bug in fid: normalization

[youngjung]

Hi, I found a bug in fid score.

The normalization before inception model is written as:

pytorch.sngan_projection/models/inception.py

Lines 128 to 132 in 7afdca4

	if self.normalize_input:
	x = x.clone()
	x[:, 0] = x[:, 0] * (0.229 / 0.5) + (0.485 - 0.5) / 0.5
	x[:, 1] = x[:, 1] * (0.224 / 0.5) + (0.456 - 0.5) / 0.5
	x[:, 2] = x[:, 2] * (0.225 / 0.5) + (0.406 - 0.5) / 0.5

but it should be:
https://github.com/mseitzer/pytorch-fid/blob/4e366b2fc9fb933bec9f6f24c5e87c3bd9452eda/inception.py#L130-L131

or it should be fixed like:

if self.normalize_input:
    mean = torch.tensor((0.485, 0.456, 0.406)).view(1, 3, 1, 1).cuda()
    std = torch.tensor((0.229, 0.224, 0.225)).view(1, 3, 1, 1).cuda()
    x = (x - mean) / std

[crcrpar]

Thank you for your report & suggestion!

I'm sorry for my late response and bothering you by my bad.
To be honest, I did not test/use FID scores.

[youngjung]

Oh, I did not mean to blame you but just wanted to help your repo and other users.
Thanks for your response :)

[crcrpar]

Oh, I did not mean to blame you but just wanted to help your repo and other users.

Thank you for your kind words.

I added the link to this issue tentatively to clarify what is wrong & how to fix this.