Start Tensorboard from the UI

If your runs have Tensorboard, Polyaxon will automatically show a new button to start a Tensorboard service based on that run, this will reduce the need to copy the UUID and start the operation with the CLI:

single-run-tensorboard

If the current run has already a downstream Tensorboard, the UI will show an icon to redirect the user directly to the Tensorboard service.

running-tensorboard

If a user has a custom Tensorboard component, they just need to use their own reference instead of the default Tensorboard component.

Custom UI Actions

Similar to Tensorboard, which is just a specific use case, Polyaxon UI has two new buttons to trigger custom downstream operations. These buttons allow triggering an operation manually using the UI based on the context of the current run. In order to leverage those buttons, you will need to register a component that expects a run or multiple runs and performs some logic on their outputs or artifacts.

For a single run:

custom-action-single-run

For a selection of runs:

custom-action-multi-runs

Learn More about Polyaxon

This blog post just goes over a couple of features that we shipped since our last product update, several other features and fixes are worth checking. To learn more about all the features, fixes, and enhancements, please visit the release notes.

Polyaxon continues to grow quickly and keeps improving and providing the simplest machine learning abstraction. We hope that these updates will improve your workflows and increase your productivity, and again, thank you for your continued feedback and support.