A Jittor Implementation of Conditional GAN

This is a Jittor implementation of Conditional GAN. The main framework can be found at gitlink. Here I provide a fulfilled example of this framework.

Setup

pip install -r requirements.txt

A detailed tutorial for Jittor installation can be found at Jittor.

Usage

cd src
python CGAN.py

The program will automatically load the MNIST dataset from web and start traning.

The last check point will be stored in src/generator_last.pkl and src/discriminator_last.pkl.

The pictures generated during training will be stored in src folder.