https://github.com/ekzhang/vae-cnn-mnist

Conditional variational autoencoder applied to EMNIST + an interactive demo to explore the latent space.
https://github.com/ekzhang/vae-cnn-mnist

keras mnist variational-autoencoder

Last synced: 3 months ago
JSON representation

Conditional variational autoencoder applied to EMNIST + an interactive demo to explore the latent space.

• Host: GitHub
• URL: https://github.com/ekzhang/vae-cnn-mnist
• Owner: ekzhang
• Created: 2018-08-07T01:26:39.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2018-08-09T00:56:17.000Z (almost 7 years ago)
• Last Synced: 2025-03-23T21:23:19.621Z (3 months ago)
• Topics: keras, mnist, variational-autoencoder
• Language: Jupyter Notebook
• Homepage: https://ekzhang.github.io/vae-cnn-mnist/
• Size: 22.2 MB
• Stars: 23
• Watchers: 2
• Forks: 4
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Variational Autoencoder for MNIST digits

![Sample](sample.png)

## Architecture

### Encoder

![Encoder](vae_cnn_encoder.png)

### Decoder

![Decoder](vae_cnn_decoder.png)

### VAE

![VAE](vae_cnn.png)