Developer API

org.apache.spark.ml.evaluation

Class Evaluator

java.lang.Object

org.apache.spark.ml.evaluation.Evaluator

All Implemented Interfaces:

java.io.Serializable, Params, Identifiable

Direct Known Subclasses:

BinaryClassificationEvaluator, MulticlassClassificationEvaluator, RegressionEvaluator

public abstract class Evaluator
extends java.lang.Object
implements Params

Abstract class for evaluators that compute metrics from predictions.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
Evaluator()

Method Summary

Methods

Modifier and Type	Method and Description
abstract Evaluator	copy(ParamMap extra) Creates a copy of this instance with the same UID and some extra params.
abstract double	evaluate(DataFrame dataset) Evaluates the output.
double	evaluate(DataFrame dataset, ParamMap paramMap) Evaluates model output and returns a scalar metric (larger is better).
boolean	isLargerBetter() Indicates whether the metric returned by evaluate() should be maximized (true, default) or minimized (false).

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Methods inherited from interface org.apache.spark.ml.param.Params

clear, copyValues, defaultCopy, defaultParamMap, explainParam, explainParams, extractParamMap, extractParamMap, get, getDefault, getOrDefault, getParam, hasDefault, hasParam, isDefined, isSet, paramMap, params, set, set, set, setDefault, setDefault, shouldOwn, validateParams

Methods inherited from interface org.apache.spark.ml.util.Identifiable

toString, uid

Constructor Detail

Evaluator

public Evaluator()

Method Detail

evaluate

public double evaluate(DataFrame dataset,
              ParamMap paramMap)

Evaluates model output and returns a scalar metric (larger is better).

Parameters:

dataset - a dataset that contains labels/observations and predictions.

paramMap - parameter map that specifies the input columns and output metrics

Returns:

metric

evaluate

public abstract double evaluate(DataFrame dataset)

Evaluates the output.

Parameters:

dataset - a dataset that contains labels/observations and predictions.

Returns:

metric

isLargerBetter

public boolean isLargerBetter()

Indicates whether the metric returned by evaluate() should be maximized (true, default) or minimized (false). A given evaluator may support multiple metrics which may be maximized or minimized.

Returns:

(undocumented)

copy

public abstract Evaluator copy(ParamMap extra)

Description copied from interface: Params

Creates a copy of this instance with the same UID and some extra params. Subclasses should implement this method and set the return type properly.

Specified by:

copy in interface Params

Parameters:

extra - (undocumented)

Returns:

(undocumented)

See Also:

defaultCopy()