| description |
|---|
Changelog for Clarifai Release 7.1 |
| New Feature | Improvement | Bug Fix | Enterprise Only |
|---|---|---|---|
 |
 |
 |
 |
| Status | Details |
|---|---|
|
Add apps and keys scopes so they can be created with personal access tokens. |
|
Change /keys to work with PATs |
|
Change /apps to work with personal access tokens |
|
Split java proto files into multiple and add package name. |
|
Write script to prune the proto files into public only rpcs and messages |
|
Mark the appropriate fields in protos as cl_private_rpc to release grpc clients. |
|
copy app count and last_inputs added in app duplication |
|
Rewrite input counting in the API to be more scalable and robust. |
|
Fix collector scopes so that predict keys don't need Collectors:Get |
|
Check the GetInputsKey of collectors has access to userA's information. |
|
Return “All” scopes when listing available scopes so that you have that option when creating new keys. |
|
properly return err if AddAssets failed to insert into DB |
|
POST inputs wasn’t using batch model optimizations correctly. |
|
created_at field in sharing table is incorrect |
|
Missing Apps_Get scope in session token auth caused creation of keys to fail temporarily. |
|
API services do not function once Queue goes down and comes back up has been fixed. This makes on premise deployments more resilient to power failures. |
|
List of missing scopes is not correct in error messages |
|
Fix node ID validation logic in Bug in workflows |
| Status | Details |
|---|---|
|
Create endpoint for taking down a spire |
|
Create endpoint to deploy deep training models |
|
Calculate detection pr and roc curves using score buckets |
|
Support for pytorch inference in spire |
|
able to overwrite default max conn for Citus |
|
Upgrade to go version for performance boost. |
|
Make runtime config to remote the extra round trip to storage in predict pathway. |
|
Improve JSON serialization performance in our servers by using an optimized third party library. |
|
Fix the WorkflowInput field name in proto to workflow_input |
|
When reaching the final page, network request responds with 500 internal service error. Fixed |
|
Spire fails to launch in local-k8s-USER with error "persistentvolumeclaim not found". Fixed |
|
Video processing fails with 'caseids' error. Fixed |
|
Fix a connection issue from Golang backend service to media processing service |
|
Allocated resources for faster model performance |
| Status | Details |
|---|---|
|
Return deep training evals through the API |
|
OCR Model |
|
Update templates to have more straightforward names and more friendly defaults |
|
Add 'Face' Default workflow to `https://api-dev.clarifai.com' |
|
Make custom/transfer training evaluations for large models stable. |
|
training progress is saved too frequently, causing very slow training |
|
Return friendlier errors for incorrect parameters passed to templates |
|
Fix a bug in tracing setup for custom trainer and evaluator |
|
Accellerated training for specific cases |
|
Training System failed to train some layers |
| Status | Details |
|---|---|
|
Implement image crop model |
|
Implement RandomSample model type |
|
Allow models that need outputs from previous nodes in a workflow to have access to those outputs to support chaining complex graphs of models. |
| Status | Details |
|---|---|
|
make UI for personal access token |
|
Send event to Hubspot when a user signs up on portal |
|
Change HTML tag of ImagePile component from '' to '' |
|
Optimize Video Detection Frame Rate on Front end |
|
Remove classification/detection toggle in image details view |
|
Add colors to differentiate region results |
|
Scroll active thumb into view in image details carousel |
|
Unable to upload same file(s) through browse files. Fixed |
|
Added a favicon for Portal |
|
Can’t create a new API key or edit the information of API key. Fixed |
|
Delete input while having other inputs selected deselects everything. Fixed |
|
Cannot view workflow results in a face app. Fixed |
|
Setting useCustomConfig isn't checked at login. Fixed |
|
ffmpeg can produce no frames for very short videos. Fixed |
|
Video spire tests are not running correctly. Fixed |
|
Verify your email |
|
Last concept used for bounding boxes is retained between apps. Fixed |
|
Intercom links to old community site (and maybe old FAQ page). Fixed |
|
Add Inputs/View Explorer does not display in new app anymore |
|
Clicking video thumbs in detail view does not reload a video |
|
The Add Positives / Add Negatives buttons on a Concept details view breaks portal |
|
Collaboration apps have race condition where wrong user id is used |
|
Don't load collaborations for search demo/logged-out users |
|
Keyboard navigation in image details view highlights incorrect thumb |
|
Render Video Assets in Search Bar |
|
AppDetailsPanel add inputs/view in explorer no longer displays in devel |
|
Editing geo/json search items no longer work after adding the search bar tooltip |
|
TypeError: Cannot read 'get' of undefined when clicking image thumbnails in explorer search bar |
|
Explorer Visibility in small resolution screen |
|
No Prompt when uploading an image to explorer through url |
|
CFR rectangles on grid view do not correlate |
|
Prevent users from evaluating models that are not trainable. |
| Status | Details |
|---|---|
|
Mark the /annotation endpoints with cl_private_rpc |
|
No longer able to copy an app that has been shared with you via Collaborators. Fixed |
|
Collaborators can not see workers |
| Status | Details |
|---|---|
|
Implement visual search in another app as a model type you can add to a workflow. |
|
Add click to search metadata attributes in image details sidebar |
|
Ability to keep concepts sorted by alpha. |
|
Refactor search |
|
Add metadata to collector added inputs so that you can filter by collector ID |