Namespaces

Types in MathNet.Numerics

Type Distance

Namespace MathNet.Numerics

Metrics to measure the distance between two structures.

Static Functions

Public Static Functions

float Canberra(Single[] a, Single[] b)

Canberra Distance, a weighted version of the L1-norm of the difference.

double Canberra(Double[] a, Double[] b)

Canberra Distance, a weighted version of the L1-norm of the difference.

float Chebyshev(Single[] a, Single[] b)

Chebyshev Distance, i.e. the Infinity-norm of the difference.

double Chebyshev(Double[] a, Double[] b)

Chebyshev Distance, i.e. the Infinity-norm of the difference.

double Chebyshev<T>(Vector<T> a, Vector<T> b)

Chebyshev Distance, i.e. the Infinity-norm of the difference.

float Cosine(Single[] a, Single[] b)

Cosine Distance, representing the angular distance while ignoring the scale.

double Cosine(Double[] a, Double[] b)

Cosine Distance, representing the angular distance while ignoring the scale.

float Euclidean(Single[] a, Single[] b)

Euclidean Distance, i.e. the L2-norm of the difference.

double Euclidean(Double[] a, Double[] b)

Euclidean Distance, i.e. the L2-norm of the difference.

double Euclidean<T>(Vector<T> a, Vector<T> b)

Euclidean Distance, i.e. the L2-norm of the difference.

float Hamming(Single[] a, Single[] b)

Hamming Distance, i.e. the number of positions that have different values in the vectors.

double Hamming(Double[] a, Double[] b)

Hamming Distance, i.e. the number of positions that have different values in the vectors.

double Jaccard(Single[] a, Single[] b)

Jaccard distance, i.e. 1 - the Jaccard index.
Return
double

Jaccard distance.

double Jaccard(Double[] a, Double[] b)

Jaccard distance, i.e. 1 - the Jaccard index.
Return
double

Jaccard distance.

float MAE(Single[] a, Single[] b)

Mean-Absolute Error (MAE), i.e. the normalized L1-norm (Manhattan) of the difference.

double MAE(Double[] a, Double[] b)

Mean-Absolute Error (MAE), i.e. the normalized L1-norm (Manhattan) of the difference.

double MAE<T>(Vector<T> a, Vector<T> b)

Mean-Absolute Error (MAE), i.e. the normalized L1-norm (Manhattan) of the difference.

double Manhattan(Double[] a, Double[] b)

Manhattan Distance, i.e. the L1-norm of the difference.

float Manhattan(Single[] a, Single[] b)

Manhattan Distance, i.e. the L1-norm of the difference.

double Manhattan<T>(Vector<T> a, Vector<T> b)

Manhattan Distance, i.e. the L1-norm of the difference.

double Minkowski(double p, Double[] a, Double[] b)

Minkowski Distance, i.e. the generalized p-norm of the difference.

float Minkowski(double p, Single[] a, Single[] b)

Minkowski Distance, i.e. the generalized p-norm of the difference.

double Minkowski<T>(double p, Vector<T> a, Vector<T> b)

Minkowski Distance, i.e. the generalized p-norm of the difference.

double MSE(Double[] a, Double[] b)

Mean-Squared Error (MSE), i.e. the normalized squared L2-norm (Euclidean) of the difference.

float MSE(Single[] a, Single[] b)

Mean-Squared Error (MSE), i.e. the normalized squared L2-norm (Euclidean) of the difference.

double MSE<T>(Vector<T> a, Vector<T> b)

Mean-Squared Error (MSE), i.e. the normalized squared L2-norm (Euclidean) of the difference.

double Pearson(IEnumerable<double> a, IEnumerable<double> b)

Pearson's distance, i.e. 1 - the person correlation coefficient.

float SAD(Single[] a, Single[] b)

Sum of Absolute Difference (SAD), i.e. the L1-norm (Manhattan) of the difference.

double SAD(Double[] a, Double[] b)

Sum of Absolute Difference (SAD), i.e. the L1-norm (Manhattan) of the difference.

double SAD<T>(Vector<T> a, Vector<T> b)

Sum of Absolute Difference (SAD), i.e. the L1-norm (Manhattan) of the difference.

double SSD(Double[] a, Double[] b)

Sum of Squared Difference (SSD), i.e. the squared L2-norm (Euclidean) of the difference.

float SSD(Single[] a, Single[] b)

Sum of Squared Difference (SSD), i.e. the squared L2-norm (Euclidean) of the difference.

double SSD<T>(Vector<T> a, Vector<T> b)

Sum of Squared Difference (SSD), i.e. the squared L2-norm (Euclidean) of the difference.