Skip to content

Conversation

@EgorShternov2
Copy link

This PR adds a new self-contained template vae-mnist/.

Features:

  • Minimal PyTorch VAE (encoder/decoder + reparameterization)
  • ELBO loss: reconstruction (BCE) + beta * KL
  • Train/val/test split with reproducible seed
  • Qualitative artifacts: reconstructions and samples
  • Model checkpoint and loss curve

No changes to fenn core.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant