Skip to main content

Publish data for modeling (gold)

What you'll accomplish: Create a gold artifact that ML or Forecast can use in the training wizard.

Before you start​

  • A bronze or silver dataset in the Data catalog

Steps​

  1. Open the dataset you want to publish in Data.
  2. Click Publish gold.
  3. Enter a name for the gold artifact.
  4. Select the use case:
    • Machine learning — choose a target column to predict.
    • Forecast — choose a time column; data must be valid time series.
  5. Run Validate and fix any errors shown.
  6. Click Publish. The artifact appears under Gold with status ready.

Start training from gold​

  • ML: Machine Learning → New model → select your gold dataset (or open gold detail → Start ML project).
  • Forecast: Forecast → Build → select your gold dataset.

Requirements (at a glance)​

Use caseMinimum rowsNotes
ML100Target column required; needs variation in the target
Forecast30Parseable datetime + numeric values

What happens next​

Common questions​

No gold shows in the wizard
Confirm publish succeeded, status is ready, and you enabled the correct use case (ML vs Forecast).

Validation failed on forecast
Ensure the first column is datetime (or set time column explicitly) and the value column is numeric. See Build a forecast.