New Features in October: Bulk Actions, Label Filtering, Faster Machine Start-up, and More!

Source: Deep Learning on Medium

New Features in October: Bulk Actions, Label Filtering, Faster Machine Start-up, and More!

Performing bulk actions on runs

Start Runs from Repositories with Uncommitted Changes

Spell automatically syncs uncommitted changes and use those changes as a git patch within your run. We provide a link that enables you to download the changes from your web console. Iterate and run your code faster while still getting all the benefits of reproducibility with Spell. Read more in our docs.

Manage Runs with Bulk Actions

Perform actions on multiple runs at once in the web console. Stop, kill, archive, and add labels with bulk actions.

Filter by Label

Filter by label easily by clicking on a label or by using the filter menu dropdown.

Faster Machine Start-up Time for Private Clusters

Start-up new machines in ~3–4 minutes (down from 15 minutes previously). Performance improvements to our machine start-up logic mean less waiting to get started on your work.

Set Early Stopping Conditions

Stop runs that aren’t meeting specified conditions by adding early stopping logic.

Mount Paths Are Now Relative

Mounts to a relative path are now also relative to your current working directory. If you start a run within a subdirectory of your repository, then the resources mounted to a relative path will also be relative to that directory.

Increased Visibility on Runs in Your Workflow

View and monitor active runs in your workflow on the new workflow list and details pages. Try an example workflow here.

For Teams: Mount Resources from Multiple Buckets

For Teams subscribers, kick-off new runs without requiring data to be in one bucket. Mount resources from separate buckets (e.g. run outputs and a private s3 bucket) into a run or Jupyter Workspace.