1

Taking cue from Is it possible to obtain class probabilities using GradientBoostedTrees with spark mllib?

The answer to this question is given last December. I wanted to know if there is anything in Spark 1.6.1.

I tried to edit the above question but somehow it did not happen.

My query is like most other algorithms in Spark mllib library we can find predicted probabilities. Can we get the same in Gradient Boosting Trees-Classification(http://spark.apache.org/docs/latest/mllib-ensembles.html#gradient-boosted-trees-vs-random-forests) and change the model threshold?

That way I can change the threshold to improve the model performance.

Community
  • 1
  • 1
PARTHA TALUKDER
  • 321
  • 5
  • 17
  • Beside answering your first question with "not yet available", I don't understand what you are asking for in the rest of the question. – eliasah Apr 27 '16 at 06:55
  • Thanks for your reply. Is there any way other than model.setThreshold(), I can change the threshold? I know its repetitive. But I am searching for this solution as this is available in R. – PARTHA TALUKDER Apr 27 '16 at 07:17
  • Just to add to the question if we were to treat the classification as a regression problem(with the response being still binary 0|1) and get predicted values which if required can be normalized in 0-1 range will this be effective indicator of response probability? – sourabh Apr 27 '16 at 08:06

0 Answers0