FID sensitive to image saving about pytorch-fid HOT 4 OPEN

I have the same issue. Now I save and load [0,1] range raw data instead of images for the evaluation, and numbers make sense to me.

from pytorch-fid.

This should not be the case because generated images are often quantized to uint8 before computing the FID score.
(because real images are not real values).
The different in FID might be cause due to JPEG compression.
The table 4 and figure 2 in this paper might be relevant to this issue.
( https://arxiv.org/pdf/2104.11222.pdf )

from pytorch-fid.

I have the same issue. Now I save and load [0,1] range raw data instead of images for the evaluation, and numbers make sense to me.

As saving it to be image format is not okay, how to save it to be [0, 1] range raw data, something like .npz?

from pytorch-fid.

Thank you for the solution.
Here is the reference architecture of the InceptionV3 model

Also refer to this solution

from pytorch-fid.