graphrag/examples/interdependent_workflows/pipeline.yml

workflows:
  - name: aggregate_workflow
    steps:
      - verb: "aggregate"  # https://github.com/microsoft/datashaper/blob/main/python/datashaper/datashaper/engine/verbs/aggregate.py
        args:
            groupby: "type"
            column: "col_multiplied"
            to: "aggregated_output"
            operation: "sum"
        input:
          source: "workflow:derive_workflow" # reference the derive_workflow, cause this one requires that one to run first
            # Notice, these are out of order, the indexing engine will figure out the right order to run them in

  - name: derive_workflow
    steps:
      - verb: "derive" # https://github.com/microsoft/datashaper/blob/main/python/datashaper/datashaper/engine/verbs/derive.py
        args:
          column1: "col1"  # from above
          column2: "col2"  # from above
          to: "col_multiplied"  # new column name
          operator: "*"  # multiply the two columns,
    # Since we're trying to act on the dataset, we don't need explicitly to specify an input
      # "input": { "source": "source" } # use the dataset as the input to this verb. This is the default, so you can omit it.