I was able to successfully serialize a pyspark generated RandomForestModel as Mleap bundle and run it using mleap-runtime. I want to get the confidence score and score for each feature. Is there a way to achieve that without using any spark dependencies?
Asked
Active
Viewed 14 times
