• Enumerations

Enumeration for matrix and vector operations

This section describes the enumerations that are used in various matrix and vector methods.

 

ENUM_AVERAGE_MODE

Smoothing type enumeration.

ID

Description

AVERAGE_NONE

No averaging. Results are provided for each label separately

AVERAGE_BINARY

Label 1 result for binary classification

AVERAGE_MICRO

Average error matrix result (confusion matrix)

AVERAGE_MACRO

Average result from the results of the error matrices of each label

AVERAGE_WEIGHTED

Weighted average result

 

ENUM_VECTOR_NORM

Enumeration of vector norms for vector::Norm.

ID

Description

VECTOR_NORM_INF

Inf norm

VECTOR_NORM_MINUS_INF

Minus Inf norm

VECTOR_NORM_P

Norm P

 

ENUM_MATRIX_NORM

Enumeration of matrix norms for matrix::Norm and for obtaining the matrix::Cond matrix condition number.

ID

Description

MATRIX_NORM_FROBENIUS

Frobenius norm

MATRIX_NORM_SPECTRAL

Spectral norm

MATRIX_NORM_NUCLEAR

Nuclear norm

MATRIX_NORM_INF

Inf norm

MATRIX_NORM_P1

P1 norm

MATRIX_NORM_P2

P2 norm

MATRIX_NORM_MINUS_INF

Minus Inf norm

MATRIX_NORM_MINUS_P1

Minus P1 norm

MATRIX_NORM_MINUS_P2

Minus P2 norm

 

ENUM_VECTOR_CONVOLVE

Enumeration for convolution vector::Convolve and cross-correlation vector::Correlate.

ID

Description

VECTOR_CONVOLVE_FULL

Convolve full

VECTOR_CONVOLVE_SAME

Convolve same

VECTOR_CONVOLVE_VALID

Convolve valid

 

ENUM_REGRESSION_METRIC

Enumeration of regression metrics for vector::RegressionMetric.

ID

Description

REGRESSION_MAE

Mean Absolute Error

REGRESSION_MSE

Mean Squared Error

REGRESSION_RMSE

Root Mean Squared Error

REGRESSION_R2

R-Squared

REGRESSION_MAPE

Mean Absolute Percentage Error

REGRESSION_MSPE

Mean Squared Percentage Error

REGRESSION_RMSLE

Root Mean Squared Logarithmic Error

REGRESSION_SMAPE

Symmetric Mean Absolute Percentage Error

REGRESSION_MAXE

Maximal Absolute Error

REGRESSION_MEDE

Median Absolute Error

REGRESSION_MPD

Mean Poisson Deviance

REGRESSION_MGD

Mean Gamma Deviance

REGRESSION_EXPV

Explained Variance

 

ENUM_CLASSIFICATION_METRIC

Enumeration of metrics for classification problems.

ID

Description

CLASSIFICATION_ACCURACY

Model quality in terms of prediction accuracy for all classes

CLASSIFICATION_AVERAGE_PRECISION

Average model accuracy

CLASSIFICATION_BALANCED_ACCURACY

Balanced prediction accuracy

CLASSIFICATION_F1

F1 score. Harmonic mean between the model precision and recall

CLASSIFICATION_JACCARD

Jaccard score

CLASSIFICATION_PRECISION

Model accuracy in predicting true positives for the target class

CLASSIFICATION_RECALL

Model completeness

CLASSIFICATION_ROC_AUC

Area under the error curve

CLASSIFICATION_TOP_K_ACCURACY

Frequency of the correct label appearing at the top of k predicted labels

 

ENUM_LOSS_FUNCTION

Enumeration for loss function calculations vector::Loss.

ID

Description

LOSS_MSE

Root mean square error

LOSS_MAE

Mean Absolute Error

LOSS_CCE

Categorical Crossentropy

LOSS_BCE

Binary Crossentropy

LOSS_MAPE

Mean Absolute Percentage Error

LOSS_MSLE

Mean Squared Logarithmic Error

LOSS_KLD

Kullback-Leibler Divergence

LOSS_COSINE

Cosine similarity/proximity

LOSS_POISSON

Poisson

LOSS_HINGE

Hinge

LOSS_SQ_HINGE

Squared Hinge

LOSS_CAT_HINGE

Categorical Hinge

LOSS_LOG_COSH

Logarithm of the Hyperbolic Cosine

LOSS_HUBER

Huber

 

ENUM_ACTIVATION_FUNCTION

Enumeration for the activation function vector::Activation and for the activation function derivative vector::Derivative.

ID

Description

AF_ELU

Exponential Linear Unit

AF_EXP

Exponential

AF_GELU

Gaussian Error Linear Unit

AF_HARD_SIGMOID

Hard Sigmoid

AF_LINEAR

Linear

AF_LRELU

Leaky Rectified Linear Unit

AF_RELU

REctified Linear Unit

AF_SELU

Scaled Exponential Linear Unit

AF_SIGMOID

Sigmoid

AF_SOFTMAX

Softmax

AF_SOFTPLUS

Softplus

AF_SOFTSIGN

Softsign

AF_SWISH

Swish

AF_TANH

The hyperbolic tangent function

AF_TRELU

Thresholded Rectified Linear Unit

 

ENUM_SORT_MODE

Enumeration of sort types for the Sort function.

ID

Description

SORT_ASCENDING

Sort ascending

SORT_DESCENDING

Sort descending

 

ENUM_MATRIX_AXIS

Enumeration for specifying the axis in all statistical functions for matrices.

ID

Description

AXIS_NONE

The axis is not specified. Calculation is performed over all matrix elements, as if it were a vector (see the Flat method).

AXIS_HORZ

Horizontal axis

AXIS_VERT

Vertical axis