Apizr for VS Code

Inspect Python and notebooks, browse local experiment history and compare Runs.

These are actual SDK and Jupyter executions on the public UCI Wine dataset, shown in the 0.4.6 extension. Parameters, accuracy and macro F1 come from the retained Runs; the interface delegates history validation to the Apizr engine.

Short walkthrough using captured views of the running extension:

Repeat the illustrated Academy lesson to train the models, inspect saved source and compare their results.

Get started

1. Install the Apizr VSIX in VS Code 1.140+ and open a trusted project.
2. Accept Install CLI. Apizr installs its CLI and notebook support in a managed environment, then selects it automatically. Python 3.11–3.15 is required.
3. Configure your apizr.toml and operator policy in Apizr: Open settings.

Use the Apizr bee in the Activity Bar for the experiment dashboard, metrics, parameters and Run details. Python editor shortcuts and the notebook toolbar inspect saved code without executing it. Apizr: Open CLI terminal makes the installed CLI available in a terminal. Existing CLI environments remain selectable.

The managed installer uses the exact PyPI release outerspace-apizr[notebook]==0.4.6.

Usage telemetry follows VS Code’s permission by default. Disable apizr.telemetry.enabled to opt out. Only operation/outcome counts are sent; code, paths and experiment data stay local. Privacy policy.

Setup guide · Contributing · Engine