Release v.6.1.0

Version 6.1.0 (2024-11-12)

Added

• Support for step reasoning ontology tool(#1894, #1882, #1879)

Fixed

• Use valid lbox.exception in client IAMIntegration validation(#1887)
• Improve SDK method name detection for Datadog dashboard(#1880)
• Limit labeling front end visibility and mentions of a custom editor(#1877)
• Remove all unused python imports(#1876)
• Remove bulk import request from readthedocs(#1881)

Release v.6.0.1

Version 6.0.1 (2024-10-30)

Fixed

• Updated labelbox.exception to lbox.exception for IAMIntegration check(#1887)

Release v.6.0.0

Version 6.0.0 (2024-10-15)

Added

• Deprecation messages to all export_v2 methods(#1865)
• Logic to extract sdk method to pass to datadog dashboard(#1865)
• Graphql client as a separate component lbox.request_client(#1834)

Updated

• Minimum requirement numpy version from 1.24 to 1.25 and cleaned up _TypedArray validations(#1845)
• Refactored client create_project inputs to be named parameters, not kwargs(#1844)
• Fixed pydantic v2 deprecation warnings(#1838)
• Remove experimental from UserGroup(#1840)
• Improved labeling service error reporting(#1836)
• Switched to ruff linter(#1822)

Fixed

• Labeling dashboard serialization (#1870)
• Handling of global key in ModelEvaluationTemplate(#1869)

Removed

• Various deprecated classes, methods and attributes(#1853)
  • QueueMode for project
  • LabelGenerator assign_feature_schema_ids
  • Label assign_feature_schema_ids
  • AttachmentType attachment_type TEXT
  • DataRowMetadata bulk_delete deletes paramerter - support for string for data row ids or global keys, use UniqueId or GlobalKey class instead
  • Dataset create_data_rows_sync. Use create_data_rows instead
  • Slice get_data_row_ids. Use get_data_row_identifiers instead
• More deprecations(#1852)
  • LabelingParameterOverrideInput remove instance of DataRow as a type of input
  • Project _update_queue_mode, get_queue_mode
• Support for Python 3.8(#1847)
• Project setup(#1843)
• Deprecated get stream - related classes from ExportTask(#1839)
  • JsonConverter, JsonConverterOutput, FileConverter, FileConverterOutput
• Project setup_editor(#1841)
• Label *Data* classes as Label data attribute. Use GenericDataRowData
  • NOTE MaskData was NOT removed
• COCO package(#1820)
• NDJsonConverter deserialize(#1818)
• BulkImportRequest package(#1821), including
  • BulkImportRequest
  • BulkImportRequestState
  • Project upload_annotations
  • Project bulk_import_requests

Release v.5.2.1

Version 5.2.1 (2024-10-09)

Fixed

• Exporter encoding

Release v.5.2.0

Version 5.2.0 (2024-10-09)

Fixed

• Support data row / batch for live mmc projects(#1856)

Release v.5.1.0

Version 5.1.0 (2024-09-27)

Fixed

• Support self-signed SSL certs(#1811)
• Rectangle units now correctly support percent inputs(#1848)

Release v.5.0.0

Version 5.0.0 (2024-09-16)

Updated

• Set tasks_remaining_count to None LabelingServiceDashboard if labeling has not started (#1817)
• Improve error messaging when creating LLM project with invalid dataset id parameter(#1799)

Removed

• BREAKING CHANGE SDK methods for exports v1(#1800)
• BREAKING CHANGE Unused labelbox_v1 serialization package(#1803)

Fixed

• Cuid dependencies that cause a crash if numpy is not installed (#1807)

Release v.4.0.0

Version 4.0.0 (2024-09-10)

Added

• BREAKING CHANGE for pydantic V1 users: Converted SDK to use pydantic V2(#1738)
• Automation test support for multiple sdk versions(#1792)

Fixed

• Flaky tests(#1793)

Release v.3.78.1

Version 3.78.1 (2024-09-10)

Fixed

• Labeling dashboard query for tags

Release v.3.78.0

Version 3.78.0 (2024-09-03)

Added

• Project get_labeling_service_dashboard() aka project detail (Miltiple PRs)
• Client get_labeling_service_dashboards() aka project list (Miltiple PRs)
• Client get_task_by_id()(#1767)
• Support for MMC tasks annotations(#1787)

Fixed

• Build a test-pypi sdk instance even if tests fail(#1774)