🤗 Diffusers: State-of-the-art diffusion models for image and audio generation in PyTorch
Go to file
Pedro Cuenca bfe37f3159
Reproducible images by supplying latents to pipeline (#247)
* Accept latents as input for StableDiffusionPipeline.

* Notebook to demonstrate reusable seeds (latents).

* More accurate type annotation

Co-authored-by: Suraj Patil <surajp815@gmail.com>

* Review comments: move to device, raise instead of assert.

* Actually commit the test notebook.

I had mistakenly pushed an empty file instead.

* Adapt notebook to Colab.

* Update examples readme.

* Move notebook to personal repo.

Co-authored-by: Suraj Patil <surajp815@gmail.com>
2022-08-25 19:17:05 +05:30
.github added test workflow and fixed failing test (#237) 2022-08-24 13:46:53 +02:00
docs/source README improvements: credits and roadmap (#116) 2022-07-21 10:06:16 +02:00
examples Reproducible images by supplying latents to pipeline (#247) 2022-08-25 19:17:05 +05:30
scripts Style the `scripts` directory (#250) 2022-08-25 15:46:09 +02:00
src/diffusers Reproducible images by supplying latents to pipeline (#247) 2022-08-25 19:17:05 +05:30
tests added test workflow and fixed failing test (#237) 2022-08-24 13:46:53 +02:00
utils Add `is_torch_available`, `is_flax_available` (#204) 2022-08-17 16:47:20 +02:00
.gitignore upload some cleaning tools 2022-05-31 10:17:19 +02:00
LICENSE init upload 2022-05-30 18:21:15 +02:00
MANIFEST.in Fix manifest to include model card (#136) 2022-07-24 16:18:59 +02:00
Makefile Style the `scripts` directory (#250) 2022-08-25 15:46:09 +02:00
README.md Add image2image example script. (#231) 2022-08-23 16:27:28 +05:30
pyproject.toml init upload 2022-05-30 18:21:15 +02:00
setup.cfg init upload 2022-05-30 18:21:15 +02:00
setup.py Pin black==22.3 to keep a stable --preview flag (#249) 2022-08-25 15:19:59 +02:00

README.md



GitHub GitHub release Contributor Covenant

🤗 Diffusers provides pretrained diffusion models across multiple modalities, such as vision and audio, and serves as a modular toolbox for inference and training of diffusion models.

More precisely, 🤗 Diffusers offers:

  • State-of-the-art diffusion pipelines that can be run in inference with just a couple of lines of code (see src/diffusers/pipelines).
  • Various noise schedulers that can be used interchangeably for the prefered speed vs. quality trade-off in inference (see src/diffusers/schedulers).
  • Multiple types of models, such as UNet, can be used as building blocks in an end-to-end diffusion system (see src/diffusers/models).
  • Training examples to show how to train the most popular diffusion models (see examples/training).
  • Inference examples to show how to create custom pipelines for advanced tasks such as image2image, in-painting (see examples/inference)

Quickstart

In order to get started, we recommend taking a look at two notebooks:

  • The Getting started with Diffusers Open In Colab notebook, which showcases an end-to-end example of usage for diffusion models, schedulers and pipelines. Take a look at this notebook to learn how to use the pipeline abstraction, which takes care of everything (model, scheduler, noise handling) for you, and also to understand each independent building block in the library.
  • The Training a diffusers model Open In Colab notebook summarizes diffuser model training methods. This notebook takes a step-by-step approach to training your diffuser model on an image dataset, with explanatory graphics.

New 🎨🎨🎨 Stable Diffusion is now fully compatible with diffusers!

Stable Diffusion is a text-to-image latent diffusion model created by the researchers and engineers from CompVis, Stability AI and LAION. It's trained on 512x512 images from a subset of the LAION-5B database. This model uses a frozen CLIP ViT-L/14 text encoder to condition the model on text prompts. With its 860M UNet and 123M text encoder, the model is relatively lightweight and runs on a GPU with at least 10GB VRAM. See the model card for more information.

You need to accept the model license before downloading or using the Stable Diffusion weights. Please, visit the model card, read the license and tick the checkbox if you agree. You have to be a registered user in 🤗 Hugging Face Hub, and you'll also need to use an access token for the code to work. For more information on access tokens, please refer to this section of the documentation.

# make sure you're logged in with `huggingface-cli login`
from torch import autocast
from diffusers import StableDiffusionPipeline, LMSDiscreteScheduler

lms = LMSDiscreteScheduler(
    beta_start=0.00085, 
    beta_end=0.012, 
    beta_schedule="scaled_linear"
)

pipe = StableDiffusionPipeline.from_pretrained(
    "CompVis/stable-diffusion-v1-3", 
    scheduler=lms,
    use_auth_token=True
).to("cuda")

prompt = "a photo of an astronaut riding a horse on mars"
with autocast("cuda"):
    image = pipe(prompt)["sample"][0]  
    
image.save("astronaut_rides_horse.png")

For more details, check out the Stable Diffusion notebook Open In Colab and have a look into the release notes.

Examples

If you want to run the code yourself 💻, you can try out:

# !pip install diffusers transformers
from diffusers import DiffusionPipeline

model_id = "CompVis/ldm-text2im-large-256"

# load model and scheduler
ldm = DiffusionPipeline.from_pretrained(model_id)

# run pipeline in inference (sample random noise and denoise)
prompt = "A painting of a squirrel eating a burger"
images = ldm([prompt], num_inference_steps=50, eta=0.3, guidance_scale=6)["sample"]

# save images
for idx, image in enumerate(images):
    image.save(f"squirrel-{idx}.png")
# !pip install diffusers
from diffusers import DDPMPipeline, DDIMPipeline, PNDMPipeline

model_id = "google/ddpm-celebahq-256"

# load model and scheduler
ddpm = DDPMPipeline.from_pretrained(model_id)  # you can replace DDPMPipeline with DDIMPipeline or PNDMPipeline for faster inference

# run pipeline in inference (sample random noise and denoise)
image = ddpm()["sample"]

# save image
image[0].save("ddpm_generated_image.png")

If you just want to play around with some web demos, you can try out the following 🚀 Spaces:

Model Hugging Face Spaces
Text-to-Image Latent Diffusion Hugging Face Spaces
Faces generator Hugging Face Spaces
DDPM with different schedulers Hugging Face Spaces

Definitions

Models: Neural network that models p_\theta(\mathbf{x}_{t-1}|\mathbf{x}_t) (see image below) and is trained end-to-end to denoise a noisy input to an image. Examples: UNet, Conditioned UNet, 3D UNet, Transformer UNet


Figure from DDPM paper (https://arxiv.org/abs/2006.11239).

Schedulers: Algorithm class for both inference and training. The class provides functionality to compute previous image according to alpha, beta schedule as well as predict noise for training. Examples: DDPM, DDIM, PNDM, DEIS


Sampling and training algorithms. Figure from DDPM paper (https://arxiv.org/abs/2006.11239).

Diffusion Pipeline: End-to-end pipeline that includes multiple diffusion models, possible text encoders, ... Examples: Glide, Latent-Diffusion, Imagen, DALL-E 2


Figure from ImageGen (https://imagen.research.google/).

Philosophy

  • Readability and clarity is prefered over highly optimized code. A strong importance is put on providing readable, intuitive and elementary code design. E.g., the provided schedulers are separated from the provided models and provide well-commented code that can be read alongside the original paper.
  • Diffusers is modality independent and focuses on providing pretrained models and tools to build systems that generate continous outputs, e.g. vision and audio.
  • Diffusion models and schedulers are provided as concise, elementary building blocks. In contrast, diffusion pipelines are a collection of end-to-end diffusion systems that can be used out-of-the-box, should stay as close as possible to their original implementation and can include components of another library, such as text-encoders. Examples for diffusion pipelines are Glide and Latent Diffusion.

Installation

With pip

pip install --upgrade diffusers  # should install diffusers 0.2.4

With conda

conda install -c conda-forge diffusers

In the works

For the first release, 🤗 Diffusers focuses on text-to-image diffusion techniques. However, diffusers can be used for much more than that! Over the upcoming releases, we'll be focusing on:

A few pipeline components are already being worked on, namely:

  • BDDMPipeline for spectrogram-to-sound vocoding
  • GLIDEPipeline to support OpenAI's GLIDE model
  • Grad-TTS for text to audio generation / conditional audio generation

We want diffusers to be a toolbox useful for diffusers models in general; if you find yourself limited in any way by the current API, or would like to see additional models, schedulers, or techniques, please open a GitHub issue mentioning what you would like to see.

Credits

This library concretizes previous work by many different authors and would not have been possible without their great research and implementations. We'd like to thank, in particular, the following implementations which have helped us in our development and without which the API could not have been as polished today:

  • @CompVis' latent diffusion models library, available here
  • @hojonathanho original DDPM implementation, available here as well as the extremely useful translation into PyTorch by @pesser, available here
  • @ermongroup's DDIM implementation, available here.
  • @yang-song's Score-VE and Score-VP implementations, available here

We also want to thank @heejkoo for the very helpful overview of papers, code and resources on diffusion models, available here as well as @crowsonkb and @rromb for useful discussions and insights.