Type Alias ModelMetricsMulticlass

Multiclass metrics for categorical values with 3+ options. Details: https://help.qlik.com/en-US/cloud-services/Subsystems/Hub/Content/Sense_Hub/AutoML/scoring-multiclass-classification.htm

type ModelMetricsMulticlass = {
    accuracy?: number;
    accuracyTest?: number;
    confusionMatrix?: string;
    confusionMatrixTest?: string;
    f1Macro?: number;
    f1MacroTest?: number;
    f1Micro?: number;
    f1MicroTest?: number;
    f1Weighted?: number;
    f1WeightedTest?: number;
}
Index

Properties

accuracy? accuracyTest? confusionMatrix? confusionMatrixTest? f1Macro? f1MacroTest? f1Micro? f1MicroTest? f1Weighted? f1WeightedTest?

Properties

accuracy?: number

Average of how often the model made a correct prediction (training data)

accuracyTest?: number

Average of how often the model made a correct prediction (test data)

confusionMatrix?: string

A matrix summary of the accuracy of predictions in a classification model (training data)

confusionMatrixTest?: string

A matrix summary of the accuracy of predictions in a classification model (training data)

f1Macro?: number

Macro F1 is the averaged F1 value for each class without weighting (training data)

f1MacroTest?: number

Macro F1 is the averaged F1 value for each class without weighting (test data)

f1Micro?: number

Micro F1 is the F1 value calculated across the entire confusion matrix (training data)

f1MicroTest?: number

Micro F1 is the F1 value calculated across the entire confusion matrix (test data)

f1Weighted?: number

Weighted F1 (training data)

f1WeightedTest?: number

Weighted F1 (test data)

Settings

Member Visibility

On This Page

Properties