[R] Watching tune parameters for SVM?

[Noah Silverman]

Hi,

I'm switch over from RapidMiner to R.  (The learning curve is steep, but 
there is so much more I can do with R and it runs much faster overall.)

In RapidMiner, I can "tune" a parameter of my svm in a nice cross 
validation loop.  The process will print out the progress as it goes. 
So for a 5x cross tuning for the value of C with auc as my performance 
measure, I see

XV    C    AUC
1    1    .5
2    1    .48
3    1    .51
4    1    .52
1    2    .52
2    2    .54
3    2    .53
4    2    .52
1    3    .6
2    3    .61
3    3    .6
4    3    .59
etc...

RapidMiner then takes the average for each value of C and reports back 
the best one.

RapidMiner will even graph the progress with C on the X axis and AUC on 
the Y axis so I can see how the process is doing.  (This is nice when 
training a large range of C so I can tell if it is worth stopping the 
process early.

I can even train 2 or more variables at a time in a big "grid" of 
combination is 5x cross validation for each combo. i.e. 10 values for C 
and 10 values for gamma gives 100 combinations with 5x cross validation 
for each for a total of 500 executions.

Is there anyway to do something similar to this in R.  I've come across 
the "tune" function, but don't entirely understand all the options. 
Additionally, I don't see how to output the progress as it goes.  (also, 
setting number of XV, averaging of scores, etc.)

Does anybody have any ideas?

Thanks






	[[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.