Can anybody help me to sort out this error?

I am running neural network for cancer data set with 50000 hidden layer neurons but getting following error:

??? Error using ==> horzcat Out of memory. Type HELP MEMORY for your options.

Error in ==> calcgrad at 96 Nc = [N{i,:}];

Error in ==> calcgx at 98 [gB,gIW,gLW] = feval(net.gradientFcn,net,Q,PD,BZ,IWZ,LWZ,N,Ac,gE,TS);

Error in ==> trainscg at 241 gX = -calcgx(net,X,trainV.Pd,Zb,Zi,Zl,N,Ac,El,perf,Q,TS);

Error in ==> network.train at 216 [net,tr] = feval(net.trainFcn,net,tr,trainV,valV,testV);

Error in ==> create_pr_net at 10 [net,tr] = train(net,inputs,targets);

                                  Please help me to solve this error..

Answer 1

[Greg Heath]

Use the default number of hidden nodes.

Post any error messages.

Answer 2

[Jan]

If you take the chance to search the error message in this forum, you will find the standard solutions:

1. install more RAM
2. run a 64 bit version of the OS and Matlab
3. install even more RAM
4. Increase the virtual memory
5. as much RAM as you can effort is a good point to start from

Only a small number of these errors are based on bugs like creating a huge matrix by rand(10000) although a vector rand(1, 10000) was meant.