Codex_Extractor / smoke_signal

Commit History

  1. feat: Gill Sans Infant Tesseract model (BCER 0.071%) + font OTF via LFS

    Claude
    ec14162
  2. chore: track *.traineddata via LFS + Gill Sans font->model mapping

    Claude
    57311b9
  3. fix: gruffalo lowercase, remove dangerous so->Go word correction

    Claude
    5d7e317
  4. fix: ocr_done state triggers review load after OCR completes; add Gruffalo word corrections

    Claude
    4233f08
  5. feat: prev button, step instructions all tabs, noise threshold 3.2, word corrections I->T G->s

    Claude
    c481e30
  6. Add persistent book page scope, safe titles, and OCR stall guards

    Pointf5ive
    7152efa
  7. Upload tessdata/gill_sans_infant.traineddata.otf

    Verified signature
    d109898

  1. Integrate punctuation corrector into OCR routing and review learning

    Pointf5ive
    4ac6999
  2. Add Smoke Signal diagnostic audit trail and bug report template

    Pointf5ive
    9f024f9
  3. Fix Smoke Signal Surya OCR for new predictor API

    Pointf5ive
    be1ecc2
  4. Training phase: scripts 09+10, --model flag in 03, gold schema patch

    Pointf5ive
    5ffc645
  5. Stages 9+10: Codex exporter and validation report

    Pointf5ive
    bad810b
  6. Stages 7+8: LLM normalisation and review workbench

    Pointf5ive
    46d8f34
  7. test push

    Pointf5ive
    bfe9493
  8. Upload 04_region_detector.py

    Verified signature
    ddb6b3c
  9. Update smoke_signal/scripts/03_ocr_bakeoff.py

    Verified signature
    ace839d
  10. Create 03_ocr_bakeoff.py

    Verified signature
    a3735e7
  11. Create 02_profile_pdfs.py

    Verified signature
    d4ee6c9
  12. Upload 13 files

    Verified signature
    532d429