https://github.com/laknath1996/gen-models

Course project for EN.553.741 Machine Learning II at JHU. Implements Wasserstein GAN, variational autoencoder and denoising diffusion probabilistic models.
https://github.com/laknath1996/gen-models

diffusion-models generative-adversarial-network generative-model

Last synced: about 1 month ago
JSON representation

Course project for EN.553.741 Machine Learning II at JHU. Implements Wasserstein GAN, variational autoencoder and denoising diffusion probabilistic models.

• Host: GitHub
• URL: https://github.com/laknath1996/gen-models
• Owner: Laknath1996
• Created: 2024-03-25T03:31:27.000Z (11 months ago)
• Default Branch: main
• Last Pushed: 2024-05-14T03:02:26.000Z (9 months ago)
• Last Synced: 2024-05-15T12:20:13.786Z (9 months ago)
• Topics: diffusion-models, generative-adversarial-network, generative-model
• Language: Jupyter Notebook
• Homepage:
• Size: 8.28 MB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Description 

Course project for EN.553.741 Machine Learning II at JHU. This repo contains code for implementing Wasserstein GAN, variational autoencoder and denoising diffusion probabilistic models from scratch.

## Training metrics

* VAE (Gaussian encoder + Bernoulli decoder) [1] trained on MNIST: [[logs]](https://wandb.ai/ashwin1996/vae/runs/ye2lf3sr?nw=nwuserashwin1996)

* WGAN (with weight clipping) [2] trained on MNIST: [[logs]](https://wandb.ai/ashwin1996/wgan/runs/kfa382kb?nw=nwuserashwin1996)

* DDPM [3] trained on MNIST: [[logs]](https://wandb.ai/ashwin1996/ddpm/runs/c362xa13?nw=nwuserashwin1996)

## Report & Slides

* [[Report]](https://github.com/Laknath1996/gen-models/blob/main/assets/report.pdf)

## References

[1] Kingma, Diederik P., and Max Welling. "Auto-encoding variational bayes." arXiv preprint arXiv:1312.6114 (2013).

[2] Arjovsky, Martin, Soumith Chintala, and Léon Bottou. "Wasserstein generative adversarial networks." International conference on machine learning. PMLR, 2017.

[3] Ho, Jonathan, Ajay Jain, and Pieter Abbeel. "Denoising diffusion probabilistic models." Advances in neural information processing systems 33 (2020): 6840-6851.