Daniel@0: % Here the training data is adapted from UCI ML repository, 'housing' data
Daniel@0: % Input variables: 12 continous, one binary
Daniel@0: % Ouput variables: continous
Daniel@0: % The testing result trace is in the end of this script, it is same to the graph in page 219 of 
Daniel@0: % Leo Brieman etc. 1984 book titled "Classification and regression trees".
Daniel@0: 
Daniel@0: dtreeCPD=tree_CPD;
Daniel@0: 
Daniel@0: % load data
Daniel@0: fname = fullfile(BNT_HOME, 'examples', 'static', 'uci_data', 'housing', 'housing.data');
Daniel@0: data=load(fname);
Daniel@0: data=data';
Daniel@0: data=transform_data_into_bnt_format(data,[1:3,5:14]); 
Daniel@0: 
Daniel@0: % learn decision tree from data 
Daniel@0: ns=1*ones(1,14);
Daniel@0: ns(4)=2;
Daniel@0: dtreeCPD1=learn_params(dtreeCPD,1:14,data,ns,[1:3,5:14],'stop_cases',5,'min_gain',0.006); 
Daniel@0: 
Daniel@0: % evaluate on data
Daniel@0: [score,outputs]=evaluate_tree_performance(dtreeCPD1,1:14,data,ns,[1:3,5:14]);
Daniel@0: fprintf('Mean square deviation (using regression tree to predict) in old training data %6.3f\n',score);
Daniel@0: 
Daniel@0: 
Daniel@0: % show decision tree using graphpad
Daniel@0: % It should be easy, but still not implemented
Daniel@0: 
Daniel@0: 
Daniel@0: 
Daniel@0: % >> test_housing
Daniel@0: % Create node 1 split at 6 gain  38.2205 Th 6.939000e+000. Mean  22.5328 Cases 506
Daniel@0: % Create node 2 split at 13 gain  14.4503 Th 1.437000e+001. Mean  19.9337 Cases 430
Daniel@0: % Create node 3 split at 8 gain   4.9809 Th 1.358000e+000. Mean  23.3498 Cases 255
Daniel@0: % Create node 4 split at 1 gain   0.7722 Th 1.023300e+001. Mean  45.5800 Cases 5
Daniel@0: % Create leaf node(samevalue) 5. Mean  50.0000 Std   0.0000 Cases 4 
Daniel@0: % Add subtree node 5 to 4. #nodes 5
Daniel@0: % Create leaf node(samevalue) 6. Mean  27.9000 Std   0.0000 Cases 1 
Daniel@0: % Add subtree node 6 to 4. #nodes 6
Daniel@0: % Add subtree node 4 to 3. #nodes 6
Daniel@0: % Create node 7 split at 6 gain   2.8497 Th 6.540000e+000. Mean  22.9052 Cases 250
Daniel@0: % Create node 8 split at 13 gain   0.5970 Th 7.560000e+000. Mean  21.6297 Cases 195
Daniel@0: % Create leaf node(nogain) 9. Mean  23.9698 Std   1.7568 Cases 43 
Daniel@0: % Add subtree node 9 to 8. #nodes 9
Daniel@0: % Create leaf node(nogain) 10. Mean  20.9678 Std   2.8242 Cases 152 
Daniel@0: % Add subtree node 10 to 8. #nodes 10
Daniel@0: % Add subtree node 8 to 7. #nodes 10
Daniel@0: % Create leaf node(nogain) 11. Mean  27.4273 Std   3.4512 Cases 55 
Daniel@0: % Add subtree node 11 to 7. #nodes 11
Daniel@0: % Add subtree node 7 to 3. #nodes 11
Daniel@0: % Add subtree node 3 to 2. #nodes 11
Daniel@0: % Create node 12 split at 1 gain   2.2467 Th 6.962150e+000. Mean  14.9560 Cases 175
Daniel@0: % Create node 13 split at 5 gain   0.5172 Th 5.240000e-001. Mean  17.1376 Cases 101
Daniel@0: % Create leaf node(nogain) 14. Mean  20.0208 Std   3.0672 Cases 24 
Daniel@0: % Add subtree node 14 to 13. #nodes 14
Daniel@0: % Create leaf node(nogain) 15. Mean  16.2390 Std   2.9746 Cases 77 
Daniel@0: % Add subtree node 15 to 13. #nodes 15
Daniel@0: % Add subtree node 13 to 12. #nodes 15
Daniel@0: % Create node 16 split at 5 gain   0.6133 Th 6.050000e-001. Mean  11.9784 Cases 74
Daniel@0: % Create leaf node(nogain) 17. Mean  16.6333 Std   4.5052 Cases 12 
Daniel@0: % Add subtree node 17 to 16. #nodes 17
Daniel@0: % Create leaf node(nogain) 18. Mean  11.0774 Std   3.0090 Cases 62 
Daniel@0: % Add subtree node 18 to 16. #nodes 18
Daniel@0: % Add subtree node 16 to 12. #nodes 18
Daniel@0: % Add subtree node 12 to 2. #nodes 18
Daniel@0: % Add subtree node 2 to 1. #nodes 18
Daniel@0: % Create node 19 split at 6 gain   6.0493 Th 7.420000e+000. Mean  37.2382 Cases 76
Daniel@0: % Create node 20 split at 1 gain   1.9900 Th 7.367110e+000. Mean  32.1130 Cases 46
Daniel@0: % Create node 21 split at 8 gain   0.6273 Th 1.877300e+000. Mean  33.3488 Cases 43
Daniel@0: % Create leaf node(samevalue) 22. Mean  45.6500 Std   6.1518 Cases 2 
Daniel@0: % Add subtree node 22 to 21. #nodes 22
Daniel@0: % Create leaf node(nogain) 23. Mean  32.7488 Std   3.5690 Cases 41 
Daniel@0: % Add subtree node 23 to 21. #nodes 23
Daniel@0: % Add subtree node 21 to 20. #nodes 23
Daniel@0: % Create leaf node(samevalue) 24. Mean  14.4000 Std   3.7363 Cases 3 
Daniel@0: % Add subtree node 24 to 20. #nodes 24
Daniel@0: % Add subtree node 20 to 19. #nodes 24
Daniel@0: % Create node 25 split at 1 gain   1.1001 Th 2.733970e+000. Mean  45.0967 Cases 30
Daniel@0: % Create leaf node(nogain) 26. Mean  45.8966 Std   4.4005 Cases 29 
Daniel@0: % Add subtree node 26 to 25. #nodes 26
Daniel@0: % Create leaf node(samevalue) 27. Mean  21.9000 Std   0.0000 Cases 1 
Daniel@0: % Add subtree node 27 to 25. #nodes 27
Daniel@0: % Add subtree node 25 to 19. #nodes 27
Daniel@0: % Add subtree node 19 to 1. #nodes 27
Daniel@0: % Mean square deviation (using regression tree to predict) in old training data  9.405
Daniel@0: % 
Daniel@0: 
Daniel@0: