concur

(To be removed) Create concurrent bias vectors

concur 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

concur(B,Q)


Description

concur(B,Q)

B	S-by-1 bias vector (or an Nl-by-1 cell array of vectors)
Q	Concurrent size

and returns an S-by-B matrix of copies of B (or an Nl-by-1 cell array of matrices).

Examples

Here concur creates three copies of a bias vector.

b = [1; 3; 2; -1];
concur(b,3)

Network Use

To calculate a layer’s net input, the layer’s weighted inputs must be combined with its biases. The following expression calculates the net input for a layer with the netsum net input function, two input weights, and a bias:

n = netsum(z1,z2,b)

The above expression works if Z1, Z2, and B are all S-by-1 vectors. However, if the network is being simulated by sim (or adapt or train) in response to Q concurrent vectors, then Z1 and Z2 will be S-by-Q matrices. Before B can be combined with Z1 and Z2, you must make Q copies of it.

n = netsum(z1,z2,concur(b,q))

Version History

Introduced before R2006a

R2026a: concur will be removed

concur will be removed in a future release.

To learn more about initializing and customizing neural network learnable parameters, see these examples and topics:

• Compare Layer Weight Initializers

• Specify Custom Weight Initialization Function

• Neural Network Weight Tying

• Initialize Learnable Parameters for Model Function

For more information, see Transition Legacy Neural Network Code to dlnetwork Workflows.