vmlrelud - ReLU activation (first derivative)

4.6 vmlrelud - ReLU activation (first derivative)

#include <mvecml.h> 
 
void vmlrelud (int n, double *restrict y, int incy, const double *restrict x, int incx); 
void vmlreludf(int n, float *restrict y, int incy, const float *restrict x, int incx); 
 
void vmlrelud_64(long n, double *restrict y, long incy, const double *restrict x, long incx); 
void vmlreludf_64(long n, float *restrict y, long incy, const float *restrict x, long incx);

Given an input vector $x$ and a result vector $y$ this function computes the first derivative of the ReLU activation function of the values in the $x$ vector and stores the result in the $y$ vector.

{ y .= 0 where x< 0∨ x= 0 1 where x> 0

The case where $x =0$ is normally undefined, for proper behavior in the application scenario of machine learning this function substitutes $0$ as a result on those elements to handle this condition gracefully.

4.6.1 Parameters

N - INTEGER: ENTRY: Number of elements of $x$ and $y$ .
CONSTRAINT: $n ≥1$ .
Y - ARRAY OF REAL: EXIT: Result vector $y$ .
CONSTRAINT: Must contain $n× incy$ elements.
CONSTRAINT: Must not overlap with array $x$ .
INCY - INTEGER: ENTRY: Stride for the vector $y$ .
CONSTRAINT: $incy̸= 0$ .
BEHAVIOR: A negative stride will traverse the array in reverse.
X - ARRAY OF REAL: ENTRY: Input vector $x$ .
CONSTRAINT: Must contain $n× incx$ elements.
CONSTRAINT: Must not overlap with array $y$ .
INCX - INTEGER: ENTRY: Stride for the vector $x$ .
CONSTRAINT: $incx̸= 0$ .
BEHAVIOR: A negative stride will traverse the array in reverse.

[next] [prev] [prev-tail] [front] [up]