Skip to content

Releases: Labbeti/conette-audio-captioning

Version 0.3.2

13 Dec 13:47
Compare
Choose a tag to compare

[0.3.2] 2024-12-13

Added

  • Add MIT LICENCE file.

Changed

  • Update README.md and CITATION.cff file.

Version 0.3.1

10 Jun 14:57
Compare
Choose a tag to compare

[0.3.1] 2024-06-10

Fixed

  • Remove duplicated files in results folder.
  • CNN10 and CNN14D pre-process transform instantiation.

Version 0.3.0

19 Apr 07:34
Compare
Choose a tag to compare

[0.3.0] 2024-04-18

Changed

  • Update dependencies with torchoutil, and clean a lot of dead code.
  • Convnext model (BL version) is now automatically selected during HDF preparation. (#5)

Fixed

  • Requirements versions specified during installation. (#4)
  • Preparation now correctly loads scale_layers in ConvNext during preparation. (#5)
  • ConvNeXt 75 iterations is now loaded instead of ConvNeXt 70 iterations for CoNeTTE.

Version 0.2.2

15 Jan 14:08
Compare
Choose a tag to compare

[0.2.2] 2024-01-15

Added

  • Multiple candidates, predictions and probabilities in model outputs.
  • train_and_enable_grad method in CoNeTTEModel class.

Changed

  • Rename eval_and_detach to eval_and_disable_grad in CoNeTTEModel class.

Version 0.2.1

12 Jan 17:27
Compare
Choose a tag to compare

[0.2.1] 2024-01-12

Added

  • conette-predict now support CNext-trans (baseline) model.

Version 0.2.0

12 Jan 16:44
Compare
Choose a tag to compare

[0.2.0] 2024-01-12

Added

  • CoNeTTE training source code, with entire data processing.
  • ConvNeXt-trans baseline training source code, with entire data processing.
  • ConvNeXt tag logits to CoNeTTE model outputs during inference.

Version 0.1.4

20 Nov 16:23
Compare
Choose a tag to compare

[0.1.4] 2023-11-20

Fixed

  • Fix forbid repetition mode argument.

Version 0.1.3

20 Nov 14:18
Compare
Choose a tag to compare

[0.1.3] 2023-11-20

Added

  • Forbid repetition mode argument to LightningModule and HuggingFace wrapper.

Version 0.1.2

17 Nov 12:48
Compare
Choose a tag to compare

[0.1.2] 2023-11-17

Fixed

  • Task embeddings inputs wavcaps_audioset_sl and wavcaps_bbc_sound_effects.

Version 0.1.1

09 Nov 17:43
Compare
Choose a tag to compare

[0.1.1] 2023-11-09

Added

  • Unittests for hf model.

Fixed

  • Fix sample path for PyPI package.