Two time-scale update rule for training GANs
This repository contains code accompanying the paper GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium.
The FID is the performance measure used to evaluate the experiments in the paper. There, a detailed description can be found in the experiment section as well as in the the appendix in section A1.
In short: The Fréchet distance between two multivariate Gaussians X1 ~ N(mu1, C1) and X2 ~ N(mu2, C2) is
d^2 = ||mu_1 - mu_2||^2 + Tr(C_1 + C_2 - 2*sqrt(C_1*C_2)).
The FID is calculated by assuming that X1 and X2 are the activations of the coding layer pool3 of the inception model (see below) for generated samples and real world samples respectivly. mun is the mean and C_n the covariance of the activations of the coding layer over all real world or generated samples.
IMPORTANT: The number of samples to calculate the Gaussian statistics (mean and covariance) should be greater than the dimension of the coding layer, here 2048 for the Inception pool 3 layer. Otherwise the covariance is not full rank resulting in complex numbers and nans by calculating the square root.
We recommend using a minimum sample size of 10,000 to calculate the FID otherwise the true FID of the generator is underestimated.
Previous versions of this repository contained two implementations to calculate the FID, a "unbatched" and a "batched" version. The "unbatched" version should not be used anymore. If you've downloaded this code previously, please update it immediately to the new version. The old version included a bug!
If you're looking for a pytorch implementation we recommend https://github.com/mseitzer/pytorch-fid
Requirements: TF 1.1+, Python 3.x
This file contains the implementation of all necessary functions to calculate the FID. It can be used either as a python module imported into your own code, or as a standalone script to calculate the FID between precalculated (training set) statistics and a directory full of images, or between two directories of images.
To compare directories with pre-calculated statistics (e.g. the ones from http://bioinf.jku.at/research/ttur/), use:
fid.py /path/to/images /path/to/precalculated_stats.npz
To compare two directories, use
fid.py /path/to/images /path/to/other_images
fid.py --helpfor more details.
Example code to show the usage of
fid.pyin your own Python scripts.
Example code to show how to calculate and save training set statistics.
Improved WGAN (WGAN-GP) implementation forked from https://github.com/igul222/improvedwgantraining with added FID evaluation for the image model and switchable TTUR/orig settings. Lanuage model with JSD Tensorboard logging and switchable TTUR/orig settings.
Precalculated statistics for datasets - cropped CelebA (64x64, calculated on all samples) - LSUN bedroom (calculated on all training samples) - CIFAR 10 (calculated on all training samples) - SVHN (calculated on all training samples) - ImageNet Train (calculated on all training samples) - ImageNet Valid (calculated on all validation samples)
are provided at: http://bioinf.jku.at/research/ttur/
For FID evaluation download the Inception modelf from http://download.tensorflow.org/models/image/imagenet/inception-2015-12-05.tgz
The cropped CelebA dataset can be downloaded here http://mmlab.ie.cuhk.edu.hk/projects/CelebA.html
To download the LSUN bedroom dataset go to: http://www.yf.io/p/lsun
The 64x64 downsampled ImageNet training and validation datasets can be found here http://image-net.org/small/download.php