Skip to content

Releases: docling-project/docling-ibm-models

v3.2.0

21 Jan 15:04

Choose a tag to compare

Feature

v3.1.2

10 Jan 12:41

Choose a tag to compare

Fix

  • Use old transformers version with old torch version (#70) (b3e072e)

v3.1.1

09 Jan 12:53

Choose a tag to compare

Fix

  • Force numpy < 2.0.0 on mac intel (#69) (7f9365f)

v3.1.0

13 Dec 13:24

Choose a tag to compare

Feature

  • Add arguments for LayoutPredictor (#66) (fe6a476)

v3.0.0

11 Dec 12:06

Choose a tag to compare

Feature

  • New API for models initialization with accelerators parameters. Use HF implementation for LayoutPredictor. Migrate models to safetensors format. (#50) (04295b2)

Breaking

  • New API for models initialization with accelerators parameters. Use HF implementation for LayoutPredictor. Migrate models to safetensors format. (#50) (04295b2)

v2.0.8

11 Dec 11:53

Choose a tag to compare

Fix

v2.0.7

02 Dec 16:00

Choose a tag to compare

Fix

  • Improve numpy compatibility pinning (#57) (de2f241)

v2.0.6

20 Nov 12:21

Choose a tag to compare

Fix

v2.0.5

20 Nov 08:46

Choose a tag to compare

Fix

  • Removing dependency from mean_average_precision package (not in use) (#53) (65affef)

v2.0.4

18 Nov 16:38

Choose a tag to compare

Fix