org.apache.spark.ml.evaluation

Class MultilabelClassificationEvaluator

Object

org.apache.spark.ml.evaluation.Evaluator

org.apache.spark.ml.evaluation.MultilabelClassificationEvaluator

All Implemented Interfaces:

java.io.Serializable, Params, HasLabelCol, HasPredictionCol, DefaultParamsWritable, Identifiable, MLWritable

public class MultilabelClassificationEvaluator
extends Evaluator
implements HasPredictionCol, HasLabelCol, DefaultParamsWritable

:: Experimental :: Evaluator for multi-label classification, which expects two input columns: prediction and label.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
MultilabelClassificationEvaluator()
MultilabelClassificationEvaluator(String uid)

Method Summary

Modifier and Type	Method and Description
MultilabelClassificationEvaluator	copy(ParamMap extra) Creates a copy of this instance with the same UID and some extra params.
double	evaluate(Dataset<?> dataset) Evaluates model output and returns a scalar metric.
double	getMetricLabel()
String	getMetricName()
MultilabelMetrics	getMetrics(Dataset<?> dataset) Get a MultilabelMetrics, which can be used to get multilabel classification metrics such as accuracy, precision, precisionByLabel, etc.
boolean	isLargerBetter() Indicates whether the metric returned by evaluate should be maximized (true, default) or minimized (false).
Param<String>	labelCol() Param for label column name.
static MultilabelClassificationEvaluator	load(String path)
DoubleParam	metricLabel() param for the class whose metric will be computed in "precisionByLabel", "recallByLabel", "f1MeasureByLabel".
Param<String>	metricName() param for metric name in evaluation (supports "f1Measure" (default), "subsetAccuracy", "accuracy", "hammingLoss", "precision", "recall", "precisionByLabel", "recallByLabel", "f1MeasureByLabel", "microPrecision", "microRecall", "microF1Measure")
Param<String>	predictionCol() Param for prediction column name.
static MLReader<T>	read()
MultilabelClassificationEvaluator	setLabelCol(String value)
MultilabelClassificationEvaluator	setMetricLabel(double value)
MultilabelClassificationEvaluator	setMetricName(String value)
MultilabelClassificationEvaluator	setPredictionCol(String value)
String	toString()
String	uid() An immutable unique ID for the object and its derivatives.

Methods inherited from class org.apache.spark.ml.evaluation.Evaluator

evaluate, params

Methods inherited from class Object

equals, getClass, hashCode, notify, notifyAll, wait, wait, wait

Methods inherited from interface org.apache.spark.ml.param.shared.HasPredictionCol

getPredictionCol

Methods inherited from interface org.apache.spark.ml.param.shared.HasLabelCol

getLabelCol

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

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

write

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

save

Constructor Detail

MultilabelClassificationEvaluator

public MultilabelClassificationEvaluator(String uid)

MultilabelClassificationEvaluator

public MultilabelClassificationEvaluator()

Method Detail

load

public static MultilabelClassificationEvaluator load(String path)

read

public static MLReader<T> read()

labelCol

public final Param<String> labelCol()

Description copied from interface: HasLabelCol

Param for label column name.

Specified by:

labelCol in interface HasLabelCol

Returns:

(undocumented)

predictionCol

public final Param<String> predictionCol()

Description copied from interface: HasPredictionCol

Param for prediction column name.

Specified by:

predictionCol in interface HasPredictionCol

Returns:

(undocumented)

uid

public String uid()

Description copied from interface: Identifiable

An immutable unique ID for the object and its derivatives.

Specified by:

uid in interface Identifiable

Returns:

(undocumented)

metricName

public final Param<String> metricName()

param for metric name in evaluation (supports "f1Measure" (default), "subsetAccuracy", "accuracy", "hammingLoss", "precision", "recall", "precisionByLabel", "recallByLabel", "f1MeasureByLabel", "microPrecision", "microRecall", "microF1Measure")

Returns:

(undocumented)

getMetricName

public String getMetricName()

setMetricName

public MultilabelClassificationEvaluator setMetricName(String value)

metricLabel

public final DoubleParam metricLabel()

param for the class whose metric will be computed in "precisionByLabel", "recallByLabel", "f1MeasureByLabel".

Returns:

(undocumented)

getMetricLabel

public double getMetricLabel()

setMetricLabel

public MultilabelClassificationEvaluator setMetricLabel(double value)

setPredictionCol

public MultilabelClassificationEvaluator setPredictionCol(String value)

setLabelCol

public MultilabelClassificationEvaluator setLabelCol(String value)

evaluate

public double evaluate(Dataset<?> dataset)

Description copied from class: Evaluator

Evaluates model output and returns a scalar metric. The value of isLargerBetter specifies whether larger values are better.

Specified by:

evaluate in class Evaluator

Parameters:

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

Returns:

metric

getMetrics

public MultilabelMetrics getMetrics(Dataset<?> dataset)

Get a MultilabelMetrics, which can be used to get multilabel classification metrics such as accuracy, precision, precisionByLabel, etc.

Parameters:

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

Returns:

MultilabelMetrics

isLargerBetter

public boolean isLargerBetter()

Description copied from class: Evaluator

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.

Overrides:

isLargerBetter in class Evaluator

Returns:

(undocumented)

copy

public MultilabelClassificationEvaluator 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. See defaultCopy().

Specified by:

copy in interface Params

Specified by:

copy in class Evaluator

Parameters:

extra - (undocumented)

Returns:

(undocumented)

toString

public String toString()

Specified by:

toString in interface Identifiable

Overrides:

toString in class Object