mirror of
https://github.com/openai/CLIP.git
synced 2026-11-05 01:37:26 +00:00
[GH-ISSUE #193] Different accuracy for different batch sizes #166
Labels
No labels
pull-request
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
openai/CLIP#166
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @harryfin on GitHub (Dec 6, 2021).
Original GitHub issue: https://github.com/openai/CLIP/issues/193
The different accuracies occur when the model is tested with different batch sizes.
I assume it is related to Batch Normalization, even though the model has been set to evaluation mode before testing.
Used Model: 'RN50'
Is there a known solution to this problem?