Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
Drexel University or The Math Forum.
|
|
|
|
Re: Hopfield Neural Network
Posted:
Feb 20, 2013 11:51 AM
|
|
"Prateek " <malikprateek22@gmail.com> wrote in message
news:kg2tb8$lp0$1@newscl01ah.mathworks.com...
> I have been trying to make a Hopfield Network to store the data i have
> which is a matrix of dimension 26X10000 but with no success following is
> the code i am writing and the errors which i am getting please help me out
>>> net=newhop(mat')
> ??? Out of memory. Type HELP MEMORY for your options.
"newhop(T) takes an RxQ matrix of Q target vectors T with element
values of +1 or -1, and returns a new Hopfield recurrent neural
network with stable points at the vectors in T."
So R is 10000 and Q is 26 in this case (since you transposed mat when you
called NEWHOP.) You're trying to take 26 10000-dimension vectors and
generate a network from them? Seems to me you have MUCH too small a data set
to be able to draw any sort of conclusions.
Do you instead want to operate on 10000 26-dimension vectors? That's
probably going to give you a better chance of success, though I'd guess
you'd do better with more data. If so, don't transpose.
*snip*
--
Steve Lord
slord@mathworks.com
To contact Technical Support use the Contact Us link on
http://www.mathworks.com
|
|
|
|
