Neural Network: Single Input, Multiple Output

Guys,

I need your help ASAP! I have a vector 1x1000 entering my NN, and my target output shoudld be a 7x5x1000 3D matrix, meaning a a 2D 7x5 matrix describes each point of my input vector, since there are 1000 periods, its a 3D matrix 7x5x1000.

How can I do this? Usually NNs have a mulitple input, single output, and I want to do the opposite.

Would appreciate your help!

JC

Answer 1

[Greg Heath]

Unfold your target images to 35x1 column vectors using (:) or reshape. Then

[I N ] = size(input) % [ 1 1000]

[ O N ] =size(target) %[ 35 1000]

Hope this helps.

Thank you for formally accepting my answer

Greg