forum.alglib.net
http://forum.alglib.net/

Decision Forest Classification
http://forum.alglib.net/viewtopic.php?f=2&t=508		 Page 1 of 1
Author:  the_max [ Tue Dec 20, 2011 2:26 pm ]
Post subject:  Decision Forest Classification
Hi there,
I'm trying to use alglib for a classification task in cpp. I'm using the decision forest with a 1500-dimensional feature vector (image features, float 0-1) and 40 classes. Therefore I use 5000 samples to train the df. Unfortunately after training the result of all classification tasks is class 0. Also the report errors after training are all 0.00. The input vector xy is built properly according to the manual. I use the following code the build the forest.

Code:
dfbuildrandomdecisionforest(
            xy,
            samples.size(),
            nvars,
            nclasses,
            80,
            0.9f,
            info,
            df,
            rep);


test_c.cpp runs without errors. there are no compiler optimization flags set.

Are there any limitations concerning the feature or class size? Any ideas what might be wrong?

Thanks in advance!

best regards
Markus
Page 1 of 1 All times are UTC
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
http://www.phpbb.com/