PrincipalComponentAnalysis.GetQResiduals Method

public Vector<double> GetQResiduals(
	Matrix<double> observations
)

Public Function GetQResiduals ( 
	observations As Matrix(Of Double)
) As Vector(Of Double)

public:
Vector<double>^ GetQResiduals(
	Matrix<double>^ observations
)

member GetQResiduals : 
        observations : Matrix<float> -> Vector<float> 

Parameters

observations  Matrix<Double>

A matrix with one observation per row and the same feature order as the training data.

Return Value

The fitted matrix preprocessing is applied to the observations. Each result is the squared Euclidean norm of the difference between a preprocessed row and its reconstruction using the currently retained NumberOfComponents principal components. The calculation is performed in PCA preprocessing space, so it generally differs from squared reconstruction error in the original measurement units when scaling is enabled.

This method is also available on deployed and deserialized models.

Reference