How can I reuse the activations from a layer and then merge it with another coming layer in the same neural network like in the figure?

Answer 1

[Nomit Jangid]

Hi Cinar,

It is not very clear in figure. if you want to add layers output you can use additionLayer function in MATLAB.

https://in.mathworks.com/help/deeplearning/ref/nnet.cnn.layer.additionlayer.html

A depth concatenation layer takes inputs that have the same height and width and concatenates them along the third dimension (the channel dimension).

https://in.mathworks.com/help/deeplearning/ref/nnet.cnn.layer.depthconcatenationlayer.html

I hope this helps!

[Cinar Gulleroglu]

Thank you so much, this was really helpful!