A crowded low-cardinality dimension no longer pushes the true grain out of the composite-key probe - #237
Merged
Merged
Conversation
marcociav-exmergo
changed the title
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes : #168
_probe_composite_keys (explore/profile.py) raises _COMPOSITE_PAIR_CAP from 3 to 5, and adds a new _COMPOSITE_REDUNDANCY_RATIO = 3.0 dedup pass before the cut: a pair sharing a column with an already-kept pair is dropped as a near-duplicate ("same dimension, different filler") only when its product is within 3x of the kept pair's a genuinely different hypothesis still gets its own slot even if it reuses that column.
Fixes the crowding case from the issue: several junk pairs anchored on one low-cardinality dimension no longer consume the entire cap and starve out the true grain, which can legitimately score worse on raw distinct-count product.
No change to cost-guard behavior: each extra slot is still a real scan batched into one statement, and a metered adapter that can't cover the wider batch already degrades to "grain unknown" via distinct_combination_counts's existing contract.
The issue :
After fix:
