mirror of
https://github.com/finegrain-ai/refiners.git
synced 2024-09-19 19:05:28 +00:00
108 lines
7 KiB
Markdown
108 lines
7 KiB
Markdown
<div align="center">
|
|
|
|
<picture>
|
|
<source media="(prefers-color-scheme: dark)" srcset="https://raw.githubusercontent.com/finegrain-ai/refiners/main/assets/logo_dark.png">
|
|
<source media="(prefers-color-scheme: light)" srcset="https://raw.githubusercontent.com/finegrain-ai/refiners/main/assets/logo_light.png">
|
|
<img alt="Finegrain Refiners Library" width="352" height="128" style="max-width: 100%;">
|
|
</picture>
|
|
|
|
**The simplest way to train and run adapters on top of foundation models**
|
|
|
|
[**Manifesto**](https://refine.rs/home/why/) |
|
|
[**Docs**](https://refine.rs) |
|
|
[**Guides**](https://refine.rs/guides/adapting_sdxl/) |
|
|
[**Discussions**](https://github.com/finegrain-ai/refiners/discussions) |
|
|
[**Discord**](https://discord.gg/mCmjNUVV7d)
|
|
|
|
______________________________________________________________________
|
|
|
|
[](https://github.com/astral-sh/rye)
|
|
[](https://github.com/astral-sh/ruff)
|
|
[](https://github.com/pypa/hatch)
|
|
[](https://pypi.org/project/refiners/)
|
|
[](https://pypi.org/project/refiners/)
|
|
[](/LICENSE) \
|
|
[](https://finegrain.ai/bounties)
|
|
[](https://discord.gg/mCmjNUVV7d)
|
|
[](https://huggingface.co/refiners)
|
|
[](https://huggingface.co/finegrain)
|
|
[](https://registry.comfy.org/publishers/finegrain/nodes/comfyui-refiners)
|
|
|
|
</div>
|
|
|
|
## Latest News 🔥
|
|
|
|
- Added [ELLA](https://arxiv.org/abs/2403.05135) for better prompts handling (contributed by [@ily-R](https://github.com/ily-R))
|
|
- Added the Box Segmenter all-in-one solution ([model](https://huggingface.co/finegrain/finegrain-box-segmenter), [HF Space](https://huggingface.co/spaces/finegrain/finegrain-object-cutter))
|
|
- Added [MVANet](https://arxiv.org/abs/2404.07445) for high resolution segmentation
|
|
- Added [IC-Light](https://github.com/lllyasviel/IC-Light) to manipulate the illumination of images
|
|
- Added Multi Upscaler for high-resolution image generation, inspired from [Clarity Upscaler](https://github.com/philz1337x/clarity-upscaler) ([HF Space](https://huggingface.co/spaces/finegrain/enhancer))
|
|
- Added [HQ-SAM](https://arxiv.org/abs/2306.01567) for high quality mask prediction with Segment Anything
|
|
- Added [SDXL-Lightning](https://arxiv.org/abs/2402.13929)
|
|
- Added [Latent Consistency Models](https://arxiv.org/abs/2310.04378) and [LCM-LoRA](https://arxiv.org/abs/2311.05556) for Stable Diffusion XL
|
|
- Added [Style Aligned adapter](https://arxiv.org/abs/2312.02133) to Stable Diffusion models
|
|
- Added [ControlLoRA (v2) adapter](https://github.com/HighCWu/control-lora-v2) to Stable Diffusion XL
|
|
- Added [Euler's method](https://arxiv.org/abs/2206.00364) to solvers (contributed by [@israfelsr](https://github.com/israfelsr))
|
|
- Added [DINOv2](https://github.com/facebookresearch/dinov2) for high-performance visual features (contributed by [@Laurent2916](https://github.com/Laurent2916))
|
|
- Added [FreeU](https://github.com/ChenyangSi/FreeU) for improved quality at no cost (contributed by [@isamu-isozaki](https://github.com/isamu-isozaki))
|
|
- Added [Restart Sampling](https://github.com/Newbeeer/diffusion_restart_sampling) for improved image generation ([example](https://github.com/Newbeeer/diffusion_restart_sampling/issues/4))
|
|
- Added [Self-Attention Guidance](https://github.com/KU-CVLAB/Self-Attention-Guidance/) to avoid e.g. too smooth images ([example](https://github.com/SusungHong/Self-Attention-Guidance/issues/4))
|
|
- Added [T2I-Adapter](https://github.com/TencentARC/T2I-Adapter) for extra guidance ([example](https://github.com/TencentARC/T2I-Adapter/discussions/93))
|
|
- Added [MultiDiffusion](https://github.com/omerbt/MultiDiffusion) for e.g. panorama images
|
|
- Added [IP-Adapter](https://github.com/tencent-ailab/IP-Adapter), aka image prompt ([example](https://github.com/tencent-ailab/IP-Adapter/issues/92))
|
|
- Added [Segment Anything](https://github.com/facebookresearch/segment-anything) to foundation models
|
|
- Added [SDXL 1.0](https://github.com/Stability-AI/generative-models) to foundation models
|
|
- Made possible to add new concepts to the CLIP text encoder, e.g. via [Textual Inversion](https://arxiv.org/abs/2208.01618)
|
|
|
|
## Installation
|
|
|
|
The current recommended way to install Refiners is from source using [Rye](https://rye-up.com/):
|
|
|
|
```bash
|
|
git clone "git@github.com:finegrain-ai/refiners.git"
|
|
cd refiners
|
|
rye sync --all-features
|
|
```
|
|
|
|
## Documentation
|
|
|
|
Refiners comes with a MkDocs-based documentation website available at https://refine.rs. You will find there a [quick start guide](https://refine.rs/getting-started/recommended/), a description of the [key concepts](https://refine.rs/concepts/chain/), as well as in-depth foundation model adaptation [guides](https://refine.rs/guides/adapting_sdxl/).
|
|
|
|
## Awesome Adaptation Papers
|
|
|
|
If you're interested in understanding the diversity of use cases for foundation model adaptation (potentially beyond the specific adapters supported by Refiners), we suggest you take a look at these outstanding papers:
|
|
|
|
- [ControlNet](https://arxiv.org/abs/2302.05543)
|
|
- [T2I-Adapter](https://arxiv.org/abs/2302.08453)
|
|
- [IP-Adapter](https://arxiv.org/abs/2308.06721)
|
|
- [Medical SAM Adapter](https://arxiv.org/abs/2304.12620)
|
|
- [3DSAM-adapter](https://arxiv.org/abs/2306.13465)
|
|
- [SAM-adapter](https://arxiv.org/abs/2304.09148)
|
|
- [Cross Modality Attention Adapter](https://arxiv.org/abs/2307.01124)
|
|
- [UniAdapter](https://arxiv.org/abs/2302.06605)
|
|
|
|
## Projects using Refiners
|
|
|
|
- https://github.com/brycedrennan/imaginAIry
|
|
|
|
## Credits
|
|
|
|
We took inspiration from these great projects:
|
|
|
|
- [tinygrad](https://github.com/tinygrad/tinygrad) - For something between PyTorch and [karpathy/micrograd](https://github.com/karpathy/micrograd)
|
|
- [Composer](https://github.com/mosaicml/composer) - A PyTorch Library for Efficient Neural Network Training
|
|
- [Keras](https://github.com/keras-team/keras) - Deep Learning for humans
|
|
|
|
## Citation
|
|
|
|
```bibtex
|
|
@misc{the-finegrain-team-2023-refiners,
|
|
author = {Benjamin Trom and Pierre Chapuis and Cédric Deltheil},
|
|
title = {Refiners: The simplest way to train and run adapters on top of foundation models},
|
|
year = {2023},
|
|
publisher = {GitHub},
|
|
journal = {GitHub repository},
|
|
howpublished = {\url{https://github.com/finegrain-ai/refiners}}
|
|
}
|
|
```
|