To view Product Release Notes for 2021 and beyond, please visit this page.
Workflows API (Beta)
Release Date: 5/6/20
Improvement: Once you've created and validated a workflow via UI, tap into the new API to take your automation with Workflows a step further. A brand new set of endpoints allow you to
- Upload Data (via CSV/JSON)
- Control & Poll Workflow Status
- Download Reports
- View Workflow Configurations
- Copy Workflows
Learn more here.
Filter API job listings by state, created_at, and tags
Release Date: 5/15/20
Improvement:
- With some customer job listings becoming increasingly large and difficult to manage, we created a way to filter the results. We've updated the job listing API endpoint to be filterable by state, created_at, and tag so customers can trim potentially dozens of pages of jobs down to exactly what they need.
- While we were updating API endpoints, we have also created endpoints to geotarget contributors and enable the rows remain finalized and rows complete in order settings.
Learn more here.
Image and Video Annotation Efficiency Updates
Release Date: 5/18/20
Improvement:
- We have updated the following features in Video and Image annotation jobs:
- Undo/Redo has been enabled for individual points on lines and polygons
- When zooming while drawing, the lines and points will change in size dynamically to scale with the image
- In Pixel Labeling jobs:
- Enabled add new points to a polygon annotation before rasterizing
Copy Job with unfinalized rows
Release Date: 5/22/20
Improvement:
- All users on the platform now have the ability to copy jobs with only unfinalized rows from the source job.
- This simplifies data management during the job iteration process as, previously, users could only copy jobs with either "All rows" or "No rows"
Learn more here.
CML:group and multiple="true" Aggregation
Release Date: 5/29/20
Improvement: CML:group and multiple=“true” now supports a new linked entity aggregation method for text fields in the aggregate report. Users can now parse through the JSON output to identify linked entities and determine the confidence of aggregate responses. This will improve data quality reporting and processing in complex use cases, such as PDF transcription and data collection.
Learn more here.