I am trying to run the \'prediction\' method of a scikit-learn model at scale using Pyspark. The model is accompanied by supporting feature engineering and utility methods i