search

lucidrains / denoising-diffusion-pytorch

Implementation of Denoising Diffusion Probabilistic Model in Pytorch
MIT License
8.34k stars 1.03k forks source link

Information about FID metric to evaluate the model #68

Open rinodipaola15 opened 2 years ago

rinodipaola15 commented 2 years ago

I trained the network using the training images of Cifar10's 'horse' class (5000 samples). I want to compute FID metric to evaluate the model, computing the "distance" between my generated images and dataset images. To do this, do I have to pass, in the FID formula, the train dataset, the test dataset or both?

!python pytorch_fid path1 path2 path1 = /generated_images_folder path2 = ?

Ps. Congratulations for your excellent work!

pimakshay commented 1 year ago

path1 = /generated_images_folder (fake images) path2 = /testing_dataset (real images)