Repository navigation
FIX: key model-less objective targets by content hash in analytics v1 - #3171
Open
Utkarsh Bahuguna (u7k4rs6) wants to merge 1 commit into
Open
Utkarsh Bahuguna (u7k4rs6) wants to merge 1 commit into
Utkarsh Bahuguna (u7k4rs6) wants to merge 1 commit into
Conversation
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Analytics keyed objective targets by model name, temperature and top_p, so every target without a model name (HTTP, Gandalf and so on) of one class ended up in a single unlabeled group. As Roman suggested, this is a pre-release correction to v1 rather than a v2. Targets with a model name keep their key, and the ones without fall back to their content hash, after unwrapping a wrapper the same way the eval rules do. Their labels skip the blank model name and fall back to the class name, so the SDK shows something like HTTPTarget with a short key and no endpoint details.
A new migration reruns the v1 backfill, so databases that already ran the original one get the corrected keys too, and the frozen-v1 docs note the correction. The tests cover distinct HTTP endpoints and Gandalf levels, blank model names, wrappers, unchanged model grouping, grouping and filtering across reports, facets and drill-downs, and a database that already applied the original migration.
Fixes #3087