How to define x and y in this neural code ??

Hi All

I have this code , but I literally do not know how to work with it , I am not yet familiar so much with neural networks . I'd please ask someone to explain me how to work with it :

clc clear clear all load x load y %normalizing data [xn_tr,xs_tr] = mapstd(x); [yn_tr,ys_tr] = mapstd(y); %% network net=newff(xn_tr,yn_tr,[7 7],{'tansig'},'traingda');%7 hidden tanh layer gradian descent adaptive net.trainParam.epochs =70; net.trainParam.lr = 0.05; net.trainParam.lr_inc = 1.05; %training network net.trainFcn='traingda'; [net,tr] = train(net,xn_tr,yn_tr); %randomizing initial value f weight matrix net = init(net); net.trainParam.show = NaN; u_t=mapstd('apply',x,xs_tr); %simulating output y_hat=sim(net,u_t); %plotting performance plotperform(tr) mse=mse(y-y_hat)

; ; ;;;;;;;;;;;;;;;;;;,;;;;;;;;;;;;;;;;;;;;;;;;;;;

I have an input vector as : [alfa , t, h, dp] and the output is a 6 by 6 matrix

I would like to know first , how to work with the code , to define these inputs , maybe they are the x and y in the code I don't know

and what is the precision of this case, I mean regarding to the ratio of the output parameters to input parameters

thank yo so much

[Greg Heath]

1. When you post code, always format it so that it can be run when cut and pasted into the command line.

2. Also include the performance on one of the MATLAB example datasets.

 help nndatasets
 doc nndatasets

NEWFF is obsolete. Type the command

 ver

to see what version of MATLAB you have. Do you have FITNET (regression) and PATTERNNET (classification)? Type

 help fitnet
 doc fitnet

etc

[farzad]

Thank you Greg so much

I actually had problem and Did not know who to put that code correctly that everyone could copy it correctly , I just have attached the file as well

I won't use toolbox , I will use this code

yes matlab has FITNET and PATTERNET

so reading the nndatasets , will give me all the answers I got about this code ?

I also don't get notification email when I receive a comment ,do I need a correction in settings ?

Answer 1

[Greg Heath]

I don't recommend that code. Use FITNET or PATTERNNET.

 The documentations and trivial examples are readily available using the help and doc commands.

Additional practice datasets are available via

 help datasets
 doc datasets

Less trivial examples are obtained via NEWSGROUP or ANSWER searches.

Since you are going to spend a nontrivial considerable amount of time, don't waste it

Greg

[farzad]

Thank you so much Greg

Sure I will follow your advices