Minor Fix - Final Batch Size if Total Samples to use in FID Calculation is a multiple of Batch Size #17
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…ple of batch size on the final batch.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Thank you so much for open sourcing this work! I think I may have found a very, very slight issue.
When using the
--eval-total-sizeargument, the final batch in the evaluator will be0if the value passed to--eval-total-sizeis a multiple of the batch size, i.e.self.eval_total_size % self.eval_batch_size = 0.This will cause a
ValueErrorlater in the FID calculation since we compute the mean of an empty array.