Address sporadic hanging of evals on certain samples #1482
Merged
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
thesofakillers
requested review from
andrew-openai,
etr2460 and
katyhshi
as code owners
This was referenced Mar 21, 2024
Merged
JunShern
pushed a commit
that referenced
this pull request
Mar 26, 2024
We're now implementing solvers for new APIs we're calling (Anthropic, Gemini, ...). Each solver was implementing the same logic for backing off and retrying when the API query limit was hit. This PR created a generic create_retrying function, which retries when specific exceptions are raised. These exceptions are passed as arguments. This uses the changes from #1482
JunShern
added a commit
that referenced
this pull request
Mar 26, 2024
Adds a solver for Gemini 1.5 Pro. Stacked on #1501 and #1482. Using the solver requires the `GEMINI_API_KEY` environment variable Test with: ``` oaieval generation/direct/gemini-pro bugged_tools ``` --------- Co-authored-by: Chan Jun Shern <[email protected]>
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Outdated suggestions cannot be applied.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.
As has been brought up before (#1384, #1292, #270), evals suffer from a hanging issue, where an evaluation run will hang for a very long time (if not indefinitely) at the end of a run (say, on the 99th sample of out 100).
This PR addresses this issue, by replacing a seemingly redundant single-threaded thread creation that was happening when making requests, nested inside the already multi-threaded eval loop. My impression is that this nested multithreading was causing overhead that resulted in the hanging experienced.
I had also noticed this hanging issue in
EVALS_SEQUENTIAL=1mode (where it no longer occurs at the end, but instead randomly in the middle of the run).I was able to identify the source of this issue though debugging print statements that ultimately pointed to the
request_with_timeoutfunction as the culprit.We have tested the new
request_with_timeoutcode on a fork where we have run multiple new and pre-existing evals, including with 3rd party solvers, and found no change in behaviour or errors, and a clear improvement on the hanging issue.