search

danzafar / tidyspark

tidyspark: a tidyverse implementation of SparkR built for simplicity, elegance, and ease of use.
Other
22 stars 0 forks source link

Add ML Support #28

Closed estern95 closed 4 years ago

estern95 commented 4 years ago

supervised approaches are finished on first pass

danzafar commented 4 years ago

Hey @estern95 once we get these tests sorted out we can go ahead and pull in. I might use your branch to publish a summary function for these object types.

estern95 commented 4 years ago

I fixed the tests :) I royally screwed up my java installation today :) that was fun

danzafar commented 4 years ago

hey @estern95 when I do load_all on this branch I get message:

 Error in setMethod("ml_perplexity", signature(object = "LDAModel", data = "SparkDataFrame"),  : 
  no existing definition for function ‘ml_perplexity’

Any idea? By the way I'm thinking to just port over SparkR's tests for this. They seem conprehensive. https://github.com/apache/spark/blob/master/R/pkg/tests/fulltests/test_mllib_regression.R