Hitachi Vantara Pentaho Community Wiki
Child pages
  • ClassifierSubsetEval
Skip to end of metadata
Go to start of metadata

Package

weka.attributeSelection

Synopsis

Classifier subset evaluator:

Evaluates attribute subsets on training data or a seperate hold out testing set. Uses a classifier to estimate the 'merit' of a set of attributes.

Available in Weka 3.6.x - 3.7.1. Available via the package management system for Weka >= 3.7.2 (classifierBasedAttributeSelection).

Options

The table below describes the options available for ClassifierSubsetEval.

Option

Description

classifier

Classifier to use for estimating the accuracy of subsets

holdOutFile

File containing hold out/test instances.

useTraining

Use training data instead of hold out/test instances.

Capabilities

The table below describes the capabilites of ClassifierSubsetEval.

Capability

Supported

Class

Binary class, Date class, Missing class values, Numeric class, Nominal class

Attributes

Missing values, Empty nominal attributes, Relational attributes, Unary attributes, Nominal attributes, Binary attributes, Date attributes, Numeric attributes, String attributes

Min # of instances

0

  • No labels