Skip to content
CSA Loom — the Microsoft Fabric experience for Azure tenants where Fabric isn't yet available: lakehouses, warehouses, notebooks, semantic models, Activator rules, Data Agents, across Commercial, GCC, GCC-High, and DoD IL5

Tutorial: AutoML editor

CSA Loom automl editor — a low-code Automated ML wizard that submits real Azure Machine Learning AutoML jobs and monitors them live. No Microsoft Fabric required.

What it is

AutoML is a low-code wizard for Automated machine learning. In Loom it runs real Azure Machine Learning AutoML jobs (Microsoft.MachineLearningServices/workspaces/<ws>/jobs, jobType: 'AutoML'). Pick a task, point at a dataset and target column, choose compute, and AutoML trains and ranks candidate models while you watch the run.

When to use it

  • You want a trained model without hand-writing training code — AutoML sweeps algorithms and hyperparameters for you.
  • You have tabular data and a clear label column to predict.
  • You need a governed, reproducible AML job (visible in the AML workspace) — not a black box.

Step-by-step in Loom

  1. Create the item. Choose + New item → AutoML (Data Science). The editor opens at /items/automl/<id>.
  2. Pick a task type. Choose Classification (binary or multi-class), Regression, or Forecasting — AutoML applies the right family of algorithms for the task.
  3. Choose dataset + target. Select a datastore and the MLTable folder that holds your tabular data, then name the target (label) column AutoML should learn to predict.
  4. Select compute. Pick an AmlCompute cluster from the workspace to run the model sweep on.
  5. Set limits and submit. Choose the primary metric and limits (timeout, max trials, concurrency), then submit — a real AutoML job — and watch it on the Runs tab.

The Azure backend it rides on

  • Jobs: Azure Machine Learning workspace jobs REST (jobType: 'AutoML') via the Console managed identity.
  • Compute: an AmlCompute cluster in the bound AML workspace.
  • Gate: a missing AML workspace surfaces the exact LOOM_AML_* env vars to set — the surface never fakes a run.

No Fabric required

AutoML jobs run entirely on Azure Machine Learning; no Fabric capacity, workspace, or OneLake is involved.

Learn more