Skip to content

Experimenting with Generative Adversarial Networks (GAN) to generate images similar to the Fashion MNIST dataset using PyTorch.

Notifications You must be signed in to change notification settings

alvayliu/Fashion-GAN

Repository files navigation

Generative Adversarial Network

Experimenting with Generative Adversarial Networks (GAN) using PyTorch. The GAN built in this project can generate images of clothes. Here are some of the images it generated:

alt text

Data: Fashion MNIST, 60000 samples of greyscale 28x28 px images.

To train the models from scratch, download all python files in this folder and run with

python3 train.py

To generate samples with the trained models, download the python files and the models from the trained_models folder. Run with

python3 test.py

About

Experimenting with Generative Adversarial Networks (GAN) to generate images similar to the Fashion MNIST dataset using PyTorch.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published