I would like to implement a pandasUDF function in Pyspark which returns a matrix of float numbers. I have tried, but the bellow appeared.
RuntimeError: (\'Excepti