Repository navigation
Conversation
added 2 commits
October 22, 2026 16:11
VRR scores a prediction valid iff it matches ANY dictionary-licensed reading of its skeleton (single-gold DER undercounts legitimate ambiguity — the WO56 structural rule). build_lexicon groups vocalized forms by skeleton (bare forms excluded — a form with no haraka is not a reading); score_text_vrr does word-level paragraph scoring for SadeedDiac-style rows. Convention declared in the result, per C2.
…positional-truncate by design
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Implements the ambiguity-aware metric from WO56's vocalization-as-composition rule: a prediction is valid when it matches ANY dictionary-licensed reading of its skeleton.
build_lexicon: skeleton → licensed readings from pinned vocalized corpora (bare forms excluded — no haraka, not a reading).score_predictions_vrr: row-level scoring with the convention declared in the result (C2).score_text_vrr: word-level paragraph scoring for SadeedDiac-style rows.TDD: 6 tests, skeletons/lexicon-building/word-level semantics. First published number lands via the lexicon-build job (tashkeela, pinned) against the frozen frontier-predictions — reported alongside DER, never instead.