Source

orange-multitarget / docs / rst / ClassifierChain.rst

Full commit

ClassifierChain

Widget icon

Signals

Inputs:
  • Learner

    The base learner used in the ensemble technique.

Outputs:
  • Learner or Classifier

Description

Usage example

Classifier chain learner takes a single-target learner and with it creates a classifier for every class variable in the data. Every time a classifier is created, the values of that class variable are added to features. The order in which the class variables are chosen is random.

Setting:

  • Use actual values

    If checked, the values added into features are actual values from the data. Otherwise the values predicted by the classifier are used.