Skip to content

Sort files to ensure deterministic runs - #832

Closed
mrpositron wants to merge 2 commits into
mainfrom
nauryzbay-data-should-be-sorted
Closed

Sort files to ensure deterministic runs#832
mrpositron wants to merge 2 commits into
mainfrom
nauryzbay-data-should-be-sorted

Conversation

@mrpositron

Copy link
Copy Markdown
Contributor

What has changed and why?

The input filenames are now sorted before processing. Previously, the file processing order depended on the order returned by the filesystem, which is not guaranteed to be consistent across platforms or architectures. As a result, the processing sequence could differ even with shuffle=False, leading to different validation results. Sorting the filenames ensures a deterministic processing order across runs and environments.

How has it been tested?

NA

Did you update CHANGELOG.md?

  • Yes
  • Not needed (internal change)

Did you update the documentation?

  • Yes
  • Not needed (internal change without effects for user)

@mrpositron mrpositron closed this Jul 21, 2026
@mrpositron
mrpositron deleted the nauryzbay-data-should-be-sorted branch July 21, 2026 13:30
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant