AnimateDiff Local Fine Tuning

This repo is only for fine tuning motion modules, locally. It is based on the work of Tumurzakov, and with much help from Cubey and many others.

Results "may" vary, use at your own discretion, offer void where prohibited. No gaurantees or promises.

HEAVILY WIP

Usage:

linux see end

Requires 3090/4090 for now, maybe...

git clone https://github.com/B34STW4RS/AD-Evo-Tuner/
cd AD-Evo-Tuner
conda env create -f environment.yaml
conda activate adt

place motion modules in models/motion_module/

run the download_SD1.5_prereq or EZ_Facehugger.py while the virtual environment is active. place runwayml sd1.5 files in models/stablediffusion/

unpacked don't need safetensors etc.

make a dataset similar to the default example set, including a populated caption.txt and validate.txt*

validate currently mostly non-functional wip sample dataset is a baseline of what you could probably train report issues with terminal outputs there are some nags you can ignore don't worry about them for now

run adt-user.bat
name your project
select the motion module you wish to fine tune
select your dataset folder
hit start
????
profit

models will be output to models/motion_module/{project_name+time}

config will be saved to configs/training/{project_name+time.yaml}

To Do:

rewrite gui
implement more viable parameters
fix validation
improve documentation
post sample models to civit.ai
post sample gifs from finetuned motion modules
data preprocessor
purple monkey dishwasher

On Linux // Temporary solution

After solving the environment and activating it

pip uninstall xformers
pip install xformers==0.0.20
pip install triton==2.0.0

#Advanced Advanced users can forgo using the gui entirely, and directly copy the default.yaml in the training directory and modify it as they wish.

run python train.py --config pathto/config.yaml

AnimateDiff

This repository is the official implementation of AnimateDiff.

AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning
Yuwei Guo, Ceyuan Yang*, Anyi Rao, Yaohui Wang, Yu Qiao, Dahua Lin, Bo Dai

*Corresponding Author

Arxiv Report | Project Page

Download Base T2I & Motion Module Checkpoints

We provide two versions of our Motion Module, which are trained on stable-diffusion-v1-4 and finetuned on v1-5 seperately. It's recommanded to try both of them for best results.

git lfs install
git clone https://huggingface.co/runwayml/stable-diffusion-v1-5 models/StableDiffusion/

bash download_bashscripts/0-MotionModule.sh

You may also directly download the motion module checkpoints from Google Drive, then put them in models/Motion_Module/ folder.

Prepare Personalize T2I

Here we provide inference configs for 6 demo T2I on CivitAI. You may run the following bash scripts to download these checkpoints.

bash download_bashscripts/1-ToonYou.sh
bash download_bashscripts/2-Lyriel.sh
bash download_bashscripts/3-RcnzCartoon.sh
bash download_bashscripts/4-MajicMix.sh
bash download_bashscripts/5-RealisticVision.sh
bash download_bashscripts/6-Tusun.sh
bash download_bashscripts/7-FilmVelvia.sh
bash download_bashscripts/8-GhibliBackground.sh

Inference

After downloading the above peronalized T2I checkpoints, run the following commands to generate animations. The results will automatically be saved to samples/ folder.

python -m scripts.animate --config configs/prompts/1-ToonYou.yaml
python -m scripts.animate --config configs/prompts/2-Lyriel.yaml
python -m scripts.animate --config configs/prompts/3-RcnzCartoon.yaml
python -m scripts.animate --config configs/prompts/4-MajicMix.yaml
python -m scripts.animate --config configs/prompts/5-RealisticVision.yaml
python -m scripts.animate --config configs/prompts/6-Tusun.yaml
python -m scripts.animate --config configs/prompts/7-FilmVelvia.yaml
python -m scripts.animate --config configs/prompts/8-GhibliBackground.yaml

To generate animations with a new DreamBooth/LoRA model, you may create a new config .yaml file in the following format:

NewModel:
  path: "[path to your DreamBooth/LoRA model .safetensors file]"
  base: "[path to LoRA base model .safetensors file, leave it empty string if not needed]"

  motion_module:
    - "models/Motion_Module/mm_sd_v14.ckpt"
    - "models/Motion_Module/mm_sd_v15.ckpt"
    
  steps:          25
  guidance_scale: 7.5

  prompt:
    - "[positive prompt]"

  n_prompt:
    - "[negative prompt]"

Then run the following commands:

python -m scripts.animate --config [path to the config file]

@misc{guo2023animatediff, title={AnimateDiff: Animate Your Personalized Text-to-Image Diffusion Models without Specific Tuning}, author={Yuwei Guo, Ceyuan Yang, Anyi Rao, Yaohui Wang, Yu Qiao, Dahua Lin, Bo Dai}, year={2023}, eprint={2307.04725}, archivePrefix={arXiv}, primaryClass={cs.CV} }

Contact Us

Yuwei Guo: guoyuwei@pjlab.org.cn
Ceyuan Yang: yangceyuan@pjlab.org.cn
Bo Dai: daibo@pjlab.org.cn

Acknowledgements

Codebase built upon Tune-a-Video.

artmamedov/AD-Evo-Tuner