robinsingh-ai/dcgan

DCGAN

Loss Curve for DCGAN on Fashion MNIST

Generated Images for DCGAN on Fashion MNIST

Loss Curve for DCGAN on MNIST

Generated Images for DCGAN on MNIST

Introduction

This is an implementation of DCGAN on Fashion MNIST and MNIST datasets using PyTorch.

Dataset

The Fashion MNIST dataset is used for training and testing the model. It consists of 60,000 28x28 grayscale images of 10 fashion categories, along with a test set of 10,000 images. The MNIST dataset is also used for training and testing the model. It consists of 60,000 28x28 grayscale images of the 10 digits, along with a test set of 10,000 images.

Model

The model consists of a generator and a discriminator. The generator takes a random noise vector as input and generates a 28x28 grayscale image. The discriminator takes an image as input and outputs a scalar value representing the probability that the image is real (as opposed to generated). The generator and discriminator are trained simultaneously using the adversarial loss.