An empirical study on evaluation metrics of generative adversarial networks.
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
framework Update README.md Jun 21, 2018
.gitignore rename Jun 17, 2018
README.md Update README.md Jun 21, 2018
demo.gif update gif Jun 20, 2018
demo_dcgan.py minor fix + add demo Jun 16, 2018
metric.py minor fix + add demo Jun 16, 2018

README.md

GAN Metrics

This repository provides the code for An empirical study on evaluation metrics of generative adversarial networks.

Requirement

  • Python 3.6.4
  • torch 0.4.0
  • torchvision 0.2.1
  • ot 0.4.0
  • tqdm 4.19.6
  • numpy, scipy, math

Usage

We create a demo for DCGAN training as well as computing all the metrics after each epoch.

python3 demo_dcgan.py \
--dataset cifar10 \
--cuda \
--dataroot <data_folder> \
--outf <output_folder> \
--sampleSize 2000

demo