Content safety evals aggregate max from conversations #39083
+129
−3
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
MilesHolland
changed the title
github-actions
bot
added
the
Evaluation
nagkumar91
approved these changes
Jan 8, 2025
|
API change check API changes are not detected in this pull request. |
There was a problem hiding this comment.
Copilot reviewed 5 out of 10 changed files in this pull request and generated no comments.
Files not reviewed (5)
- sdk/evaluation/azure-ai-evaluation/tests/unittests/data/evaluate_test_data_conversation.jsonl: Language not supported
- sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_evaluators/_content_safety/_violence.py: Evaluated as low risk
- sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_evaluators/_content_safety/_hate_unfairness.py: Evaluated as low risk
- sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_evaluators/_content_safety/_sexual.py: Evaluated as low risk
- sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_evaluators/_content_safety/_self_harm.py: Evaluated as low risk
Comments suppressed due to low confidence (3)
sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_evaluators/_common/_base_eval.py:77
- [nitpick] Update the docstring to reflect the correct class name if it is renamed to
ConversationNumericAggregationType.
:type conversation_aggregation_type: ~azure.ai.evaluation._constants._ConversationNumericAggregationType
sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_evaluators/_common/_base_rai_svc_eval.py:41
- Missing period at the end of the docstring.
Default is ~azure.ai.evaluation._constants.ConversationNumericAggregationType.MEAN.
sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/_evaluators/_common/_base_rai_svc_eval.py:42
- The
conversation_aggregation_typeparameter should be explicitly mentioned in the constructor's docstring.
:type conversation_aggregation_type: ~azure.ai.evaluation._constants.ConversationNumericAggregationType
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.
Adds a new private argument to the base and rai base evaluator classes which determines how numeric results from conversations are evaluated into a single overall result. Note that this has no impact on the across-multiple-evaluations aggregation performed by the
evaluatefunction.Current options are mean (the default), max, and min. Content safety evaluators have their default aggregation set to max.
Added tests to ensure that the aggregation options all work as intended, and that the content safety evals have the proper value set.
Note that the aggregate content safety evaluator does not need to set anything, since its internal evaluators handle their own aggregation methods.