[R] neural networks

From: Donna Tucker <ladylilly_at_hotmail.com>
Date: Wed, 02 Jul 2008 18:24:52 -0500

Hello all,
I am trying to use the neuralnet function from the neural package to train neural networks. I am finishing my thesis over statistical learning in which I am comparring the MSE values for various algorithms with random data sets, so I do not have a specific application. I have two issues that I am looking for help with.  

  1. When I train a network with a real data set, I ALWAYS get the mean value as all of the predictions! I get beautiful results when I use a generated data set, but only the mean for real data. Is there any explanation of this?
  2. For some data sets I get this error: "Error in neurons[[length.weights]] : attempt to select less than one element" I can not find anything about what this means.

I would be grateful for any help over either or both of these issues.  

Thanks,
Donna



[[elided Hotmail spam]]

        [[alternative HTML version deleted]]



R-help_at_r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Wed 02 Jul 2008 - 23:28:30 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Wed 02 Jul 2008 - 23:31:22 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive