https://github.com/gazeux33/variational-autoencoders

The aim is to create a VAEs in with PyTorch to create synthetic data
https://github.com/gazeux33/variational-autoencoders

Last synced: 19 days ago
JSON representation

The aim is to create a VAEs in with PyTorch to create synthetic data

• Host: GitHub
• URL: https://github.com/gazeux33/variational-autoencoders
• Owner: Gazeux33
• Created: 2024-04-25T21:16:18.000Z (8 months ago)
• Default Branch: main
• Last Pushed: 2024-05-04T15:05:18.000Z (8 months ago)
• Last Synced: 2024-05-05T11:37:10.026Z (8 months ago)
• Language: Jupyter Notebook
• Homepage:
• Size: 407 KB
• Stars: 0
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Variational-Autoencoders(VAEs)

## The project 

This project is an implementation of Variational Autoencoders (VAEs) using PyTorch. VAEs are generative models that learn to represent high-dimensional data in a lower-dimensional latent space and can generate new data samples similar to the training data.

## Features

+ Implementation of a Variational Autoencoder architecture.

+ Training pipeline for learning latent representations.

+ Generation of new data samples from the learned latent space.




## What is a VAE ?

 


 

 


 


## Technical specifications

| Property       | Value         |

|----------------|---------------|

| Device         | MAC M2        |

| Training Time  | ~20 min       |

| Epochs         | 5             |

| Training Data  | FashionMNIST  |

| Framework      | PyTorch       |

| Learning rate  | 0.001         |

|Z_DIM           |10             |

|BATCH_SIZE      |32             |

 ## Examples of constructions and reconstructions

  


 

 


 


 ## Examples of images extracted directly from latent space (fictives images)