need help in tune.nnet

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

need help in tune.nnet

madhurima bhattacharjee
Dear R people,

I want to use the tune.nnet function of e1071 package to tune nnet .
I am unable to understand the parameters of tune.nnet from the e1071 pdf
document.
I have performed nnet on a traindata and want to test it for class
prediction with a testdata.
I want to know the values of size,decay,range etc. parameters for which
the prediction of testdata is best.
Can anyone please tell me how to do this with the tune.nnet function.

Anticipating quick response.

Thanks and Regards,
Madhurima.

______________________________________________
[hidden email] mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
Reply | Threaded
Open this post in threaded view
|

Re: need help in tune.nnet

Tuszynski, Jaroslaw W.
You use tune function to find optimal parameters needed for particular
classification algorithm. I had more experience with tune.svm but, I
would try first to put parameters covering the whole possible range of
each variable (in which algorithm do not crash), for example c(4^-2,
4^-1, 4^0, 4^1, 4^2) look at the results and than narrow down the search
in the best ranges. For allowed ranges of parameters you will need to
experiment or study documentation of your chosen classifier (nnet).

Jarek Tuszynski

-----Original Message-----
From: [hidden email]
[mailto:[hidden email]] On Behalf Of madhurima
bhattacharjee
Sent: Friday, March 10, 2006 5:38 AM
To: [hidden email]; Bioconductor
Subject: [R] need help in tune.nnet

Dear R people,

I want to use the tune.nnet function of e1071 package to tune nnet .
I am unable to understand the parameters of tune.nnet from the e1071 pdf

document.
I have performed nnet on a traindata and want to test it for class
prediction with a testdata.
I want to know the values of size,decay,range etc. parameters for which
the prediction of testdata is best.
Can anyone please tell me how to do this with the tune.nnet function.

Anticipating quick response.

Thanks and Regards,
Madhurima.

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

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