org.apache.spark.ml.classification

Interface BinaryLogisticRegressionSummary

All Superinterfaces:

LogisticRegressionSummary, java.io.Serializable

All Known Subinterfaces:

BinaryLogisticRegressionTrainingSummary

All Known Implementing Classes:

BinaryLogisticRegressionSummaryImpl, BinaryLogisticRegressionTrainingSummaryImpl

public interface BinaryLogisticRegressionSummary
extends LogisticRegressionSummary

Abstraction for binary logistic regression results for a given model.

Currently, the summary ignores the instance weights.

Method Summary

Modifier and Type	Method and Description
double	areaUnderROC() Computes the area under the receiver operating characteristic (ROC) curve.
Dataset<Row>	fMeasureByThreshold() Returns a dataframe with two fields (threshold, F-Measure) curve with beta = 1.0.
Dataset<Row>	pr() Returns the precision-recall curve, which is a Dataframe containing two fields recall, precision with (0.0, 1.0) prepended to it.
Dataset<Row>	precisionByThreshold() Returns a dataframe with two fields (threshold, precision) curve.
Dataset<Row>	recallByThreshold() Returns a dataframe with two fields (threshold, recall) curve.
Dataset<Row>	roc() Returns the receiver operating characteristic (ROC) curve, which is a Dataframe having two fields (FPR, TPR) with (0.0, 0.0) prepended and (1.0, 1.0) appended to it.

Methods inherited from interface org.apache.spark.ml.classification.LogisticRegressionSummary

accuracy, asBinary, falsePositiveRateByLabel, featuresCol, fMeasureByLabel, fMeasureByLabel, labelCol, labels, precisionByLabel, predictionCol, predictions, probabilityCol, recallByLabel, truePositiveRateByLabel, weightedFalsePositiveRate, weightedFMeasure, weightedFMeasure, weightedPrecision, weightedRecall, weightedTruePositiveRate

Method Detail

areaUnderROC

double areaUnderROC()

Computes the area under the receiver operating characteristic (ROC) curve.

Returns:

(undocumented)

Note:

This ignores instance weights (setting all to 1.0) from LogisticRegression.weightCol. This will change in later Spark versions.

fMeasureByThreshold

Dataset<Row> fMeasureByThreshold()

Returns a dataframe with two fields (threshold, F-Measure) curve with beta = 1.0.

Returns:

(undocumented)

Note:

This ignores instance weights (setting all to 1.0) from LogisticRegression.weightCol. This will change in later Spark versions.

pr

Dataset<Row> pr()

Returns the precision-recall curve, which is a Dataframe containing two fields recall, precision with (0.0, 1.0) prepended to it.

Returns:

(undocumented)

Note:

This ignores instance weights (setting all to 1.0) from LogisticRegression.weightCol. This will change in later Spark versions.

precisionByThreshold

Dataset<Row> precisionByThreshold()

Returns a dataframe with two fields (threshold, precision) curve. Every possible probability obtained in transforming the dataset are used as thresholds used in calculating the precision.

Returns:

(undocumented)

Note:

This ignores instance weights (setting all to 1.0) from LogisticRegression.weightCol. This will change in later Spark versions.

recallByThreshold

Dataset<Row> recallByThreshold()

Returns a dataframe with two fields (threshold, recall) curve. Every possible probability obtained in transforming the dataset are used as thresholds used in calculating the recall.

Returns:

(undocumented)

Note:

This ignores instance weights (setting all to 1.0) from LogisticRegression.weightCol. This will change in later Spark versions.

roc

Dataset<Row> roc()

Returns the receiver operating characteristic (ROC) curve, which is a Dataframe having two fields (FPR, TPR) with (0.0, 0.0) prepended and (1.0, 1.0) appended to it. See http://en.wikipedia.org/wiki/Receiver_operating_characteristic

Returns:

(undocumented)

Note:

This ignores instance weights (setting all to 1.0) from LogisticRegression.weightCol. This will change in later Spark versions.