Adding inputs to CNNs after convolutional layers

I am trying to add another input to a CNN right after the output of the convolutional layers and before the input to the fully connected layer.

Is there a clever way of adding an input in such a way?

Here's how I want it to look

Image input
2dconvultionallayer
RELU layer
maxpooling layer
<add another interger input>
fully connected layer
classification layer

I couldn't find anything in the MATLAB neural networkin documentation.

Thank you.

Answer 1

[Srivardhan Gadila]

Training DAG networks with multiple input layers is not supported. I have heard that this issue is known and the concerned parties are may be investigating on this.

[Srivardhan Gadila]

@Siddharth Varughese, The above answer was w.r.t MATLAB R2019a.

As of MATLAB R2019b please refer to Multiple-Input and Multiple-Output Networks.