R help - Jul 2005 - about nnet package

Dear All,

I'm learning to train a neural network with my training data by using nnet
package, then evaluate it with a evaluation set. My problem here is that, I need
the trained network to be used in future, so, what should I store? and How? Any
other options other than nnet package? Any example will be highly appreciated!

Best, 
 Baoqiang Cao

Baoqiang Cao wrote:
> Dear All,
> 
> I'm learning to train a neural network with my training data by using
nnet package, then evaluate it with a evaluation set. My problem here is that, I
need the trained network to be used in future, so, what should I store? and How?
Any other options other than nnet package? Any example will be highly
appreciated!
> 
> Best, 
>  Baoqiang Cao

See ?nnet which point you to its predict method.

You say

nnetObject <- nnet(.....)

Now store the nnetObject and use it later as in:

predict(nnetObject, newdata)



Uwe Ligges





> 
> 
> ------------------------------------------------------------------------
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html

Hello Uwe Ligges,
    I might misdiscribe my question. What I thought is that, after
nnetObject<-nnet(...), this nnetObject will be gone after I exit R. Since I
need the trained nnetObject for next time without training it again, how can I
save this nnetObject (to some files I guess)? Thanks.

Best regards, 
  Baoqiang Cao

======= At 2005-07-22, 11:17:37 you wrote: ======
>Baoqiang Cao wrote:
>
>> Dear All,
>> 
>> I'm learning to train a neural network with my training data by
using nnet package, then evaluate it with a evaluation set. My problem here is
that, I need the trained network to be used in future, so, what should I store?
and How? Any other options other than nnet package? Any example will be highly
appreciated!
>> 
>> Best, 
>>  Baoqiang Cao
>
>
>See ?nnet which point you to its predict method.
>
>You say
>
>nnetObject <- nnet(.....)
>
>Now store the nnetObject and use it later as in:
>
>predict(nnetObject, newdata)
>
>
>
>Uwe Ligges
>
>
>
>
>
>
>> 
>> 
>>
------------------------------------------------------------------------
>> 
>> ______________________________________________
>> R-help at stat.math.ethz.ch mailing list
>> https://stat.ethz.ch/mailman/listinfo/r-help
>> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html
>
>.
= = = = = = = = = = = = = = = = = = = =