
Polyaxon v1.18: New Tensorboard component and initializer
You can now start Tensorboard using the experiment names instead of the experiment uuids
Page 2 of 5

You can now start Tensorboard using the experiment names instead of the experiment uuids

We recently worked on an effort to improve the visual experience our CLI

You can now define annotations and environment variables in your connections' schema

Users can now filter by connections and artifacts

Admin users with access to kubectl can now force kill operations in a namespace

Users can now package and pull runs to local paths or push runs from local paths to a Polyaxon deployment

Using the line chart widget you can now inspect the underlying data as an interactive table.

Users can now transfer runs, models, components, & artifacts from one project to another to reorganize their workspaces
Users can now log the progress of their runs and have a progress bar in the UI to get a quick overview of their runs.

Improved inspection and monitoring capabilities and Container Shell debugging capabilities in the CLI.

Polyaxon UI can now render CSV/TSV/Dataframes/... automatically as a table

Users can search organizations they belong to and project they have access and access recently viewed items.

Users can quickly search by a specific component and see recent runs for a specific component.

Organization's owners and managers can now invite read-only seats and billing seats to their organization.

Organization's owners and managers can create service accounts to issue tokens for CI/CD/CT services.

How to turn the runs table into a heatmap using the new heat field feature.

The queues table shows now a saturation bar for queues with quotas or concurrency limits and stats about the runs assigned to each queue.

A new logger for the HuggingFace transformer library.

Users can now use the cli to pass hyperparaeter directlu from the CI without creating a YAML file.

Polyaxon v1.12 release of our MLOps platform builds on top of the previous version and brings new features and improvements to the specification and dashboard experience.

Improved init section with support for promoted models and artifacts.

New hyperparam generators for dates and datetimes and Backfill scheduling strategy.

A new feature for IO and Params sectiosn to turn parameters to environment variables `toEnv`.

Managing cross project runs and moving runs from one project to another to re-organize or refactor previous work.