Skip to content

GitLab

  • Menu
Projects Groups Snippets
    • Loading...
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in / Register
  • ketos ketos
  • Project information
    • Project information
    • Activity
    • Labels
    • Planning hierarchy
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 27
    • Issues 27
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 0
    • Merge requests 0
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Schedules
  • Deployments
    • Deployments
    • Environments
    • Releases
  • Monitor
    • Monitor
    • Incidents
  • Analytics
    • Analytics
    • Value stream
    • CI/CD
    • Repository
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar
  • public_projects
  • ketosketos
  • Merge requests
  • !243

Merged
Created Apr 15, 2022 by Oliver Kirsebom@kirsebomOwner

Ketos v2.6 (includes non-backward compatible changes)

  • Overview 4
  • Commits 47
  • Pipelines 39
  • Changes 23

Copy-paste from the Changelog file:

  • New :class:MultiModalBatchGen <ketos.data_handling.data_feeding.MultiModalBatchGen> class.
  • Fixed bug in :class:BatchGenerator <ketos.data_handling.data_feeding.BatchGenerator> related to loading of labels from separate annotation table.
  • :class:AudioSelectionLoader <ketos.audio.audio_loader.AudioSelectionLoader> ignores duration parameter in the audio representation dictionary. Also, the option to standardize selection window size across the selection table has been removed. This must now be done by the user before passing the selection table to the audio loader.
  • :meth:create_database <ketos.data_handling.database_interface.create_database> allows multiple audio representations to be specified as a nested dictionary, using keys as names for the data fields in the output table.
  • Changes to the interface and default behaviour of the :meth:standardize <ketos.data_handling.selection_table.standardize> method.
  • :class:AudioSelectionLoader <ketos.audio.audio_loader.AudioSelectionLoader> generalized to be able to handle new format for selection tables where a single selection may span several files.
  • Extended Python compatibility to <=3.9.12
  • Modified (and simplified) :meth:load_audio_for_spec <ketos.audio.spectrogram.load_audio_for_spec> so that it re-uses :meth:from_wav <ketos.audio.waveform.Waveform.from_wav>.
  • Added pad_mode argument to :meth:from_wav <ketos.audio.waveform.Waveform.from_wav> class method.
  • Changed default behaviour of :meth:num_samples <ketos.audio.utils.misc.pad_reflect> to not invert the signal
  • Minor bug fix in :meth:num_samples <ketos.audio.utils.misc.num_samples> for rounding time intervals to integer number of samples.
  • New method :meth:append <ketos.audio.waveform.merge> for stitching together multiple waveforms
  • :meth:append <ketos.audio.waveform.Waveform.append> preserves duration
  • `from_wav`` methods can load from multiple audio files
  • AudioLoader stores annotation start time and end time without truncation.

I believe this is now ready for review.

Edited May 01, 2022 by Oliver Kirsebom
Assignee
Assign to
Reviewer
Request review from
Time tracking
Source branch: sel_tbl_dev