From: Rajeev Kumar on
Hi to all,
i am designing feedforward backpropagation network throgh matlab. I am new to matlab.if i have the input samles like this
p=[2 3 ];
t=[0.1 0.2 0.3 0.4 ;
0.6 0.7 0.8 0.9 ];
then matlab dont take these samples.becoz the input columns and output columns must be same.if i put these values repeatedly just like
p =[2 2 3 3];
t=[0.1 0.2 0.3 0.4 ;
0.1 0.2 0.3 0.4 ;
0.6 0.7 0.8 0.9 ;
0.6 0.7 0.8 0.9];
to make the no.of input columns equal to the output columns.or
i add zeroes in the input and ouput matrix just like
p=[0 0 2 3];
t=[0 0 0 0;
0 0 0 0;
0.1 0.2 0.3 0.4;
0.5 0.6 0.7 0.8]; to make input and ouput columns same.
becoz i have calculated if i have 0 input then in output i have 0 0 0 0 i.e four outputs.
are these ways of entering the matrix are correct,if yes then which one,if no then please suggest me. at er.rajeevdogra(a)rediff.com
I shall be thankful to you.
From: Max on
Which function are these parameters for? Please, explain better, what you are trying to do and what you get.
From: Rajeev Kumar on











Hi to all,
i am designing feedforward backpropagation network throgh matlab. I am new to matlab.if i have the input ,output samples like this
p=[2 3 ];
t=[0.1 0.2 0.3 0.4 ;
0.6 0.7 0.8 0.9 ];
then matlab dont take these samples.becoz the input columns and output columns must be same. matlab gives error as ??? Error using ==> network.train at 228
Targets are incorrectly sized for network.
Matrix must have 4 rows.
if i put values repeatedly just like
p =[2 2 3 3];
t=[0.1 0.2 0.3 0.4 ;
0.1 0.2 0.3 0.4 ;
0.6 0.7 0.8 0.9 ;
0.6 0.7 0.8 0.9];
to make the no.of input columns equal to the output columns.
or i add zeroes in the input and ouput matrix just like
p=[0 0 2 3];
t=[0 0 0 0;
0 0 0 0;
0.1 0.2 0.3 0.4;
0.5 0.6 0.7 0.8]; to make input and ouput columns same.
becoz i have calculated if i have 0 input then in output i have 0 0 0 0 i.e four outputs.
are these ways of entering the matrix are correct,if yes then which one,if no then please suggest me. at er.rajeevdogra(a)rediff.com
I shall be thankful to you.