January 3, 2022

Release notes

App

Added

  • Video segmentation is now available in closed beta. To request access to this feature, sign up here.
  • You can now hold shift+click to select data rows in the Catalog tab - allowing you to select data more quickly.
  • You can now filter classifications and object subclassifications in the Catalog.
  • Quickly find your projects using the new search functionality in the Projects tab.

Changed

  • The annotation table is now sorted in descending order by count.
  • You can now do the following in the Data Rows tab:
    • Hide data rows from labelers (for sensitive data)
    • Bulk approve or reject Data Rows from the table. This is especially useful when using the grid view.
    • See status of Consensus Data Rows. You will see an “ongoing” status when there are labels submitted but not all labels have been submitted yet.
    • Filter for nested answers

Python SDK

Our latest version of the Python SDK is v3.11.0. See our full changelog in Github for more details on what was added recently.

Fixed

  • Dataset.create_data_rows() now accepts an iterable of data row information instead of a list
  • project.upsert_instructions()
    • now only supports pdfs since that is what the editor requires
    • There was a bug that could cause this to modify the project ontology
  • Nested checklists can have extra brackets. This would cause the annotation type converter to break.

Removed

  • DataRowMetadataSchema.id use DataRowMetadataSchema.uid instead
  • ModelRun.delete_annotation_groups() use ModelRun.delete_model_run_data_rows() instead
  • ModelRun.annotation_groups() use ModelRun.model_run_data_rows() instead

Added

  • AnnotationImport.wait_until_done() accepts a show_progress param. This is set to False by default.
    • If enabled, a tqdm progress bar will indicate the import progress.
    • This works for all classes that inherit from AnnotationImport: LabelImport, MALPredictionImport, MEAPredictionImport
    • This is not support for BulkImportRequest (which will eventually be replaced by MALPredictionImport)
  • Option.label and Option.value can now be set independently
  • ClassificationAnswers now support a new keyframe field for videos.
  • New LBV1Label.media_type field. This is a placeholder for future backend changes.