June 30, 2020

Release 6.3 Changelog

Table of Contents:
Introducing important new features and improvements with Changelog 6.3

New Feature

Improvement

Bug Fix

Enterprise Only

White Letter N in Box icon

White Letter I in Box icon

White Letter B in Box icon

White Letter E in Box icon

 

Status

Details

White Letter I in Box icon

Send collaborator emails asynchronously

White Letter B in Box icon

NLP bug fixes for non-text apps

 

Status

Details

White Letter I in Box icon

Consolidated input related status codes

White Letter I in Box icon

Add frame.id to API as well as region.track_id

White Letter B in Box icon

Granted select permission to clarifairead

 

Status

Details

White Letter N in Box icon

Added list annotations filter status

White Letter I in Box icon

Added concept selection for tasks

White Letter I in Box icon

Post/Patch annotations request now allow setting status

White Letter I in Box icon

Changed task form options

White Letter I in Box icon

Set annotation status to awaiting for review if the authorized user is not app owner

White Letter I in Box icon

Return only input_level annotation in input.data

White Letter B in Box icon

Drawing annotations: wrong embed model version id. Fixed

White Letter B in Box icon

Eliminated error if no annotation to be deleted

White Letter B in Box icon

Create one annotation for each bbox

 

Status

Details

White Letter N in Box icon

Added support for adding and training on text in the platform

White Letter N in Box icon

Created a NLP mock prediction endpoint

White Letter I in Box icon

Created test set to evaluate quick trained models or k-fold if no test search is specified

White Letter I in Box icon

Added vocab_id for demographics model concepts

White Letter I in Box icon

Fixed sorting of A.G.E. concepts in golang for demographics model so we don't chop off sets of them

White Letter I in Box icon

Deprecated Face from javascript Client

White Letter I in Box icon

Deprecated Face from Java Client

White Letter B in Box icon

Confusion matrix predicted/true are swapped in evaluation results. Fixed

White Letter B in Box icon

Explorer Image/Text Joint embedding

White Letter B in Box icon

Fixed selectEmbedModelVersionId in detection apps

White Letter B in Box icon

Fixed generalModel imports and optimized video click handlers with useCallback hooks

 

Status

Details

White Letter I in Box icon

Persisted the saved search used in train a model version

White Letter B in Box icon

Created log for annotation/search request/response

White Letter B in Box icon

Region Searches within Search Bar still use crop coordinates instead of base64 bytes. Fixed

 

Status

Details

White Letter N in Box icon

Created new Single Image View and Image Tools

White Letter N in Box icon

Enabled Display Text Thumbnails in App Grid View and App Details View

White Letter N in Box icon

Text Thumbnails display in Portal/Search Bar disabled

White Letter N in Box icon

Enabled View Text Assets in Portal's Image View

White Letter N in Box icon

Added Text Inputs To Explorer Apps

White Letter N in Box icon

Imported new icons for Labeler Image Tools into the style guide

White Letter I in Box icon

Added login tracking to analytics package in Portal

White Letter I in Box icon

Allowed pasting into the add inputs text area and clear the text box after clicking submit

White Letter B in Box icon

Search bar not visible. Fixed

White Letter B in Box icon

Removed all instances of worker_id from Explorer

White Letter B in Box icon

Fixed popover left/right overflow

White Letter B in Box icon

Disabled all search by click handlers in Portal for Text Apps

White Letter B in Box icon

Click Search button icons on Thumbs not working for localized search. Fixed

White Letter B in Box icon

Fixed details page header missing description

White Letter B in Box icon

Fixed demo font syntax

 

Status

Details

White Letter N in Box icon

Added a Range Slider to filter Workflow Predictions by Value

White Letter I in Box icon

Updated Face workflow to include the detect faces as concepts for search