clean-fid
clean-fid copied to clipboard
GaParmar
PyTorch - FID calculation with proper image resizing and quantization steps [CVPR 2022]
Hi, thanks for sharing your work. In my case, I have a reference folder (fdir1), and about 50 target folders (fdir2), each containing hundreds of images. It takes a long...
This code is from fid.py line 459-470. elif gen is not None: if not verbose: print(f"compute FID of a model with {dataset_name}-{dataset_res} statistics") score = fid_model(gen, dataset_name, dataset_res, dataset_split, model=feat_model,...
I found that when calculating custom dataset, using UPPERCASE named `custom_name` such as "dragan_Rs", it will be ERROR. And i change it to "dragan_rs", its OK. I don't know why....
From 3 mins on a100 it went to 40 mins. Also, Pytorch added antialiasing to interpolation, should it be used for speding up calculations?
FID error
RuntimeError: PytorchStreamReader failed reading zip archive: failed finding central directory How to solve it?
Default saved path `No such file or directory: '/tmp\\inception-2015-12-05.pt'` which is not so suitable for Windows ~
Hi, Thanks for your contribution on GAN research! We tried to get clean-fid for the images generated by styleganv2 using your code. However, the score was about 24, which was...
Env: python 3.9.6, clean-fid: 0.1.15 My code: ```python from cleanfid import fid score = fid.compute_fid('./fake_images', '../real_images') ``` Error: ```shell File "C:\Users\10034\AppData\Local\Programs\Python\Python39\lib\site-packages\cleanfid\fid.py", line 389, in compute_fid score = compare_folders(fdir1, fdir2, feat_model,...
Hi, Thanks for maintaining the library for GAN research! I noticed that the [KID implementation in this repository](https://github.com/GaParmar/clean-fid/blob/main/cleanfid/fid.py#L75-L77) is quite different from [the official KID implementation](https://github.com/mbinkowski/MMD-GAN/blob/678bb5e2d5f7b0bb8dd5c3591d7759e1bb3f8018/gan/compute_scores.py#L231) Could the author explain...
Dear clean-fid group, Thank you for sharing this great work, I really like it. You mention in the paper that the PIL implementation is anti alias. But the PIL library...
