Neural Network with multiple Outputs

Hi everybody :) !!

I'm creating a neural network with multiple inputs and multiple outputs. I had no problem with the input since I reshaped it into one vector and then I used the con2seq command to change it to a cell array with the number of column as samples. 1xN samples. Now the Problem is that my outputs are three matrices: Output1: 263x3 (it's a temperatur as a function of time) Output2: 263x3 (another temperature-function as a function of time) Output3: 263x3 (264 time steps)

the number of column is 3 because it's actually the number of sensors that I'm using in my experiment and the number of rows is 263 ... it's the time steps. How do I have to define my outputs since it is multidimensional?

Thank you !

Answer 1

[Greg Heath]

For N I-dimensional "I"nputs yielding N O-dimensional "O"utputs

 [ I N ] = size(input)

 [ O N ] = size(target)

You have some rearranging to do.

Hope this helps.

Thank you for formally accepting my answer

Greg

[Mohamed Amine Chaaleb]

can you plz explain more ?