multiple inputs with nntool

Hi, I'm a new user of Matlab and trying to create a neural network with the nntool as it is much easier with this interface. While I can load many inputs and targets, when I want to create the network I am having a problem loading multiple inputs and targets at the same time (there is the option of loading only one), for ex. with back-propagation. I would really appreciate any help thanks in advance

Answer 1

[mohammed]

The following code import 4 columns as inputs, and 1 column as output

may you can modify as your own dataset

inputs = xlsread('D:\MyData\data.xls', 1, 'A2:D115');

targets = xlsread('D:\MyData\data.xls', 1, 'L2:L115');

inputs=inputs';

targets=targets';

Answer 2

[mohammed]

There is may way to load your data before using nntool

the following URL shows how to load data from text file

http://www.mathworks.com/help/techdoc/import_export/f5-35378.html

after that you can use nntool

[K]

Hi, thanks for the quick response. Though, I have already imported my data from an excel file, which are different columns of the xis and yis. I cannot have them as parallel inputs and outputs of the Network I "create" with the nntool. Any ideas?

Thakns!

Answer 3

[mohammed]

can explain more regrading your dataset?

[K]

Yes! Well, the dataset is imported from an excel file with 4 columns: x1,x2...x4 (which I want to be 4 different inputs at the same time of the neural network) and 3 columns: y1, y2, y3 (which I want to be outputs of the network). However, while I create the network with the nntool, with back-propagation, there is only otion of one input and target data at a time. This is my problem