Skip to content

Releases: marl/openl3

v0.4.2

03 May 18:22
Compare
Choose a tag to compare

Release version 0.4.2 of OpenL3.

  • Fix incorrect embedding_size in load_image_embedding_model docstring
  • Add tensorflow.keras mock modules to docs/conf.py to fix docs build
  • Remove pin on sphinx version
  • Add note about training bug in README

v0.4.1

06 Aug 20:34
2b04817
Compare
Choose a tag to compare

Release version 0.4.1 of OpenL3.

  • Add librosa as an explicit dependency
  • Remove upper limit pinning for scikit-image dependency
  • Fix version number typo in README
  • Update TensorFlow information in README

v0.4.0

06 Aug 20:25
b3d924e
Compare
Choose a tag to compare

Release version 0.4.0 of OpenL3.

  • Upgraded to tensorflow>=2.0.0. Tensorflow is now included as a dependency because of dual CPU-GPU support.
  • Upgraded to kapre>=0.3.5. Reverted magnitude scaling method to match kapre<=0.1.4 as that's what the model was trained on.
  • Removed Python 2/3.5 support as they are not supported by Tensorflow 2 (and added 3.7 & 3.8)
  • Add librosa frontend, and allow frontend to be configurable between kapre and librosa
    • Added frontend='kapre' parameter to get_audio_embedding, process_audio_file, and load_audio_embedding_model
    • Added audio_frontend='kapre' parameter to process_video_file and the CLI
    • Added frontend='librosa' flag to load_audio_embedding_model for use with a librosa or other external frontend
    • Added a openl3.preprocess_audio function that computes the input features needed for each frontend
  • Model .h5 no longer have Kapre layers in them and are all importable from tf.keras
  • Made skimage and moviepy.video.io.VideoFileClip import VideoFileClip use lazy imports
  • Added new regression data for both Kapre 0.3.5 and Librosa
  • Parameterized some of the tests to reduce duplication
  • Added developer helpers for regression data, weight packaging, and .h5 file manipulation

v0.4.0rc2

30 May 02:19
b3d924e
Compare
Choose a tag to compare
v0.4.0rc2 Pre-release
Pre-release

v0.4.0rc2

v0.4.0rc1

30 May 02:10
70cfb0f
Compare
Choose a tag to compare
v0.4.0rc1 Pre-release
Pre-release

RC1, fix soundfile install

v0.4.0rc0

30 May 01:25
08b643d
Compare
Choose a tag to compare
v0.4.0rc0 Pre-release
Pre-release

RC0 for v0.4.0 which adds TF2 support and adds a librosa audio front end as a CPU alternative to the kapre front-end

v0.3.1

28 Feb 21:23
Compare
Choose a tag to compare

Release version 0.3.0 of OpenL3.

  • Require keras>=2.0.9,<2.3.0 in dependencies to avoid force installation of TF 2.x during pip installation.
  • Update README and installation docs to explicitly state that we do not yet support TF 2.x and to offer a working dependency combination.
  • Require kapre==0.1.4 in dependencies to avoid installing tensorflow>=1.14 which break regression tests.

v0.3.1rc0

28 Feb 21:10
Compare
Choose a tag to compare
v0.3.1rc0 Pre-release
Pre-release

Release candidate 0 of version 0.3.1.

  • Require keras>=2.0.9,<2.3.0 in dependencies to avoid force installation of TF 2.x during pip installation.
  • Update README and installation docs to explicitly state that we do not yet support TF 2.x and to offer a working dependency combination.
  • Require kapre==0.1.4 in dependencies to avoid installing tensorflow>=1.14 which break regression tests.

v0.3.0

23 Jan 20:53
Compare
Choose a tag to compare

Release version 0.3.0 of OpenL3.

  • Rename audio related embedding functions to indicate that they are specific to audio.
  • Add image embedding functionality to API and CLI.
  • Add video processing functionality to API and CLI.
  • Add batch processing functionality to API and CLI to more efficiently process multiple inputs.
  • Update documentation with new functionality.
  • Address build issues with updated dependencies.

v0.3.0rc0

23 Jan 20:08
Compare
Choose a tag to compare
v0.3.0rc0 Pre-release
Pre-release

Release candidate 0 of version 0.3.0.

  • Rename audio related embedding functions to indicate that they are specific to audio.
  • Add image embedding functionality to API and CLI.
  • Add video processing functionality to API and CLI.
  • Add batch processing functionality to API and CLI to more efficiently process multiple inputs.
  • Update documentation with new functionality.
  • Address build issues with updated dependencies.