.sample_size
+ # to deal with lora scaling and other possible forward hooks
+
+ # 1. Check inputs. Raise error if not correct
+ self.check_inputs(
+ prompt,
+ height,
+ width,
+ callback_steps,
+ negative_prompt,
+ prompt_embeds,
+ negative_prompt_embeds,
+ ip_adapter_image,
+ ip_adapter_image_embeds,
+ callback_on_step_end_tensor_inputs,
+ )
+
+ self._guidance_scale = guidance_scale
+ self._guidance_rescale = guidance_rescale
+ self._clip_skip = clip_skip
+ self._cross_attention_kwargs = cross_attention_kwargs
+ self._interrupt = False
+
+ # 2. Define call parameters
+ if prompt is not None and isinstance(prompt, str):
+ batch_size = 1
+ elif prompt is not None and isinstance(prompt, list):
+ batch_size = len(prompt)
+ else:
+ batch_size = prompt_embeds.shape[0]
+
+ device = self._execution_device
+
+ # 3. Encode input prompt
+ lora_scale = (
+ self.cross_attention_kwargs.get("scale", None) if self.cross_attention_kwargs is not None else None
+ )
+
+ (
+ prompt_embeds,
+ negative_prompt_embeds,
+ prompt_attention_mask,
+ negative_prompt_attention_mask,
+ ) = self.encode_prompt(
+ prompt,
+ device,
+ num_images_per_prompt,
+ self.do_classifier_free_guidance,
+ negative_prompt,
+ prompt_embeds=prompt_embeds,
+ negative_prompt_embeds=negative_prompt_embeds,
+ lora_scale=lora_scale,
+ clip_skip=self.clip_skip,
+ )
+
+ # For classifier free guidance, we need to do two forward passes.
+ # Here we concatenate the unconditional and text embeddings into a single batch
+ # to avoid doing two forward passes
+ if self.do_classifier_free_guidance:
+ prompt_embeds = torch.cat([negative_prompt_embeds.unsqueeze(0), prompt_embeds.unsqueeze(0)])
+ attention_masks = torch.cat([negative_prompt_attention_mask, prompt_attention_mask])
+ else:
+ attention_masks = prompt_attention_mask
+
+ prompt_embeds = prompt_embeds * attention_masks.unsqueeze(-1)
+
+ if ip_adapter_image is not None or ip_adapter_image_embeds is not None:
+ image_embeds = self.prepare_ip_adapter_image_embeds(
+ ip_adapter_image,
+ ip_adapter_image_embeds,
+ device,
+ batch_size * num_images_per_prompt,
+ self.do_classifier_free_guidance,
+ )
+
+ # 4. Prepare timesteps
+ timesteps, num_inference_steps = retrieve_timesteps(
+ self.scheduler, num_inference_steps, device, timesteps, sigmas
+ )
+ timesteps = timesteps[:-1]
+
+ # 5. Prepare latent variables
+ num_channels_latents = self.unet.config.in_channels
+ latents = self.prepare_latents(
+ batch_size * num_images_per_prompt,
+ num_channels_latents,
+ height,
+ width,
+ prompt_embeds.dtype,
+ device,
+ generator,
+ self.scheduler.scales,
+ latents,
+ )
+
+ # 6. Prepare extra step kwargs. TODO: Logic should ideally just be moved out of the pipeline
+ extra_step_kwargs = self.prepare_extra_step_kwargs(generator, eta)
+ extra_step_kwargs |= {"use_clipped_model_output": True}
+
+ # 6.1 Add image embeds for IP-Adapter
+ added_cond_kwargs = (
+ {"image_embeds": image_embeds}
+ if (ip_adapter_image is not None or ip_adapter_image_embeds is not None)
+ else None
+ )
+
+ # 6.2 Optionally get Guidance Scale Embedding
+ timestep_cond = None
+ if self.unet.config.time_cond_proj_dim is not None:
+ guidance_scale_tensor = torch.tensor(self.guidance_scale - 1).repeat(batch_size * num_images_per_prompt)
+ timestep_cond = self.get_guidance_scale_embedding(
+ guidance_scale_tensor, embedding_dim=self.unet.config.time_cond_proj_dim
+ ).to(device=device, dtype=latents.dtype)
+
+ # 7. Denoising loop
+ num_warmup_steps = len(timesteps) - num_inference_steps * self.scheduler.order
+ self._num_timesteps = len(timesteps)
+ with self.progress_bar(total=num_inference_steps) as progress_bar:
+ for i, t in enumerate(timesteps):
+ if self.interrupt:
+ continue
+
+ # expand the latents if we are doing classifier free guidance
+ if self.do_classifier_free_guidance and isinstance(latents, list):
+ latent_model_input = [latent.repeat(2, 1, 1, 1) for latent in latents]
+ elif self.do_classifier_free_guidance:
+ latent_model_input = latents.repeat(2, 1, 1, 1)
+ else:
+ latent_model_input = latents
+ latent_model_input = self.scheduler.scale_model_input(latent_model_input, t)
+
+ # predict the noise residual
+ noise_pred = self.unet(
+ latent_model_input,
+ t - 1,
+ encoder_hidden_states=prompt_embeds,
+ timestep_cond=timestep_cond,
+ cross_attention_kwargs=self.cross_attention_kwargs,
+ added_cond_kwargs=added_cond_kwargs,
+ encoder_attention_mask=attention_masks,
+ return_dict=False,
+ )[0]
+
+ # perform guidance
+ if isinstance(noise_pred, list) and self.do_classifier_free_guidance:
+ for i, (noise_pred_uncond, noise_pred_text) in enumerate(noise_pred):
+ noise_pred[i] = noise_pred_uncond + self.guidance_scale * (noise_pred_text - noise_pred_uncond)
+ elif self.do_classifier_free_guidance:
+ noise_pred_uncond, noise_pred_text = noise_pred.chunk(2)
+ noise_pred = noise_pred_uncond + self.guidance_scale * (noise_pred_text - noise_pred_uncond)
+
+ if self.do_classifier_free_guidance and self.guidance_rescale > 0.0:
+ # Based on 3.4. in https://arxiv.org/pdf/2305.08891.pdf
+ noise_pred = rescale_noise_cfg(noise_pred, noise_pred_text, guidance_rescale=self.guidance_rescale)
+
+ # compute the previous noisy sample x_t -> x_t-1
+ latents = self.scheduler.step(noise_pred, t, latents, **extra_step_kwargs, return_dict=False)[0]
+
+ if callback_on_step_end is not None:
+ callback_kwargs = {}
+ for k in callback_on_step_end_tensor_inputs:
+ callback_kwargs[k] = locals()[k]
+ callback_outputs = callback_on_step_end(self, i, t, callback_kwargs)
+
+ latents = callback_outputs.pop("latents", latents)
+ prompt_embeds = callback_outputs.pop("prompt_embeds", prompt_embeds)
+ negative_prompt_embeds = callback_outputs.pop("negative_prompt_embeds", negative_prompt_embeds)
+
+ # call the callback, if provided
+ if i == len(timesteps) - 1 or ((i + 1) > num_warmup_steps and (i + 1) % self.scheduler.order == 0):
+ progress_bar.update()
+ if callback is not None and i % callback_steps == 0:
+ step_idx = i // getattr(self.scheduler, "order", 1)
+ callback(step_idx, t, latents)
+
+ if XLA_AVAILABLE:
+ xm.mark_step()
+
+ image = latents
+
+ if self.scheduler.scales is not None:
+ for i, (img, scale) in enumerate(zip(image, self.scheduler.scales)):
+ image[i] = self.image_processor.postprocess(img * scale, output_type=output_type)[0]
+ else:
+ image = self.image_processor.postprocess(image, output_type=output_type)
+
+ # Offload all models
+ self.maybe_free_model_hooks()
+
+ if not return_dict:
+ return (image,)
+
+ return MatryoshkaPipelineOutput(images=image)