FullyConnectedLayer class
Fully connected layer
Description
A fully connected layer class containing input and output size, layer name, and weights and bias data.
Construction
returns a fully connected layer, in which the software multiplies the input by a matrix and then adds a bias vector.fullconnectlayer
= fullyConnectedLayer(outputSize
)
returns the fully connected layer, with additional options specified by one or morefullconnectlayer
= fullyConnectedLayer(outputSize
,Name,Value
)Name,Value
pair arguments.
For more details on the name-value pair arguments, seefullyConnectedLayer
.
Input Arguments
Properties
Copy Semantics
Value. To learn how value classes affect copy operations, seeCopying Objects(MATLAB) in the MATLAB®documentation.
Examples
Algorithms
The default for the initial weights is a Gaussian distribution with a mean of 0 and a standard deviation of 0.01. The default for the initial bias is 0. You can manually change the initialization for the weights and bias. SeeSpecify Initial Weight and Biases in Fully Connected Layer.