Sigmoid activation layer in Deep Learning Toolbox?

Hi! Kindly guide why there is no option for defining a Sigmoid / Logistic Nonlinearity like ReLU Nonlinearity in Deep Learning Toolbox? If I have to design a Conv Net with Sigmoid Activations in its various layers, how to define the layer array? Do I need to define Custom Layer? Thanks & regards.

[M. Mubashir Hussain]

Thanks for the kind reply!

Answer 1

[Monika Phadnis]

You can define a custom activation layer in Deep Learning Toolbox following this doc page:

https://www.mathworks.com/help/deeplearning/ug/define-custom-deep-learning-layer.html