MATLAB: In neural network, how can I change the sigmoid activation function to a Gaussian one, How to modify the Matlab codes to achieve this

activation functionDeep Learning Toolboxneural network

that is, change this function: a = tansig(n) = 2/(1+exp(-2*n))-1 to a = gauss(n) = exp(-(n.^2))

Best Answer

 net = fitnet;
 net.layers{1}.transferFcn = 'radbas';

Hope this helps.

Thank you for formally accepting my answer

Greg

Related Solutions

MATLAB: How to create a neural network with 1 layer only (no hidden layers)

 net = fitnet([]);     % For regression
 net = patternnet([]); % For classification

Hope this helps.

Thank you for formally accepting my answer

Greg

MATLAB: Neural network fitting for PWM synthesis

According to your code:

net = network( ...
1, ... % numInputs, number of inputs,
2, ... % numLayers, number of layers
[1; 0], ... % biasConnect, numLayers-by-1 Boolean vector,
[1; 0], ... % inputConnect, numLayers-by-numInputs Boolean matrix,
[0 0; 1 0], ... % layerConnect, numLayers-by-numLayers Boolean matrix
[0 1] ... % outputConnect, 1-by-numLayers Boolean vector
);

The network has only 2 layers.

>> view(net)
>> net.layers
ans =
  2×1 cell array
    {1×1 nnetLayer}
    {1×1 nnetLayer}

Refer to the following for more information: network, Neural Network Object Properties, Neural Network Subobject Properties

Best Answer

Related Solutions

MATLAB: How to create a neural network with 1 layer only (no hidden layers)

MATLAB: Neural network fitting for PWM synthesis

Related Question