Hugging Face's logo

Spaces:
bayan10
/
bayan-api
Sleeping

App Files Community
bayan-api / src /nlp /spelling

Commit History

  1. Update extension UI, app pipeline, and Spelling Rules

    youssefreda9
    75e0099

  2. Update extension UI, app pipeline and model loaders

    youssefreda9
    9c49fff

  1. Update extension UI, spelling rules, and benchmark artifacts

    youssefreda9
    183ba6e

  2. Fix structured data regex bug in spelling rules

    youssefreda9
    1e65c93

  3. Fix structural regressions

    youssefreda9
    e7b2a51

  4. Fix: 10 critical NLP logic bugs in grammar, spelling, and punctuation to prevent false positives

    youssefreda9
    41ea30b

  5. Fix 30 NLP edge cases in Grammar, Spelling, and Punctuation (Phase 10 results and Extension UI improvements)

    youssefreda9
    be32617

  1. update benchmark logs

    youssefreda9
    a84aad9

  2. Fix collision benchmark failures: patch overlap mapping, tokenization guard, and rule hallucinations

    youssefreda9
    df5e501

  3. Fix collision benchmark: non-overlap, destructive tokenization, colon guard, hamza punctuation

    youssefreda9
    5ed1cab

  1. fix: contextual logging, word merge overlap resolver, and responsive toolbar/footer UI

    youssefreda9
    2702d02

  1. refactor: simplify architecture and replace whitelists with CamelTools

    Bayan AI
    ebd15e7

  1. FIX-47: Expand hamza whitelist with verb+pronoun forms⁠

    youssefreda9
    eebdf74

  2. FIX-44: OOV cleanup pass between spelling and grammar stages⁠

    youssefreda9
    4655865

  3. Fix: TerminalPunctuationGuard for short phrases and comprehensive Nasb/Jazm grammar rules fixes

    Bayan AI
    76b9ec3

  1. Phase 12: Spelling pipeline integration + Benchmark integrity fixes⁠

    youssefreda9
    31f02a4

  2. Phase 12: Fix 6 batches — grammar pipeline bypass, religious/structured protection, punct rearrangement, SV/gender agreement⁠

    youssefreda9
    81c58cc

  3. fix: Apply hamza+ta-marbuta corrections AFTER validation steps to prevent reversion⁠

    youssefreda9
    b7e415b

  4. Phase 11: Master Bug Fix — 23 fixes across 8 files⁠

    youssefreda9
    bfc07c3

  1. Phase 8: Fix all critical bugs from adversarial validation⁠

    youssefreda9
    76ea593

  2. revert: restore backend/model files to cf07939 state, keep UI/UX changes only

    youssefreda9
    143c028

  3. UI: nav logo, bayyinah hover, quran icon, hero autocomplete shadow, clean tooltip, fix punctuation dup + hamza كان

    youssefreda9
    20a34f1

  1. fix: re-label grammar spelling fixes + handle prefix in hamza whitelist

    youssefreda9
    8f13258

  2. fix: all model bugs — S1 S2 S3 G1 P1 (6 fixes across 4 files)⁠

    youssefreda9
    bc71395

  1. fix: PDF dropdown visible + verify spelling context + araspell audit - Fix summary export dropdown clipping (overflow: visible + opens upward) - Verify: use_contextual=True in araspell_service.py (line 83) ✓ - Verify: ContextualCorrector fully initialized and used in pipeline ✓ - Verify: All araspell_rules.py classes used: AraSpellPostProcessor, ErrorClassifier, RulesBasedCorrector, OutputValidator, VocabularyManager, WordAligner, SplitMergeSpecialist, EditDistanceCorrector, ContextualCorrector, ArabicSpellChecker ✓ - Fix misleading comment about contextual=False

    youssefreda9
    d4b662b

  1. feat: Enable ContextualCorrector (BERT MLM) for spelling - Pre-download AutoModelForMaskedLM in Dockerfile

    youssefreda9
    bcf3607

  2. feat: Add AraSpell spelling correction endpoint (POST /api/spelling) with lazy model loading - Extract AraSpell pipeline into src/nlp/spelling/ (araspell_rules.py + araspell_service.py) - Pre-download AraBERT encoder-decoder + checkpoint in Dockerfile - Lazy load on first request, contextual=False to save RAM - Update health check to report spelling status

    youssefreda9
    53f433b