Objeck Documentation

All Bundles

Metrics

Model evaluation metrics for classification tasks

Operations

Accuracy
ConfusionMatrix
F1Score
MCC
Precision
PrintConfusionMatrix
Recall

Accuracy

Calculates accuracy: (TP + TN) / Total

function : native : Accuracy(predictions:Bool[], actuals:Bool[]) ~ Float

Parameters

Name	Type	Description
predictions	Bool	predicted boolean values
actuals	Bool	actual boolean values

Return

Type	Description
Float	accuracy score between 0.0 and 1.0

ConfusionMatrix

Creates a confusion matrix

function : native : ConfusionMatrix(predictions:Bool[], actuals:Bool[]) ~ Int[,]

Parameters

Name	Type	Description
predictions	Bool	predicted boolean values
actuals	Bool	actual boolean values

Return

Type	Description
Int	2x2 confusion matrix [[TN, FP], [FN, TP]]

F1Score

Calculates F1 score: 2 * (Precision * Recall) / (Precision + Recall)

function : native : F1Score(predictions:Bool[], actuals:Bool[]) ~ Float

Parameters

Name	Type	Description
predictions	Bool	predicted boolean values
actuals	Bool	actual boolean values

Return

Type	Description
Float	F1 score between 0.0 and 1.0

MCC

Calculates Matthews Correlation Coefficient (MCC)

function : native : MCC(predictions:Bool[], actuals:Bool[]) ~ Float

Parameters

Name	Type	Description
predictions	Bool	predicted boolean values
actuals	Bool	actual boolean values

Return

Type	Description
Float	MCC score between -1.0 and 1.0

Precision

Calculates precision: TP / (TP + FP)

function : native : Precision(predictions:Bool[], actuals:Bool[]) ~ Float

Parameters

Name	Type	Description
predictions	Bool	predicted boolean values
actuals	Bool	actual boolean values

Return

Type	Description
Float	precision score between 0.0 and 1.0

PrintConfusionMatrix

Prints a formatted confusion matrix

function : native : PrintConfusionMatrix(matrix:Int[,]) ~ Nil

Parameters

Name	Type	Description
matrix	Int	2x2 confusion matrix

Recall

Calculates recall (sensitivity): TP / (TP + FN)

function : native : Recall(predictions:Bool[], actuals:Bool[]) ~ Float

Parameters

Name	Type	Description
predictions	Bool	predicted boolean values
actuals	Bool	actual boolean values

Return

Type	Description
Float	recall score between 0.0 and 1.0