galgen

This repository contains the implementation of various generative models for generating synthetic galaxy images as part of our project. The models implemented include Variational Autoencoders (VAE), Unconditional Diffusion, Conditional Diffusion, nsf and Glow.

Model Implementations

The implementations of the models used in this project are based on the following repositories:

Variational Autoencoder (VAE)

• Original implementation: PyTorch-VAE by AntixK

Unconditional Diffusion

• Original implementation: diffusers by Hugging Face

Conditional Diffusion

• Original implementation: Conditional_Diffusion_MNIST by TeaPearce

Glow (Normalizing Flow)

• Original implementation: glow-pytorch by rosinality

Neural Spline Flows (Normalizing Flow)

• Original implementation: nsf by bayesiains

Results

Sample galaxy images generated by each model, along with quantitative results, can be found in our corresponding report.

Acknowledgements

We would like to express our gratitude to the authors of the original implementations for providing the foundation for our project. Their work has been instrumental in helping us explore and evaluate various generative models for synthetic galaxy image generation.