Fix non-deterministic cache hash for MetricGrouping tasks#1260
Open
nuthalapativarun wants to merge 1 commit into
Open
Fix non-deterministic cache hash for MetricGrouping tasks#1260nuthalapativarun wants to merge 1 commit into
nuthalapativarun wants to merge 1 commit into
Conversation
Author
|
Just checking in on this one - no urgency, but wanted to bring it back up in case it slipped through. Let me know if any changes are needed. |
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
`LightevalTaskConfig.str` (used by `SampleCache._get_task_hash` to build the cache key) reprs each metric field directly. For `MetricGrouping` metrics, `corpus_level_fn` and `higher_is_better` are dicts whose values are callables, so `repr(...)` includes the function's memory address (e.g. `<function compute at 0x7f...>`). This makes the resulting cache hash non-deterministic across runs, breaking the cache for tasks like `leaderboard|truthfulqa:mc|0`.
This adds a small helper that recurses into dict values and renders callables by name (matching the existing handling for top-level metric callables), so the hash is stable across processes/runs.
Closes #1023