Class Score

Contains scoring statistics for a given model.

Inheritance

System.Object

Score

Inherited Members

System.Object.Equals(System.Object)

System.Object.Equals(System.Object, System.Object)

System.Object.GetHashCode()

System.Object.GetType()

System.Object.MemberwiseClone()

System.Object.ReferenceEquals(System.Object, System.Object)

Namespace:numl.Supervised

Assembly:numl.dll

Syntax

public class Score

Constructors

View Source

Score()

Initializes a new Score object.

Declaration

public Score()

Properties

View Source

Accuracy

Gets or sets the total test accuracy as defined by the normalized distribution over true vs negative cases.

Declaration

public double Accuracy { get; set; }

Property Value

Type	Description
System.Double

View Source

AUC

Gets the Area Under the Curve value for the current fixed stationary point of the Precision / Recall curve.

Declaration

public double AUC { get; }

Property Value

Type	Description
System.Double

View Source

CoefRMSE

Gets or sets the Coefficient of Variation of the RMSE.

Declaration

public double CoefRMSE { get; set; }

Property Value

Type	Description
System.Double

View Source

CrossEntropy

Gets or sets the cross entropy loss.

Declaration

public double CrossEntropy { get; set; }

Property Value

Type	Description
System.Double

View Source

Examples

Gets or sets the total number of scored examples.

Declaration

public int Examples { get; set; }

Property Value

Type	Description
System.Int32

View Source

Fallout

Gets the Fallout value of the model.

A higher value indicates the model has decreased prediction accuracy, otherwise known as the False-Positive-Rate (FPR).

Declaration

public double Fallout { get; }

Property Value

Type	Description
System.Double

View Source

FalseNegatives

Gets or sets the number of incorrectly scored negative examples.

Declaration

public int FalseNegatives { get; set; }

Property Value

Type	Description
System.Int32

View Source

FalsePositives

Gets or sets the number of incorrectly scored positive examples.

Declaration

public int FalsePositives { get; set; }

Property Value

Type	Description
System.Int32

View Source

FScore

Returns the F Score of the model.

The F Score determines the tradeoff between higher prediction confidence and reducing false negative predictions. (Higher is better).

Declaration

public double FScore { get; }

Property Value

Type	Description
System.Double

View Source

MeanAbsError

Gets or sets the mean absolute error.

Declaration

public double MeanAbsError { get; set; }

Property Value

Type	Description
System.Double

View Source

MSE

Gets or sets the mean squared error of the predictions.

Declaration

public double MSE { get; set; }

Property Value

Type	Description
System.Double

View Source

NormRMSE

Gets or sets the normalised RMSE.

Declaration

public double NormRMSE { get; set; }

Property Value

Type	Description
System.Double

View Source

Precision

Gets the Precision of the model.

A higher precision indicates the model has a higher positive prediction confidence. Also known as the Positive-Predictive-Value (PPV).

Declaration

public double Precision { get; }

Property Value

Type	Description
System.Double

View Source

Recall

Gets the Recall of the model.

A higher recall indicates the model has scored better on reducing false negative predictions. Also known as the Sensitivity or True-Positive-Rate (TPR).

Declaration

public double Recall { get; }

Property Value

Type	Description
System.Double

View Source

RMSE

Gets or sets the Root Mean Squared Error of the predictions.

Declaration

public double RMSE { get; set; }

Property Value

Type	Description
System.Double

View Source

Specificity

Gets the Specificity of the model.

A higher value indicates the model has scored better at classifying negative examples, otherwise known as the True-Negative-Rate (TNR).

Declaration

public double Specificity { get; }

Property Value

Type	Description
System.Double

View Source

SSE

Gets or sets the sum of squared errors of the predictions.

Declaration

public double SSE { get; set; }

Property Value

Type	Description
System.Double

View Source

TotalNegatives

Gets or sets the total number of negative examples.

Declaration

public int TotalNegatives { get; set; }

Property Value

Type	Description
System.Int32

View Source

TotalPositives

Gets or sets the total number of positive examples.

Declaration

public int TotalPositives { get; set; }

Property Value

Type	Description
System.Int32

View Source

TrueNegatives

Gets or sets the number of correctly scored negative examples.

Declaration

public int TrueNegatives { get; set; }

Property Value

Type	Description
System.Int32

View Source

TruePositives

Gets or sets the number of correctly scored positive examples.

Declaration

public int TruePositives { get; set; }

Property Value

Type	Description
System.Int32

Methods

View Source

CombineScores(Score[])

Combines and averages metrics across all the given scores.

Declaration

public static Score CombineScores(params Score[] scores)

Parameters

Type	Name	Description
Score[]	scores	Scores.

Returns

Type	Description
Score

View Source

ComputeCoefRMSE(Vector, Vector)

Computes the Coefficient of Variation of the Root Mean Squared Error for the given inputs.

Declaration

public static double ComputeCoefRMSE(Vector y1, Vector y2)

Parameters

Type	Name	Description
Vector	y1	Predicted values.
Vector	y2	Actual values.

Returns

Type	Description
System.Double	Double.

View Source

ComputeCrossEntropy(Vector, Vector)

Computes the Cross Entropy Loss for the given inputs.

Declaration

public static double ComputeCrossEntropy(Vector y1, Vector y2)

Parameters

Type	Name	Description
Vector	y1	Predicted values.
Vector	y2	Actual values.

Returns

Type	Description
System.Double

View Source

ComputeMeanError(Vector, Vector)

Computes the Mean Absolute Error for the given inputs.

Declaration

public static double ComputeMeanError(Vector y1, Vector y2)

Parameters

Type	Name	Description
Vector	y1	Predicted values.
Vector	y2	Actual values.

Returns

Type	Description
System.Double

View Source

ComputeMSE(Vector, Vector)

Computes the Mean Squared Error of the given inputs.

Declaration

public static double ComputeMSE(Vector y1, Vector y2)

Parameters

Type	Name	Description
Vector	y1	Predicted values.
Vector	y2	Actual values.

Returns

Type	Description
System.Double	Double.

View Source

ComputeNormRMSE(Vector, Vector)

Computes the Normalized Root Mean Squared Error for the given inputs.

Declaration

public static double ComputeNormRMSE(Vector y1, Vector y2)

Parameters

Type	Name	Description
Vector	y1	Predicted values.
Vector	y2	Actual values.

Returns

Type	Description
System.Double	Double.

View Source

ComputeRMSE(Vector, Vector)

Computes the Root Mean Squared Error for the given inputs.

Declaration

public static double ComputeRMSE(Vector y1, Vector y2)

Parameters

Type	Name	Description
Vector	y1	Predicted values.
Vector	y2	Actual values.

Returns

Type	Description
System.Double	Double.

View Source

ComputeSSE(Vector, Vector)

Computes the Mean Squared Error of the given inputs.

Declaration

public static double ComputeSSE(Vector y1, Vector y2)

Parameters

Type	Name	Description
Vector	y1	Predicted values.
Vector	y2	Actual values.

Returns

Type	Description
System.Double	Double.

View Source

ScorePredictions(Vector, Vector, Double, Double)

Scores a set of predictions against the actual values.

Declaration

public static Score ScorePredictions(Vector predictions, Vector actual, double truthLabel = 1, double falseLabel = -1)

Parameters

Type	Name	Description
Vector	predictions	Predicted values.
Vector	actual	Actual values.
System.Double	truthLabel	(Optional) the truth label in the actual vector.
System.Double	falseLabel	(Optional) the false label in the actual vector.

Returns

Type	Description
Score

View Source

ToString()

Returns a string representation of the current Score object.

Declaration

public override string ToString()

Returns

Type	Description
System.String

Overrides

System.Object.ToString()