A collection of Variational AutoEncoders (VAEs) I have implemented in jax/flax, flux and pytorch with particular effort put into readability and reproducibility.
- Python >= 3.8
- jax
$ git clone https://github.com/BeeGass/Readable-VAEs.git
$ cd Readable-VAEs/vae-jax
$ python main.py
- PyTorch >= 1.10
$ cd Readable-VAEs/vae-pytorch
$ python main.py
- TODO
- TODO
$ cd Readable-VAEs/vae-flux
$ # TBA
Config File Template
TBA
Weights And Biases Integration
TBA
Model | PyTorch | Jax/Flax | Flux | Config | Paper | Reconstruction | Samples |
---|---|---|---|---|---|---|---|
VAE | ☑ | ☐ | ☐ | ☐ | Link | TBA | TBA |
Beta-VAE | ☐ | ☐ | ☐ | ☐ | Link | TBA | TBA |
Conditional VAE | ☐ | ☐ | ☐ | ☐ | Link | TBA | TBA |
VQ-VAE-2 | ☐ | ☐ | ☐ | ☐ | Link | TBA | TBA |
@software{Gass_Readable-VAEs_2021,
author = {Gass, B.A., Gass, B.A.},
doi = {10.5281/zenodo.1234},
month = {12},
title = {{Readable-VAEs}},
url = {https://github.com/BeeGass/Readable-VAEs},
version = {1.0.0},
year = {2021}
}