04-07-2017 07:01 AM
I am trying to implement NaiveBayes model using Cloudera PySpark ML libraries.(https://spark.apache.org/docs/1.6.1/api/python/pyspark.ml.html#pyspark.ml.classification.NaiveBayes)
It does not provide conditional probabilities which are the main component of the model. Examples online does not show how input and output are determined. Please help with the details.
1. What should be the input?
2. Is anyone implementing this model? How?