how to form target vector for neural network?

I have an eeg data set containing data of five tasks performed by a single subject. The data is collected from 7 electrodes for each task. Data size for each task is 2500x7. I have concatenated the data for all tasks to form 12500x7 matrix, to be classified using neural network. My input matrix is the 12500x7 matrix. how to form the target vector?

Answer 1

[Greg Heath]

If N I-dimensional vectors are to be classified into one of c exclusive classes, the target vectors should be {0,1} c-dimensional unit column vectors from the unit matrix eye(c).

 [ I N ] = size(input)
 [ c N ] = size(target)

See

 help patternnet

 doc patternnet

for documentation examples.

Hope this helps.

Thank you for formally accepting my answer

Greg

Answer 2

[Walter Roberson]

That depends: what are you trying to classify or predict? If you had a whole bunch of these datasets and your task was to figure out from measurements who was being tested, then you should be reshaping that 12500x7 into a single vector and making similar vectors for other subjects and your target vector would be identity number of each of the subjects.

[Santa Maria]

i have to classify the particular task, each element in my matrix belongs to