# denToZero¶

## Purpose¶

Converts every denormal to a 0 in a matrix or array.

## Format¶

y = denToZero(x)
Parameters: x (matrix or an N-dimensional array) – Data to be converted. y (matrix or an N-dimensional array) – with the same orders as the input. Every denormal in the input will be converted to 0 in the output.

## Examples¶

x =  1 | exp(-724.5) | 3;

// If 'x' contains any denormals set them to 0
if isden(x);
y = denToZero(x);
endif;


After the first line above, x is equal to:

1.000e+000
+DEN
3.000e+000


At the end of the example, y is equal to:

1.000e+000
0.000e+000
3.000e+000