cor-asv-ann e919020..1b8c120
Release: v0.1.12
- 📦 0.1.12
- wrapper.evaluate: allow aggregating histograms, add parameter
- scripts.compare: allow aggregating histograms
- lib.seq2seq / scripts.eval: allow aggregating histograms
- lib.alignment: allow aggregating histograms
- wrapper.evaluate: rename reported metrics and measure WER, too; expose gt_level parameter
- scripts.compare: rename reported metrics and measure WER, too
- lib.alignment: apply equivalences when aggregating confusion, too
- lib.alignment: more level 1/2 equivalences/normalizations
- lib.alignment: remove commented experimental code for other aligners
- 📦 0.1.11
- update README
- lib.alignment: no gaps for confusion counts
core d2c6216f..271dcdf6
Release: v2.25.1
- Merge pull request #701 from OCR-D/apidocs-standalone
- 📦 v2.25.1
- Merge remote-tracking branch 'origin/gds-id'
- 📦 v2.25.0
- 📝 changelog
- Merge remote-tracking branch 'bertsky/ocrd_page_etree'
- 📝 changelog
- Merge remote-tracking branch 'bertsky/fix-recrop'
- 📝 changelog
- Merge branch 'gds-id'
- 📝 changelog
- fix METS schemaLocation, fix #694
- 📦 v2.24.0
- 📝 changelog
- test for #686
- image_from_*: yield matching AlternativeImage with richest features, not last
- crop_image/image_from_polygon: use mask for background estimation, too
dinglehopper 0b9af3a..2497876
- Merge branch 'master' of github.com:qurator-spk/dinglehopper
- 🐛 dinglehopper: Call initLogging before logging
eynollah 80795c9..d784202
Release: v0.0.7
ocrd_calamari da98713..34013dd
Release: v1.0.3
ocrd_cis 426a33d..6f8a612
Release: v0.1.5
- ocrolib.morph: fix select_regions dtype for just 1 label
- segment/polygon_for_parent: skip segment if polygon cannot be made valid
ocrd_pc_segmentation 413f5d3..ff7c007
Release: v0.1.3
- rename ocrd-{pixelclasifier,pc} in code
- Remove extras_require from setup.py, no longer needed
- Remove extra flags for GPU use (see ocr-d-modul-2-segmentierung/page-segmentation#4)
- ocrd-tool.json: ocrd-{pixelclassifier,pc}-segmentation
ocrd_segment 5fd7254..34c72c0
Release: v0.1.11
- 📦 0.1.11
- extract_pages: fix 5c548b1a (BorderType has no id)
ocrd_tesserocr b755b26..c9a27bd
Release: v0.13.4
- 📦 0.13.4
- Merge pull request #179 from bertsky/improve-tessapi-reset
- recognize: fix 318cefd5 (inverted ChoiceIterator conf threshold)
- recognize: improve Tesseract API state wrapper:
- typo
- 📦 0.13.3
- recognize: on glyph level, fall back to RIL.SYMBOL of ChoiceIterator is empty
- 📦 0.13.2
- forgot requirements
- Merge pull request #175 from bertsky/osd-line-recognize-load-script-lang
- 📦 0.13.0
- Merge pull request #177 from bertsky/clipped-page-image
opencv-python fd4e604..86c3d2a
Release: 56
- Merge pull request #500 from williamjacksn/upload-sdist-to-pypi
- Merge pull request #510 from asenyaev/asen/master_update
- Merge pull request #507 from asenyaev/asen/macos_m1
- Merge pull request #483 from asenyaev/asen/arm_build
- Merge pull request #479 from asmorkalov/as/docs_update
- Fixed if condition for release job (#478)
- make it possible to compile from sources against numpy on macOS aarch64 (#475)
- Update opencv to 4.5.2
- Add GitHub actions setup for CI instead of Travis & Appveyour (where possible) (#470)
sbb_textline_detector f1d2777..325389a
- Update README.md
workflow-configuration cf51eb1..f1a858a
- make server: adapt to new multiprocessing options (num workers, per-page timeout)
- detect existing workspaces with arbitrary recursive depth and behind symlinks