This is an unofficial PyTorch implementation of the paper SODA: Bottleneck Diffusion Models for Representation Learning by Drew A. Hudson et al (2023). Built based on the DDPM source code.
Before running the model, install the required dependencies.
pip install -r requirements.txtWe have only implemented with the Cifar100 dataset; MIT-States and Imagenet1k are not yet ready to use.
python main.py --dataset cifar@article{hudson2023soda,
title={SODA: Bottleneck Diffusion Models for Representation Learning},
author={Hudson, Drew A and Zoran, Daniel and Malinowski, Mateusz and Lampinen, Andrew K and Jaegle, Andrew and McClelland, James L and Matthey, Loic and Hill, Felix and Lerchner, Alexander},
journal={arXiv preprint arXiv:2311.17901},
year={2023}
}