Change log

New Feature

  • Deploy General Detection Beta Model to recognize multiple objects with bounding boxes.

  • Create delete email endpoints in v2 to finally get off old internal endpoints to streamline operations.

  • Create Patch, Delete, Get CreditCards endpoint in v2 APIs to finally get off old internal endpoints to streamline operations.

  • Created display for scopes on collaborator invitations, allowing users to easily understand and control the scope of access allowed for app collaborators

  • Deployed new face detector for improved face detection performance over images and video

  • Created custom training enhancements that handle negatives better for improved model performance

  • Created evaluation metrics for custom facial recognition in backend for improved facial recognition performance

  • Introduced Collaborators and Collaborations endpoints in API and UIs in Portal

  • Return custom detection evaluations through the GO API

  • Custom detection training with new data-dump

  • Add ability to upload inputs from App Details screen in Portal

Improvement

  • Improved accuracy of annotation counts, improving the user experience when annotating inputs

  • Topological sort for workflows

  • Added helper text/suggestions to improve Portal user experience

  • Header Search return app_owner's user info in collaboration endpoints

  • Improved billing for collaborators

  • Created collaboration tab in Portal, making it easy to add collaborators to apps

  • Created display to show the user who invited you to collaborate on an app

  • Update email phrases for collaborator invitations

  • Cleaned up duplicate models in workflow model list

  • PATCH /inputs needs to check status of asset before patching

  • Removed sync DELETE /inputs after runtime config tested

  • Changed POST /inputs to be async always to simplify processing of workflows after API client tests updated

  • Deployed public general v1.5 in concept model

  • Create Pixel Training Hyperparameter Help Guide

  • Improved cluster page performance

  • Added pagination to clusters making for easier data management

Bug Fixes

  • Cleaned up duplicate models in the workflow model list, so that you no longer see two General models

  • Unintended behavior for private model version IDs for certain customers has been fixed

  • Sporadic inability to delete any inputs via Portal or in bulk via the API

  • Numerous third party security fixes under the hood during ongoing upgrades

  • Fix 40012 status caused by parallel deletes and adds having a race condition

  • Investigate health checks killing a prediction backend service, which could affect some predictions in the API

  • Update status_changed_at when deleting inputs so we can better track changes

  • Workflow predict sometimes was failing with 98012 status code. Many fixes here should reduce that

  • Models referencing deleted backends should be marked as deleted

  • Cache the input counts so that apps can display them in Portal efficiently

  • Handle killing URL downloading if it is processing for more than 60s. This will make URL processing much more reliable

  • Return an error if a user sends YouTube video URL as that is not a valid URL to a video we can download

  • Workflow Predict called the wrong model sometimes. Not any more!

  • The latest version of our general model wasn't always default, now it is

  • If an image is tagged with a concept that is not in the model, training fails due to KeyError, this is fixed

  • Video playback out of sync with detections in our demos

  • Fix detection labeling bug where previous images image ratio is used which would cause display issues

  • After successful sign-up, the user is not redirected to the app's dashboard in Portal

  • Explorer Search Bar - Clicking the green/red circle icons didn't reliably detect click, now it does!

  • We have updated Portal to scale to a large number of concepts with much lower resource usage

  • Fixed issues with regions predicted on inputs would be carried over between inputs in Portal

  • Fixed the flaky face recognition tests to ensure stability of our face recognition product

  • Face Detection backends were running out of memory for some predictions, this has been resolved

  • Investigate face bounding box probabilities consistency to improve user experience

  • Fixed a bug with face recognition evaluations.

  • Deleted Concepts Persisted in face recognition models

  • Prevent PostInputs from creating inputs with a user-provided Input.ID that contains a colon

  • Fixed issue with concept counts in some apps

  • Video calls failed if URLs contain parameters after the file type

  • Portal not showing the correct number of results in concept search

  • Failed to resolve DNS MX record in URL down-loader which effected some downloads

  • Investigate why some re-hosted s3 links are no longer working

  • Inability to see whether a large model is training and making progress, or hung has been addressed to better support our customers

  • Getting input counts was broken in some apps, reporting zero, which caused Portal to add an input view to display always

  • Debug UnicodeErrors in URL downloading to fix URLs with Unicode characters

  • Return more descriptive error msg for post metric endpoint

  • Fix the poor handling of video too large error message

  • PostVerifyEmail error causing some issues not being able to verify their email addresses upon sign-up

  • Fixed flaky email verification integration test to provide more stability to sign-up process

  • Unable to batch delete inputs from time to time has been fixed

  • Media processor video handling was having errors with decoding some videos

  • Bounding box creation canvas in Portal was breaking on resize of the window

  • Delete Image Button doesn't work in some scenarios

  • Left/right arrows in single image view don't switch between images with regions

  • Issues with properly consenting to GDPR not being visible in Portal and preventing usage have been fixed

  • Model won't train in some apps with no positive examples issue has been resolved

  • Fixed a link to a non-public version of our API used for development purposes which led to a lot of login issues for users who landed there

  • Fixed issues with color models failing for a short period of time

  • Fixed list of models available to workflows to only show a single General model

  • Fixed support for webp image format so it is available again

  • Clicking pencil icon to edit an API Key in Portal crashed your app

  • Fixed carousel thumbnail clicks wiping query params / trigger new search

Last updated