Open yueguoguo opened 5 years ago
I have added this in scala to MMLSpark.
https://github.com/Azure/mmlspark/blob/master/src/recommendation/src/main/scala/RankingEvaluator.scala You can either use my code as a base, or you can do something like the SAR+ notebook to import MMLSpark and use it like this python test. https://github.com/Azure/mmlspark/blob/master/src/recommendation/src/test/python/RankingSpec.py
Thanks @dciborow is there a derived PySpark / Python version of the Scala codes in mmlspark?
It is used in many known cases when only one relevant item from a recommended list is considered.