This step builds one or several "computable" items of the Flow (build datasets, build managed folders, train saved models)

For partitioned datasets (and folders), you can use partitioning specs

Build mode

{{translate('SCENARIO.STEPS.EDITORS.BUILD_FLOWITEM.BUILD_REQUIRED_DEPENDENCIES', 'Build required dependencies')}} Build datasets that are out-of-date (if their predecessors have changed, or if the recipe has changed) {{translate('SCENARIO.STEPS.EDITORS.BUILD_FLOWITEM.FORCE_BUILD', 'Force-build')}} Rebuild all datasets leading to the selected one

Update output schemas

Before each recipe runs, update the schema of the output dataset(s)

Stop at zone boundary

Don't go beyond flow zone when recursing

Mode

{{translate('SCENARIO.STEPS.EDITORS.BUILD_FLOWITEM.RUN_DOWNSTREAM', 'Run downstream recipes (with schema propagation)')}} Run all recipes "from left to right" from these items until the end of the Flow is reached. Propagate schema automatically along the way {{translate('SCENARIO.STEPS.EDITORS.BUILD_FLOWITEM.RUN_DOWNSTREAM_RECIPES', 'Run downstream recipes (no schema propagation)')}} Run all recipes "from left to right" from these items until the end of the Flow is reached

Handling of dependencies

{{translate('SCENARIO.STEPS.EDITORS.BUILD_FLOWITEM.BUILD_REQUIRED_DEPENDENCIES', 'Build required dependencies')}} Build datasets that are out-of-date (if their predecessors have changed, or if the recipe has changed) needed for the reachable outputs {{translate('SCENARIO.STEPS.EDITORS.BUILD_FLOWITEM.FORCE_BUILD_DEPENDENCIES', 'Force-build all dependencies')}} Rebuild all datasets leading to the reachable outputs

Update output schemas

Before each recipe runs, update the schema of the output dataset(s)

Stop at zone boundary

Don't go beyond flow zone when recursing