Multiclass classification with Random Forest in Apache Spark

Question

The Apache Spark's documentation (1.4.0) promises that Random Forest (the same promise is for decision trees) can be extended to multiclass classification setting. However, I can't find any way to reach this using RandomForestModel API:

randomForestModel.predict

Could you please provide a right way how to do that, if it's really possible in 1.4.0 version?

score 0 · Answer 1 · answered Jul 06 '15 at 13:58

0

All right, I've found it in an experimental package (ml): org.apache.spark.ml.classification.RandomForestClassificationModel

and transform should be used instead of predict:

RandomForestClassificationModel.transform

answered Jul 06 '15 at 13:58

Nikita Zhiltsov

654
9
15

Multiclass classification with Random Forest in Apache Spark

1 Answers1