Corban Beaird

Results 10 issues of Corban Beaird

## Description OSS-specific adjustments to ensure that GetPermissionChangeListener() functions as expected. ## Test Plan - Integration Test on EE side for relevant feature ## Commentary (optional) ## Checklist - [...

cla-signed

## Description Originally: Adds GET endpoint for URL that points to the pachyderm input repository (if configured) associated with provided experiment. This will be used to enable basic lineage views....

cla-signed

## Ticket ET-48 ## Description Add support for storing arbitrary metadata associated with a given run. Additionally, this sets the groundwork for filtering runs based on flat_key-value pairs. ## Test...

cla-signed

## Ticket ET-97 ET-101 ET-102 ET-164 ## Description Introduce custom project key (similar to JIRA's project key), to be able to refer to projects. This adds the column to the...

cla-signed

## Description Add shared cluster tag to each master & db deployments to the shared cluster + tags created checkpoint storage buckets with the same tag. ## Test Plan -...

cla-signed

## Description Currently if a single-run/trial experiment doesn't store validation metrics, then it will not be assigned a best_trial_id, so when there are single trial experiments that are using `report_metrics`...

cla-signed

## Description Adds allocations as a streamable type + some simplifications of the metric triggers. ## Test Plan - Run `stream.py` and make sure you receive proper updates regarding allocations...

cla-signed

## Ticket [CM-461] ## Description Older version of Kubernetes don't have the expected label (`batch.kubernetes.io/job-name`), instead they use `job-name`. https://kubernetes.io/docs/reference/labels-annotations-taints/#job-name This PR adds a fallback to handle backwards compatibility in...

cla-signed

## Ticket [CM-412] ## Description A bug in the webUI was found where workspaces couldn't be deleted if there had ever been an experiment present in the workspace at any...

cla-signed

## Ticket [CM-404] ## Description Update the resource aggregates query to exclude checkpoint garbage collection tasks, we were seeing excessively long queued times in the last few release parties due...

cla-signed