CedricL <[email protected]> wrote:
> 
> Hello all,
>
> last update of OTB 5.2 seems to remove SVM learning (I am not sure).
>
> Whatever, even if I am a fan of SVM in particular with parameter 
> optimization, I now that Random Forest is also powerfull (may be a little bit 
> less than svm) but strongly faster.
>
> However, because I don't know to much Random Forest I don(t know how to chose 
> best settings in comparison to SVM optimization parameters.
>
> So have you any tricks or paper to read about that ?
>

Hi Cédric,

I would suggest to have a look at this paper for RF parameter tuning:

 http://dx.doi.org/10.1016/j.isprsjprs.2011.11.002

An this one for a comparison of SVM and RF:

 http://dx.doi.org/10.3390/rs70912356

As a rule of thumb, start with

-classifier.rf.nbtrees 100
-classifier.rf.max     25
-classifier.rf.min     25

And eventually tune from there, but in my experience, performances are
not very sensitive to these parameters.

Jordi


> Best
>
>
> -- 

-- 
-- 
Check the OTB FAQ at
http://www.orfeo-toolbox.org/FAQ.html

You received this message because you are subscribed to the Google
Groups "otb-users" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/otb-users?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"otb-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to