dcgan-svhn

Cycle GAN Winter-Summer transformation. The code is based on the miniproject of the chapter on GAN of the Udacity Deep Learning course.

Content

The notebook implements the following steps

• Load in the data
• Pre-process the data reducing the size of the images to 128x128x3
• Define the Generator and Discriminator architectures
• Define loss function and optimizer
• Train the networks, save after training and generate new images

Features

If available, PyTorch trains on GPU.

References

Udacity github