Label Studio Frontend Versions Save

Introducing Time Series data labeling for classification, segmentation, forecasting, outlier/anomaly detection, and many more!

Read more on our introductory blog post. Play with predefined templates on /setup page. Import your data from CSV, TSV or JSON formatted files.

We put a big effort to rewrite some parts of the labeling engine for easier and better work with results, deserialization, an inner structure and so on. That led us to faster initialization (but we have a lot of space for improvements), more predictable regions operations, more stable per-region labeling, non-breaking region removal.

Features

• Relations visualizations
• Nested labeling improvements
• Granularity fixes - now you can enable granularity per label
• New deserialization mechanics for LSF
• LSF faster initialization
• More predictable regions operations
• More stable per-region labeling
• Non-breaking region removal

Improvements

• Show errors about content loading and CORS problems

Fixes

• Fixed dates in non-Chromium browsers
• Don't stretch out text object with long lines
• Don't save broken regions (polygons, brushes)
• Fixed per-region audio labeling with predictions

Finally new release!

Regions tree and Choices in dropdown

Tree-like view of regions for grouping and better semantic organization. You can drag'n'drop regions into another in the list to form a semantic groups like question-answer, person-actions and so on. Also overall appearance of regions and relations are improved.

Choices layout="select" for huge lists of choices added — they all now can be reduced into one useful dropdown; filtering included.

Image fixes and improvements

Weird bugs on zoomed images are fixed: moving, resizing, polygon modification — they all are good finally. Also creation of new regions works well on the edge of the image now. New rotatecontrol="yes" added to rotate all the canvas instead of your head while trying to get the whole picture.

Almost all warnings and errors are fixed, so your console won't be flooded. Other issues were fixed as well, especially related to NER.

Important fixes for Image and NER labelling

Simple Rectangle, Polygon, KeyPoint or Ellipse tag now works perfect for fast Image labelling without any labels — click and go.

Weird and annoying bugs of text labelling were fixed. They all got restored except the last ones, they were just disappeared:

• broken regions with newlines,
• broken overlapping regions,
• broken regions with special characters,
• broken lengths of regions with whitespaces,
• broken empty regions.

Side panel improved to display more relevant information about regions and relations:

• captured text part is longer now,
• relations highlighted when any of regions in it are selected,
• also relations display number and color of both regions to distinguish between them.

All this covered by tests!

Main updates in this release:

Different shapes on the same image

One limitation label studio had was the ability to use only one shape on the same image, for example, you were able to put either bounding boxes or polygons. Now this limitation is waived and you can define different label groups and connect those to the same image.

maxUsages

There are a couple of ways how you can make sure that the annotation is being performed in full. One of these concepts is a required flag, and we’ve created a new one called maxUsages. For some datasets you know how much objects of a particular type there is, therefore you can limit the usage of specific labels.

E2E tests introduced

There are couple of integration tests already added powered by codecept. We'll use them actively to check new features and provide automatic regression testing for existed functionality.

Bugfixes and Enhancements

Allow to use simple Rectangle or Polygon on the Image tag Fixing double text deserialization #85 Fix bug with groups of required choices #74 Several fixes for NER labeling — empty captured text, double clicks, labels appearance

Check out our blog post about this update and its new features. The most essentials are:

Nested Labeling

Nested labeling enables you to specify multiple classification options that show up after you’ve selected a connected parent class.

It can match based on the selected Choice or Label value, and works with a required attribute too, smart selecting the region that you’ve not labeled. To try it out check Choices documentation and look for the following attributes: visibleWhen, whenTagName, whenLabelValue, whenChoiceValue.

Per region labeling

With per region labeling you can now provide additional attributes to the labeled regions. For example, when doing audio segmentation you can further classify the region. Per region is available for any data type and the following control tags: Choices, TextArea, and Rating.

It nicely integrates with the nested labeling, for example, you can provide multiple levels of classification for any particular region.

The frontend code now lives in its repository, that's quite an upgrade! Imaging moving out of the dorm room and renting out your first private place. That's the kind of excitement we feel for the frontend.

Besides numerous bug fixes, refactorings, and code cleanups, this release makes it very easy to embed the labeling interface into your applications. NPM package stays the same and can be installed by executing

npm install label-studio

Check out the updated README file to include files from the CDN. Get up and running in minutes. Yay!