Skip to content

moshesipper/vae-torch-celeba

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

vae-torch-celeba

Accompanying code for my Medium article: A Basic Variational Autoencoder in PyTorch Trained on the CelebA Dataset .

Files:

  • vae.py: Class VAE + some definitions. You can change IMAGE_SIZE, LATENT_DIM, and CELEB_PATH.
  • trainvae.py: Main code, training and testing. You can change EPOCHS and BATCH_SIZE. The models and images are placed in a directory vaemodels-??????, where ?????? are 6 random characters.
  • utils.py: a couple of small utility functions.
  • genpics.py: creates a panel of original image + 7 reconstructed ones.
  • vae_model_20.pth: a trained VAE.

Running a trained model on a CPU is fine.

Training on a CPU is possible, but slow: ⚡👉 GPU.

Releases

No releases published

Packages

No packages published

Languages