Skip to content

A web-app that uses a vanilla generative adversarial network and unsupervised learning to generate random 28X28x1 plasma style images of digits from 0 to 9

License

Notifications You must be signed in to change notification settings

parikshitkumar1/0x9_Generator

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

23 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation


0x9 Generator

Motivation

To train a vanilla generative adversarial network to generate images from 0 to 9

Requirements

Python 3.8 or above with all requirements dependencies installed. To install run:

$ pip3 install -r requirements.txt

To run

$ streamlit run digit.py

or check it out here: https://weird-digits.herokuapp.com/

Examples

2 different vanilla architectures were trained for nearly 400 epochs

After 10 epochs:

After 100 epochs:

After 200 epochs:

After 300 epochs:

After 399 epochs:

Might Do

  • Use DCGAN, or controllable generation.

About

A web-app that uses a vanilla generative adversarial network and unsupervised learning to generate random 28X28x1 plasma style images of digits from 0 to 9

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published