ZijingWu381 / ERDiff

PyTorch implementation of "Extraction and recovery of spatio-temporal structure in latent dynamics alignment with diffusion model"

Extraction and recovery of spatio-temporal structure in latent dynamics alignment with diffusion model (ERDiff)

Setup

To install the required dependancies using conda, run:

$ conda create --name erdiff --file requirements.txt

To install the required dependancies using Python virtual environment, run:

$ python3 -m venv erdiff
$ source erdiff/bin/activate
$ python3 -m pip install --upgrade pip
$ python3 -m pip install -e .

Train & Align

1. Source Domain: Cooperative Training

$ python3 VAE_Diffusion_CoTrain.py

2. Target Domain: Maximum Likelihood Alignment

$ python3 MLA.py

Visualization

Citation

arxiv

@article{wang2023extraction,
  title={Extraction and Recovery of Spatio-Temporal Structure in Latent Dynamics Alignment with Diffusion Model},
  author={Wang, Yule and Wu, Zijing and Li, Chengrui and Wu, Anqi},
  journal={arXiv preprint arXiv:2306.06138},
  year={2023}
}