randnc

(To be removed) Normalized column weight initialization function

randnc will be removed in a future release. For more information, see Transition Legacy Neural Network Code to dlnetwork Workflows.

For advice on updating your code, see Version History.

Syntax

W = randnc(S,PR)

Description

randnc is a weight initialization function.

W = randnc(S,PR) takes two inputs,

S

Number of rows (neurons)

PR

R-by-2 matrix of input value ranges = [Pmin Pmax]

and returns an S-by-R random matrix with normalized columns.

You can also call this in the form randnc(S,R).

Examples

A random matrix of four normalized three-element columns is generated:

M = randnc(3,4)
M =
    -0.6007   -0.4715   -0.2724    0.5596
    -0.7628   -0.6967   -0.9172    0.7819
    -0.2395    0.5406   -0.2907    0.2747

Version History

Introduced before R2006a

collapse all

See Also

| (Statistics and Machine Learning Toolbox) | (Statistics and Machine Learning Toolbox) | | |

Topics