MulticlassClassificationCatalog.Evaluate Method

Definition

Namespace:

Microsoft.ML

Assembly:

Microsoft.ML.Data.dll

Package:

Microsoft.ML v3.0.1

Evaluates scored multiclass classification data.

public Microsoft.ML.Data.MulticlassClassificationMetrics Evaluate (Microsoft.ML.IDataView data, string labelColumnName = "Label", string scoreColumnName = "Score", string predictedLabelColumnName = "PredictedLabel", int topKPredictionCount = 0);

member this.Evaluate : Microsoft.ML.IDataView * string * string * string * int -> Microsoft.ML.Data.MulticlassClassificationMetrics

Public Function Evaluate (data As IDataView, Optional labelColumnName As String = "Label", Optional scoreColumnName As String = "Score", Optional predictedLabelColumnName As String = "PredictedLabel", Optional topKPredictionCount As Integer = 0) As MulticlassClassificationMetrics

Parameters

data

IDataView

The scored data.

labelColumnName

String

The name of the label column in data.

scoreColumnName

String

The name of the score column in data.

predictedLabelColumnName

String

The name of the predicted label column in data.

topKPredictionCount

Int32

If given a positive value, the TopKAccuracy will be filled with the top-K accuracy, that is, the accuracy assuming we consider an example with the correct class within the top-K values as being stored "correctly."

Returns

MulticlassClassificationMetrics

The evaluation results for these calibrated outputs.