Mercurial > hg > camir-aes2014
annotate toolboxes/FullBNT-1.0.7/bnt/examples/static/dtree/test_zoo1.m @ 0:e9a9cd732c1e tip
first hg version after svn
| author | wolffd |
|---|---|
| date | Tue, 10 Feb 2015 15:05:51 +0000 |
| parents | |
| children |
| rev | line source |
|---|---|
| wolffd@0 | 1 % Here the training data is adapted from UCI ML repository, 'zoo' data |
| wolffd@0 | 2 |
| wolffd@0 | 3 dtreeCPD=tree_CPD; |
| wolffd@0 | 4 |
| wolffd@0 | 5 % load data |
| wolffd@0 | 6 fname = fullfile(BNT_HOME, 'examples', 'static', 'uci_data', 'zoo', 'zoo1.data') |
| wolffd@0 | 7 data=load(fname); |
| wolffd@0 | 8 data=data'; |
| wolffd@0 | 9 |
| wolffd@0 | 10 data=transform_data_into_bnt_format(data, []); |
| wolffd@0 | 11 |
| wolffd@0 | 12 % learn decision tree from data |
| wolffd@0 | 13 ns=2*ones(1,17); |
| wolffd@0 | 14 ns(13)=6; |
| wolffd@0 | 15 ns(17)=7; |
| wolffd@0 | 16 dtreeCPD1=learn_params(dtreeCPD,1:17,data,ns,[],'stop_cases',5); % a node with less than 5 cases will not be splitted |
| wolffd@0 | 17 |
| wolffd@0 | 18 % evaluate on data |
| wolffd@0 | 19 [score,outputs]=evaluate_tree_performance(dtreeCPD1,1:17,data,ns,[]); |
| wolffd@0 | 20 fprintf('Accuracy in old training data %6.3f\n',score); |
| wolffd@0 | 21 |