Repository navigation
Say what FiQA's default-analysis rows are comparable with - #80
Merged
Merged
Conversation
The FiQA rows were measured under PostgreSQL's simple configuration, and the published 0.2361 is Lucene's English analysis at k1=0.9/b=0.4. The page now states that, and the aligned measurement that closes it: in-memory BM25 0.237 against 0.2361, this engine 0.2291, the same ~0.007 analysis gap as the other corpora.
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #80 +/- ##
==========================================
- Coverage 89.19% 89.16% -0.03%
==========================================
Files 270 270
Lines 21383 21383
Branches 4621 4621
==========================================
- Hits 19073 19067 -6
- Misses 1388 1391 +3
- Partials 922 925 +3 ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What changed
Two paragraphs, no code. The FiQA rows were measured under the library's default analysis — PostgreSQL
simple+unaccent, k1=1.5/b=0.75 — while the published 0.2361 is Lucene's English analysis at k1=0.9/b=0.4. The pages now say so, and give the aligned measurement that closes the gap:englishconfig, k1=0.9/b=0.4, gatesimpleconfig, k1=1.5/b=0.75, gatesimpleconfig, k1=1.5/b=0.75, no gateSo the published figure is reproduced when the analysis and operating point are aligned (0.237 against 0.2361), and the SQL row is ~0.007 short for the documented reason — PostgreSQL's English configuration is not Lucene's analyzer. The gate is not the cause: the aligned SQL row is a gated row, and on FiQA the gate's candidate set does not truncate.
docs/evaluation.mdanddocs/backends.mdcarry the numbers; no behaviour changes.