Description
With the new Pipeline API, it is possible to seamlessly support machine learning reductions as meta algorithms.
GBDT and SVM today are binary classifiers and we can implement multi class classification as a One vs All, or All vs All (or even more sophisticated reduction) using binary classifiers as primitives.
This JIRA is to track the creation of a reduction API for multi class classification.
Attachments
Issue Links
- is duplicated by
-
SPARK-7372 Multiclass SVM - One vs All wrapper
- Resolved
- relates to
-
SPARK-7573 OneVsRest cleanups
- Resolved
-
SPARK-7574 User guide update for OneVsRest
- Resolved
-
SPARK-7575 Example code for OneVsRest
- Resolved
- links to