public class MulticlassClassificationEvaluator extends Evaluator implements HasPredictionCol, HasLabelCol, HasWeightCol, HasProbabilityCol, DefaultParamsWritable
| Constructor and Description |
|---|
MulticlassClassificationEvaluator() |
MulticlassClassificationEvaluator(String uid) |
| Modifier and Type | Method and Description |
|---|---|
DoubleParam |
beta()
The beta value, which controls precision vs recall weighting,
used in
"weightedFMeasure", "fMeasureByLabel". |
MulticlassClassificationEvaluator |
copy(ParamMap extra)
Creates a copy of this instance with the same UID and some extra params.
|
DoubleParam |
eps()
param for eps.
|
double |
evaluate(Dataset<?> dataset)
Evaluates model output and returns a scalar metric.
|
double |
getBeta() |
double |
getEps() |
double |
getMetricLabel() |
String |
getMetricName() |
MulticlassMetrics |
getMetrics(Dataset<?> dataset)
Get a MulticlassMetrics, which can be used to get multiclass classification
metrics such as accuracy, weightedPrecision, 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 MulticlassClassificationEvaluator |
load(String path) |
DoubleParam |
metricLabel()
The class whose metric will be computed in
"truePositiveRateByLabel",
"falsePositiveRateByLabel", "precisionByLabel", "recallByLabel",
"fMeasureByLabel". |
Param<String> |
metricName()
param for metric name in evaluation (supports
"f1" (default), "accuracy",
"weightedPrecision", "weightedRecall", "weightedTruePositiveRate",
"weightedFalsePositiveRate", "weightedFMeasure", "truePositiveRateByLabel",
"falsePositiveRateByLabel", "precisionByLabel", "recallByLabel",
"fMeasureByLabel", "logLoss", "hammingLoss") |
Param<String> |
predictionCol()
Param for prediction column name.
|
Param<String> |
probabilityCol()
Param for Column name for predicted class conditional probabilities.
|
static MLReader<T> |
read() |
MulticlassClassificationEvaluator |
setBeta(double value) |
MulticlassClassificationEvaluator |
setEps(double value) |
MulticlassClassificationEvaluator |
setLabelCol(String value) |
MulticlassClassificationEvaluator |
setMetricLabel(double value) |
MulticlassClassificationEvaluator |
setMetricName(String value) |
MulticlassClassificationEvaluator |
setPredictionCol(String value) |
MulticlassClassificationEvaluator |
setProbabilityCol(String value) |
MulticlassClassificationEvaluator |
setWeightCol(String value) |
String |
toString() |
String |
uid()
An immutable unique ID for the object and its derivatives.
|
Param<String> |
weightCol()
Param for weight column name.
|
getPredictionColgetLabelColgetWeightColgetProbabilityColclear, copyValues, defaultCopy, defaultParamMap, explainParam, explainParams, extractParamMap, extractParamMap, get, getDefault, getOrDefault, getParam, hasDefault, hasParam, isDefined, isSet, onParamChange, paramMap, params, set, set, set, setDefault, setDefault, shouldOwnwritesavepublic MulticlassClassificationEvaluator(String uid)
public MulticlassClassificationEvaluator()
public static MulticlassClassificationEvaluator load(String path)
public static MLReader<T> read()
public final Param<String> probabilityCol()
HasProbabilityColprobabilityCol in interface HasProbabilityColpublic final Param<String> weightCol()
HasWeightColweightCol in interface HasWeightColpublic final Param<String> labelCol()
HasLabelCollabelCol in interface HasLabelColpublic final Param<String> predictionCol()
HasPredictionColpredictionCol in interface HasPredictionColpublic String uid()
Identifiableuid in interface Identifiablepublic Param<String> metricName()
"f1" (default), "accuracy",
"weightedPrecision", "weightedRecall", "weightedTruePositiveRate",
"weightedFalsePositiveRate", "weightedFMeasure", "truePositiveRateByLabel",
"falsePositiveRateByLabel", "precisionByLabel", "recallByLabel",
"fMeasureByLabel", "logLoss", "hammingLoss")
public String getMetricName()
public MulticlassClassificationEvaluator setMetricName(String value)
public MulticlassClassificationEvaluator setPredictionCol(String value)
public MulticlassClassificationEvaluator setLabelCol(String value)
public MulticlassClassificationEvaluator setWeightCol(String value)
public MulticlassClassificationEvaluator setProbabilityCol(String value)
public final DoubleParam metricLabel()
"truePositiveRateByLabel",
"falsePositiveRateByLabel", "precisionByLabel", "recallByLabel",
"fMeasureByLabel".
Must be greater than or equal to 0. The default value is 0.
public double getMetricLabel()
public MulticlassClassificationEvaluator setMetricLabel(double value)
public final DoubleParam beta()
"weightedFMeasure", "fMeasureByLabel".
Must be greater than 0. The default value is 1.
public double getBeta()
public MulticlassClassificationEvaluator setBeta(double value)
public final DoubleParam eps()
public double getEps()
public MulticlassClassificationEvaluator setEps(double value)
public double evaluate(Dataset<?> dataset)
EvaluatorisLargerBetter specifies whether larger values are better.
public MulticlassMetrics getMetrics(Dataset<?> dataset)
dataset - a dataset that contains labels/observations and predictions.public boolean isLargerBetter()
Evaluatorevaluate should be maximized (true, default)
or minimized (false).
A given evaluator may support multiple metrics which may be maximized or minimized.isLargerBetter in class Evaluatorpublic MulticlassClassificationEvaluator copy(ParamMap extra)
ParamsdefaultCopy().public String toString()
toString in interface IdentifiabletoString in class Object