From: pipa on
Hi All,
I want to pass few variables to the newrb function. I will be using these variables in the newrb function. Is there a way to do this?

Thanks.
From: Greg Heath on
On Jan 13, 7:06 pm, "pipa " <balwindersi...(a)gmail.com> wrote:
> Hi All,
> I want to pass few variables to the newrb function. I will be using these variables in the newrb function. Is there a way to do this?
>
> Thanks.

Please be more clear in what you want and why.

Hope this helps.

Greg
From: pipa on
Thanks Greg.

I have two set of data training data and test data. I want to plot SSE Vs. "# of neurons" for both the training and test data. Based on this plot, I will choose how many neurons to consider for the RBF network (based on minimum error for the test data).

I can send the training data to the network in the newrb function but I am not sure how to send the test data.

Right now, I am sending all the data (train and test) to the RBF network in newrb function. I modified the newrb function and added few lines which seprate the training and test data within the newrb function after it calls designrb. I have hardwired a number specify the data split in the designrb function.

The above arrangement works but I just want to make it more universal so that I can use it for any split (without hardwiring) of training and test data.

Thanks!