github HumanSignal/label-studio 1.12.0
on GitHub

latest release: 1.12.0.post0
21 days ago

🌟 What's New

🎉 New Features

🚀 Improved Machine Learning & LLM integrations

This release streamlines the way ML models and LLMs are connected to Label Studio with a focus on security and simplified user experience. (#5570)

Using the powerful ML backend integration, users can add models and customize automated workflows for:

  • Pre-labeling: Letting ML/AI models predict labels autonomously, which can then be reviewed by human annotators.
  • Interactive labeling: An automated process that applies initial labels to data, which are then refined through manual review, enhancing the efficiency and accuracy of data annotation for machine learning models.
  • Model evaluation and fine-tuning: Used in models like the Segment Anything Model (SAM), involves a human-in-the-loop approach where the model provides initial predictions or annotations, and a human annotator interacts directly with these predictions to correct or refine them.

Label Studio users can add custom models, or reference a new examples library to connect popular models, including Segment Anything, OpenAI, GroundingDINO, select Hugging Face models, Tesseract, and more.

Updates to the ML backend integration in this release include:

  • New support for basic auth, which means users can now connect to hosted ML backends that require a password.
  • The ability to specify additional parameters to pass to the model, which means users can now easily connect to Azure-hosted OpenAI in addition to OpenAI and popular ML models.
  • An improved UI and simplified project settings, including:
    • A new Predictions section under the Annotation page of the project settings makes it easier select whether you want to use predictions or a model in your annotation workflow.
    • A new Predictions page, where you can easily upload and manage predictions.
    • Removed obsolete settings that are no longer compatible (for example, auto-updating version).
    • Fixed various usability issues related to the annotation experience with a model connected. (#5621), (#5677), (#5678), (#5697)

For more information, see Integrate Label Studio into your machine learning pipeline.

screenshot of the Connect Model window

Screenshot of connected model

✂️ Remove Duplicated Tasks action

There is a new Remove Duplicated Tasks action available from the Data Manager. This action had previously only been available as an experimental feature. (#5572)

When you use this action, annotations from duplicated tasks are consolidated into one task (the first task found among the duplicated tasks).

Remove duplicated tasks option

🔄 Reset Cache action

There is a new Reset Cache action available from project settings under the Danger Zone section. You can use this action to reset and recalculate the labeling cache.

This action is particularly useful for situations in which you are attempting to modify the labeling configuration and you receive a validation error pointing to non-existent labels or drafts. (#5432)

Reset Cache

✅ Enhancements

  • Added support for X-Api-Key: <token> as an alternative to Authentication: Token <token>. This will make it easier to use API keys when integrating with cloud-based services. Special thanks to community member @mc-lp for making this feature request. (#5562)

  • Made several updates to the UI to improve performance and user experience, including:

    • Better formatting for longer text strings when using the grid view in the Data Manager. (#5369)

    • When applying regions to text spans, that text will appear next to region when listed in the Regions pane of the labeling interface. (#5323)

🐞 Bug Fixes

  • Fixed an issue where users could not submit annotations if the labeling configuration included the TextArea tag with the required and skipDuplicates parameters.  (#5490)

  • Fixed an issue where an empty draft was created every time a user click View all annotations. (#5417)

  • Fixed an issue with duplicate default hotkeys when working with multi-image segmentation. (#5297)

  • Fixed an issue where the Data Manager was not displaying false or 0 values as expected. (#5367)

  • Fixed an issue with the Number tag where toName was not validated. (#5408)

  • Fixed an issue with the Number tag in which the max constraint was not working. (#5220)

  • Fixed an issue where users could not use the Magic Wand tool with image preloading enabled. (#5385)

  • Fixed an issue where users were not shown a confirmation message after clicking Submit and exit in the label stream. (#5362)

  • Fixed an issue where the text next to the project pagination drop-down was not displaying properly. (#5363)

  • Fixed an issue where the toolbar in the labeling interface wasn’t properly positioned. (#5448)

  • Fixed an issue where Google Cloud Logging was not working due to a missing dependency. Special thanks to community member @tsr10 for reporting this issue.  (#5605)

  • Fixed an issue where imports were failing in cases where a user switched between organizations. (#5603)

  • Fixed an issue where users where seeing an n is not a function error when using a when configuring project settings. (#5449)

  • Fixed an issue where /api/version was not reporting all updates. (#5535)

🤩 Contributors

Full Changelog: tags/1.11.0...1.12.0
This changelog was updated in response to a push of b2fb581 Workflow run

Jira Release not found
Aha! Release 1.12.0
Release Notes are generated based on git log: No tasks found in Task Tracker.

Turned off Feature Flags (83)

ALL LINES STARTING FROM QUOTE WILL BE IGNORED

Check out latest releases or
releases around HumanSignal/label-studio 1.12.0

Don't miss a new label-studio release

NewReleases is sending notifications on new releases.

Get notifications