DeepNetworks

My implementations of deep neural networks for practice.

Implemented Methods

This repository implements the following:

• Subpixel Convolutions: paper, code
• Translated Parametric ReLU: paper, code
• Weight Normalization: paper, code
• Label Smoothing: paper, code
• Wasserstein GAN: paper, code
• Improved Training of Wasserstein GANs: paper, code
• Deep Regret Analytic Generative Adversarial Networks: paper, code
• Conditional Image Synthesis With Auxiliary Classifier GANs: paper
• Learning to Discover Cross-Domain Relations with Generative Adversarial Networks: paper, code
• Minibatch Stat Concatenation: paper, code

Requirements

• tensorflow >= 1.4.0
• Python 3