What activation function does fullyConnectedLayer uses?

I checked it here: https://se.mathworks.com/help/deeplearning/ref/nnet.cnn.layer.fullyconnectedlayer.html

It says it multiple the input with weight and add bias but I couldn't find any thing related to activation function. Can anyone help me on it?

[Mohammad Sami]

A prior answer from 2017 indicates that activation layer has to be added in as a layer.

Please see this for details.

https://www.mathworks.com/matlabcentral/answers/346488-how-to-change-activation-function-for-fully-connected-layer-in-convolutional-neural-network

Answer 1

[Srivardhan Gadila]

As mentioned by @Mohammad Sami, In order for an activation function after fullyConnectedLayer, you have to include an activation layer after the fullyConnectedLayer in your layers/layerGraph array.

Refer to Activation Layers for list of available activation layers in Deep Learning Toolbox & layerGraph.