I am trying to build a recommendation engine in PySpark. I am using collaborative filtering approach with Sparks\'s ALS library. I have trained the model with best set of parame