Class: MulticlassClassificationEvaluator

eclairjs/ml/evaluation. MulticlassClassificationEvaluator

Evaluator for multiclass classification, which expects two input columns: score and label.

Constructor

new MulticlassClassificationEvaluator(uidopt)

Parameters:
Name Type Attributes Description
uid string <optional>
Source:

Methods

(static) load(path) → {module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator}

Parameters:
Name Type Description
path string
Source:
Returns:
Type
module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator

copy(extra) → {module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator}

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.
Parameters:
Name Type Description
extra module:eclairjs/ml/param.ParamMap
Source:
Returns:
Type
module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator

evaluate(dataset) → {number}

Evaluates the output.
Parameters:
Name Type Description
dataset module:eclairjs/sql.DataFrame
Source:
Returns:
Type
number

getMetricName() → {string}

Source:
Returns:
Type
string

isLargerBetter() → {boolean}

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.
Source:
Returns:
Type
boolean

metricName() → {module:eclairjs/ml/param.Param}

param for metric name in evaluation (supports "f1" (default), "precision", "recall", "weightedPrecision", "weightedRecall")
Source:
Returns:
Type
module:eclairjs/ml/param.Param

setLabelCol(value) → {module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator}

Parameters:
Name Type Description
value string
Source:
Returns:
Type
module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator

setMetricName(value) → {module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator}

Parameters:
Name Type Description
value string
Source:
Returns:
Type
module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator

setPredictionCol(value) → {module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator}

Parameters:
Name Type Description
value string
Source:
Returns:
Type
module:eclairjs/ml/evaluation.MulticlassClassificationEvaluator

uid() → {string}

An immutable unique ID for the object and its derivatives.
Source:
Returns:
Type
string