Added benchmark object #878
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
Muennighoff
reviewed
Jun 4, 2024
There was a problem hiding this comment.
Nice, I think we also need to change the README which imports MTEB_MAIN_EN i.e. https://github.com/embeddings-benchmark/mteb?tab=readme-ov-file#dataset-selection
Added a test and it does run just fine (since it has an iter method). I also changed the type hint to iterable (weaker than sequence) to reflect that it is the only requirement. |
KennethEnevoldsen
added a commit
that referenced
this pull request
Jun 5, 2024
* Ensure result are consistently stored in the same way - (due to failing test): updated missing dataset references - (to test with more than one model) Added e5 models base and large - updated mteb.get_model to now include metadata in the model object - ensure that model name is always included when saving (with a default when it is not available) - use the ModelMeta for the model_meta.json * format * minor test fixes * docs: Minor updated to repro. workflow docs * fixed failing test * format * Apply suggestions from code review Co-authored-by: Isaac Chung <chungisaac1217@gmail.com> * docs: update PR template * fix: Added benchmark object (#878) * removed duplicate task * Added benchmark object * removed import for duplicate task * fix dataset references * added seb * Added test for running benchmarks * changed tasks to be an iterable * format * Apply suggestions from code review Co-authored-by: Niklas Muennighoff <n.muennighoff@gmail.com> Co-authored-by: Isaac Chung <chungisaac1217@gmail.com> --------- Co-authored-by: Isaac Chung <chungisaac1217@gmail.com> Co-authored-by: Niklas Muennighoff <n.muennighoff@gmail.com>
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.
Assuming we want a selection menu for benchmarks in the future leaderboard I have now added a benchmark object. Generally it shouldn't change much otherwise.
The idea is that you would select the benchmark. The description will show + some references. Potentially with a dropdown for the citation.