Question about FID calculation about ddpm-torch HOT 2 CLOSED

Hi, thank you for asking! I want to clarify that the final output of the covariance matrix by my algorithm is still an unbiased estimator, meaning that the effective ddof is 1, just like those implementations you mentioned. Let me explain why:

Actually, in my implementation, I derive a simple online algorithm for covariance matrix calculation (an unbiased estimator), where I create a covariance matrix as a running statistic and update it at each mini-batch. You can find a simple version in uni-variate case at Wikipedia link. For convenience, I choose ddof=0 (which is biased) to keep track of the running statistic, as the biased covariance estimator is always well-defined (even if you have a batch size of 1). Note that in

when I extract the running statistic once the process has gone through the whole dataset to evaluate, I use the unbiased correction by multiply it by $N/(N-1)$.

from ddpm-torch.

Thanks for your kind explanation:) It helped me a lot.
May good luck always be with you!

from ddpm-torch.