Introducing Label Distribution Charts for Label Groups and User Soft Delete
Sean Lynch
The newest version of Label Studio Enterprise includes two updates that provide granular visibility into outliers and reduce security risks from churned employees: label distribution charts for label groups and user soft delete. Read on to learn more about these new features!
Label distribution charts make it easy to understand your datasets' composition and diversity at a glance. They also help you quickly identify outliers and imbalances, inform resampling strategies, provide insights for feature engineering, and more. This release includes both label distribution donut charts for label groups and the ability to reorder charts and KPIs to match your desired configuration.
Now, when a labeling configuration contains multiple label groups, you can view a donut chart for each group of labels in the labeling configuration. This new view shows both pure label counts and percentages of the whole. For each new label group, a chart is provided. This can be found by simply clicking on the chart dropdown and selecting “Summary,” and is in addition to the existing time series label distribution charts on the project performance dashboard page.
The project performance dashboard itself is also more performant than ever, as the data and associated charts aren’t loaded until you scroll them onto your screen. This new feature provides more granular visibility into outliers for the selected period using preset or user-defined time ranges.
Also included in this release is the ability to reorder charts and KPIs at the top of the project dashboard, with your changes and preferences automatically saved as you go. This allows technical leads and project managers to create the perfect setup based on their distinct preferences or needs concerning data representation. It also allows them to prioritize information based on its relevance to decision-making and position the most critical data upfront, ensuring it receives the necessary attention.
Furthermore, this update makes it easier to adapt and adjust your dashboard as project priorities change and better align it to meet individual, project, or departmental objectives related to your training data.
With this release, owners and administrators can now delete user accounts via UI or API while preserving all associated annotations, activities, and artifacts. Unlike deactivating a user account, a soft delete removes the user account entirely, and they cannot be reactivated or assigned to tasks at a later time. This helps owners and admins differentiate between accounts that could pose a future security risk, such as former employees, and those that simply aren’t needed at the moment but may be in the future, such as a trusted freelance annotator between assigned projects.
Deleted users are automatically logged out, anonymized, and don’t appear in the regular controls for user management. Deleted users can be added back into the organization (reinstated) but must be formally invited through the usual process.
This reduces overall security risks because access is immediately rescinded, and it eliminates the possibility of accidentally reactivating the account of a churned employee or external group. It also reduces mental fatigue and streamlines user management workflows by removing the need to sift through long mixed lists of active and inactive (deactivated) users.
And that’s all for today! At HumanSignal, we’re committed to providing you with all the tools and features you need to make your data labeling, fine-tuning, and model validation workflows as efficient and effective as possible. With these latest improvements to Label Studio Enterprise, we’re confident that you’ll be able to create a data view that works best for you and have an even easier time with user management.
To learn more about this and other recent releases, check out our docs. If you’d like to learn more about Label Studio Enterprise or get a demo of these new features, our expert team of humans would be happy to chat with you.
Large Language Models (LLMs) are revolutionizing how we tackle data annotation, offering speed and efficiency that were previously unattainable. With Label Studio's Prompts, you can now handle multiple natural language processing (NLP) annotation types for a single data task in a unified Label Studio workflow.
Annotator Performance Dashboards make it easer to manage labeling teams at scale and orchestrate all the internal and external resources needed to improve data quality, without sacrificing speed.
In this post, we’ll guide you through the process of using Prompts in Label Studio Enterprise to pre-annotate data for Named Entity Recognition (NER) tasks.