Class: BinaryClassificationMetrics

eclairjs/mllib/evaluation. BinaryClassificationMetrics

new BinaryClassificationMetrics(scoreAndLabels, numBins)

Parameters:
Name Type Description
scoreAndLabels module:eclairjs.RDD
numBins number
Source:

Methods

areaUnderPR() → {number}

Computes the area under the precision-recall curve.
Source:
Returns:
Type
number

areaUnderROC() → {number}

Computes the area under the receiver operating characteristic (ROC) curve.
Source:
Returns:
Type
number

fMeasureByThreshold(betaopt) → {module:eclairjs.RDD}

Returns the (threshold, F-Measure) curve.
Parameters:
Name Type Attributes Description
beta number <optional>
 the beta factor in F-Measure computation.
Source:
See:
Returns:
an RDD of (threshold, F-Measure) pairs.
Type
module:eclairjs.RDD

pr() → {module:eclairjs.RDD}

Returns the precision-recall curve, which is an RDD of (recall, precision), NOT (precision, recall), with (0.0, 1.0) prepended to it.
Source:
See:
Returns:
Type
module:eclairjs.RDD

precisionByThreshold() → {module:eclairjs.RDD}

Returns the (threshold, precision) curve.
Source:
Returns:
Type
module:eclairjs.RDD

recallByThreshold() → {module:eclairjs.RDD}

Returns the (threshold, recall) curve.
Source:
Returns:
Type
module:eclairjs.RDD

roc() → {module:eclairjs.RDD}

Returns the receiver operating characteristic (ROC) curve, which is an RDD of (false positive rate, true positive rate) with (0.0, 0.0) prepended and (1.0, 1.0) appended to it.
Source:
See:
Returns:
Type
module:eclairjs.RDD

thresholds() → {module:eclairjs.RDD}

Returns thresholds in descending order.
Source:
Returns:
Type
module:eclairjs.RDD

unpersist()

Unpersist intermediate RDDs used in the computation.
Source: