Skip to content

Latest commit

 

History

History
91 lines (72 loc) · 2.53 KB

CHANGELOG.md

File metadata and controls

91 lines (72 loc) · 2.53 KB

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

0.1.8 - 2023-06-29

Fixed

  • workarounds for broken models (DocBank_X101, Jambo-sudo_X101)
  • make deps: add explicit reqs prior to pip step with Torch index
  • set pc:PcGts/@pcGtsId from mets:file/@ID

Added

  • CI for CLI tests (with cached models and stored result artifacts)

Changed

  • migrated model URLs from external to Github release assets

0.1.7 - 2023-03-20

Fixed

  • adapt to Numpy 1.24 (no np.bool)

Added

  • model by Jambo-sudo (PubLayNet+custom GT)
  • model by LayoutParser (PRImA Layout GT)
  • CLI tests

0.1.6 - 2023-03-10

Fixed

  • avoid colon in generated region IDs
  • make deps: add explicit deps for torch
  • fix/update file resources
  • fix model config base paths on-the-fly

Added

  • add Psarpei TD model

0.1.5 - 2023-01-15

Fixed

  • param debug_img: 1 image per page
  • URLs/specs for PubLayNet/JPLeoRX models

0.1.4 - 2022-12-02

Added

  • param postprocessing (select steps, including none)
  • param debug_img (styles to visualise raw predictions, including none)

0.1.3 - 2022-11-02

Fixed

  • make deps: fall back to Detectron2 src build

Changed

  • added various models as file resources
  • added corresponding preset files
  • updated documentation

0.1.2 - 2022-10-27

Fixed

  • make deps: fix CUDA detection even more
  • apply device param as passed

Changed

  • downscale images to no more than 150 DPI for prediction (for speed)
  • add param operation_level (default page), add table mode

0.1.1 - 2022-02-02

Fixed

  • make deps: fix CUDA detection and allow CPU as fallback

Changed

  • instance segmentation postprocessing: use asymmetric overlap criterion for non-maximum suppression
  • skip instances which belong to classes with empty category
  • annotate incrementally (by skipping candidates that overlap with pre-existing top-level regions)

0.1.0 - 2022-01-21