vjfaisal said:dear leekke1
ya i have aproblem how ,tis complex data could be represented in vector form, to import it in NN tool for analyze. i have the two possiblities here....please try to understand and help me in these...
1) To classify the three/two states of brain, it is rightto enter their respective frequencies as an input . for e.g : to classify the brain state , name ( gamma), which has standard frequency band of , let suppose 19 to 25 hz.
For input vector can i write it like this:
p=[19 20 21 22 23 24 25]
2) second possibility relate to the concept that after we extract our desired signal from brain , then filter it ( bandpass), the frequency range for gamma (19 to 25) hz
at last when we have the last filter out results, let suppose in vector k.
Can we load/import it as a input vector in classification. but here , problem is the last filtered results has vector of 1*236, for this case ,
input vector is 1*236,
How we adjust the target in this case, because in NN matlal tool , should have 2 rows for input vector, and one row for target vector with same i/o columns.
please , understand it , and help in this , so thatai could initiate my work , properly
regards
vjfaisal
payate said:hi leekk8..
do u know how to convert the signal into vector and represent it into neural net..
for your information, the signal is isolated of number and i'll train it using neural net (backprop)..
do u have any sample on that..?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?
We use cookies and similar technologies for the following purposes:
Do you accept cookies and these technologies?