New features
* Improve multiclass UX design by martinscooper and [natti](https://ibm-silvergate.slack.com/team/W4B3H000G) in https://github.com/label-sleuth/label-sleuth/pull/605
* Add SBERT representation and the SVM over SBERT model. Refactor memory and cache mechanism to support vectors by alonh in https://github.com/label-sleuth/label-sleuth/pull/626
* Add multiclass ensemble model by arielge in https://github.com/label-sleuth/label-sleuth/pull/595
* Update python version and dependencies by arielge in https://github.com/label-sleuth/label-sleuth/pull/624
* Add tutorial for multiclass workspaces by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/592, https://github.com/label-sleuth/label-sleuth/pull/593 and https://github.com/label-sleuth/label-sleuth/pull/594.
* Adapt force_train for multiclass workspaces by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/600
* Add percentage to panel filter dropdowns by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/619
* Split datasets with no document id into 10k parts by alonh in https://github.com/label-sleuth/label-sleuth/pull/622
Improvements
* UI text overflow improvements by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/601
* Adapt labeling status message and model checks conditions by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/613
Fixes
* Dataset delete checkbox remains checked after submit by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/591
* Map multiclass model scores to the corresponding class ids by arielge in https://github.com/label-sleuth/label-sleuth/pull/606
* Do not allow changing category name to a name that already exists by arielge in https://github.com/label-sleuth/label-sleuth/pull/599
* User can't edit category description or color without changing the category name by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/603
* User can bypass validation using shortcuts by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/610
* Don't select 'sort' text when double clicking on the sorting arrow by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/612
* Deleting workspaces fails if workspace isn't in memory by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/614
* Status call logic is wrong by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/615
* Create new iteration if there is an error in the should_train condition and fix test race condition by alonh in https://github.com/label-sleuth/label-sleuth/pull/617
* Fix force train test by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/620
* Bugfix in accessing `sorted_class_by_freq` when there's no training data by lenadankin in https://github.com/label-sleuth/label-sleuth/pull/616
* Escape with slash only as a fallback on import data. Convert input to string to handle float values by alonh in https://github.com/label-sleuth/label-sleuth/pull/621
* Percentage string is NaN is there are no labeled elements by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/625
* Training logic doesn't trigger new model iterations when appropriate by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/627 and https://github.com/label-sleuth/label-sleuth/pull/628
Chore
* Change create workspace parameter name by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/596
* Rename `/set_element_label` endpoint parameters and refactor by martinscooper in https://github.com/label-sleuth/label-sleuth/pull/618
**Full Changelog**: https://github.com/label-sleuth/label-sleuth/compare/v0.18.0...v0.19.0