how to pass matrix size 4x600 type double, as input neural network?

i trained multiple images (24 images) as training images, and save the rows and columns in empty matrix T=[], after training process finished , the rows and columns and save them into variable zi, so emptymatrix T=[T zi] . i checked in workspace the size of T is 4x575 double, and checked in command window the column through 575, too. i did the same process for testingimages (16 images) and the size of T for testingimages is 4x384double shown in workspace. how to pass 4 x 575 as the input and 4 x 384 target in neural network ? and each 4 rows intend to 1 class .

Answer 1

[Greg Heath]

When you have each of N O-dimensional "O"utput target vectors corresponding to one of N I-dimensional "I"nput vectors, the data should be presented to the NN in the following matrix form

 [ I N ] = size(input) 

 [ O N ] = size(target)

After training, the output matrix will have the same size as the target matrix.

Hope this helps.

Thank you for formally accepting my answer

Greg