Add E2B code interpreter reward function #364
Merged
+211
−1
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
qgallouedec
reviewed
Feb 19, 2025
| # TODO: add support for other languages in E2B: https://e2b.dev/docs/code-interpreting/supported-languages | ||
| try: | ||
| """Returns a reward function that evaluates code snippets in a sandbox.""" | ||
| evaluation_script_template = """ |
There was a problem hiding this comment.
It's surprising that you don't have any issue with the extra indentation.
There was a problem hiding this comment.
I usually use textwrap.dedent in this case, but it might not be necessary here for some reason.
There was a problem hiding this comment.
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Outdated suggestions cannot be applied.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.
This PR adds a reward function to execute Python code safely via E2B: https://e2b.dev/docs/legacy/sandbox/api/debugging
It's currently targeted for coding competitions where the ground truth is given in the form of test cases and the reward per problem is defined as the overall success rate.
Experiments to test the feature can be viewed here: https://wandb.ai/huggingface/open-r1/reports/Qwen2-5-1-5B-Instruct-with-Python-code-interpreter--VmlldzoxMTQxNTA3Mg
TODO