Re-implement DFGAN with pytorch

Acknowledgements

This repo is fork from generative-model-collections and reference tensorflow-dfgan Thanks for their sharing.

I implement noise_generator and moving average trick in code:

CGAN_dfgan.py

and test this implement with MNIST dataset.

you can only use default configuration for training MNIST dataset with dfgan.

python main.py

The full code in :

CGAN_dfgan.py

If your have any problem , please tell me.

Each row has the same noise vector and each column has the same label condition.

Name	Epoch 1	Epoch 25	Epoch 50	GIF
CGAN
DFGAN