[R] NNet value and convergence

[Uwe Ligges]

Hannah Wood wrote:

> Hi,  
> 
> I'm using the R nnet package and have a few simple (?) questions.  
> 
> What is the "value " that is output after every 10 iterations during the training of the network and how is it calculated? 
> 
> # weights:  177
> initial  value 506.134586 
> iter  10 value 128.222774
> iter  20 value 95.399782
> iter  30 value 87.184564
> ...
> 
> Is the "value" the error, if not, is there any way to see the error values during training?  

 From ?nnet:
"value of fitting criterion plus weight decay term."


> Also, how does the network decide when it has converged?  

More than one criterion. You can adjust the criterions with arguments to 
nnet(), see ?nnet.

Please read the references given in ?nnet for more details:

- Ripley, B. D. (1996) Pattern Recognition and Neural Networks. Cambridge.
- Venables, W. N. and Ripley, B. D. (2002) Modern Applied Statistics 
with S. Fourth edition. Springer.


Uwe Ligges


> Thanks
> 
> Hannah