To run the Spark application, assuming Spark binaries are in your path:
$ spark-submit --packages com.databricks:spark-csv_2.11:1.4.0 app.py
To test the pipelines and datasource files:
$ spark-submit --packages com.databricks:spark-csv_2.11:1.4.0 datasources.py
$ spark-submit --packages com.databricks:spark-csv_2.11:1.4.0 pipelines.py