1 | /* |
---|
2 | * This program is free software; you can redistribute it and/or modify |
---|
3 | * it under the terms of the GNU General Public License as published by |
---|
4 | * the Free Software Foundation; either version 2 of the License, or |
---|
5 | * (at your option) any later version. |
---|
6 | * |
---|
7 | * This program is distributed in the hope that it will be useful, |
---|
8 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
---|
9 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
---|
10 | * GNU General Public License for more details. |
---|
11 | * |
---|
12 | * You should have received a copy of the GNU General Public License |
---|
13 | * along with this program; if not, write to the Free Software |
---|
14 | * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA. |
---|
15 | */ |
---|
16 | |
---|
17 | /* |
---|
18 | * J48graft.java |
---|
19 | * Copyright (C) 2007 Geoff Webb & Janice Boughton |
---|
20 | * (adapted from code written by Eibe Frank). |
---|
21 | */ |
---|
22 | |
---|
23 | package weka.classifiers.trees; |
---|
24 | |
---|
25 | import weka.classifiers.Classifier; |
---|
26 | import weka.classifiers.AbstractClassifier; |
---|
27 | import weka.classifiers.Sourcable; |
---|
28 | import weka.classifiers.trees.j48.BinC45ModelSelection; |
---|
29 | import weka.classifiers.trees.j48.C45ModelSelection; |
---|
30 | import weka.classifiers.trees.j48.C45PruneableClassifierTreeG; |
---|
31 | import weka.classifiers.trees.j48.ClassifierTree; |
---|
32 | import weka.classifiers.trees.j48.ModelSelection; |
---|
33 | import weka.core.AdditionalMeasureProducer; |
---|
34 | import weka.core.Capabilities; |
---|
35 | import weka.core.Drawable; |
---|
36 | import weka.core.Instance; |
---|
37 | import weka.core.Instances; |
---|
38 | import weka.core.Matchable; |
---|
39 | import weka.core.Option; |
---|
40 | import weka.core.OptionHandler; |
---|
41 | import weka.core.RevisionUtils; |
---|
42 | import weka.core.Summarizable; |
---|
43 | import weka.core.TechnicalInformation; |
---|
44 | import weka.core.TechnicalInformationHandler; |
---|
45 | import weka.core.Utils; |
---|
46 | import weka.core.WeightedInstancesHandler; |
---|
47 | import weka.core.TechnicalInformation.Field; |
---|
48 | import weka.core.TechnicalInformation.Type; |
---|
49 | |
---|
50 | import java.util.Enumeration; |
---|
51 | import java.util.Vector; |
---|
52 | |
---|
53 | /** |
---|
54 | <!-- globalinfo-start --> |
---|
55 | * Class for generating a grafted (pruned or unpruned) C4.5 decision tree. For more information, see<br/> |
---|
56 | * <br/> |
---|
57 | * Geoff Webb: Decision Tree Grafting From the All-Tests-But-One Partition. In: , San Francisco, CA, 1999. |
---|
58 | * <p/> |
---|
59 | <!-- globalinfo-end --> |
---|
60 | * |
---|
61 | <!-- technical-bibtex-start --> |
---|
62 | * BibTeX: |
---|
63 | * <pre> |
---|
64 | * @inproceedings{Webb1999, |
---|
65 | * address = {San Francisco, CA}, |
---|
66 | * author = {Geoff Webb}, |
---|
67 | * publisher = {Morgan Kaufmann}, |
---|
68 | * title = {Decision Tree Grafting From the All-Tests-But-One Partition}, |
---|
69 | * year = {1999} |
---|
70 | * } |
---|
71 | * </pre> |
---|
72 | * <p/> |
---|
73 | <!-- technical-bibtex-end --> |
---|
74 | * |
---|
75 | <!-- options-start --> |
---|
76 | * Valid options are: <p/> |
---|
77 | * |
---|
78 | * <pre> -U |
---|
79 | * Use unpruned tree.</pre> |
---|
80 | * |
---|
81 | * <pre> -C <pruning confidence> |
---|
82 | * Set confidence threshold for pruning. |
---|
83 | * (default 0.25)</pre> |
---|
84 | * |
---|
85 | * <pre> -M <minimum number of instances> |
---|
86 | * Set minimum number of instances per leaf. |
---|
87 | * (default 2)</pre> |
---|
88 | * |
---|
89 | * <pre> -B |
---|
90 | * Use binary splits only.</pre> |
---|
91 | * |
---|
92 | * <pre> -S |
---|
93 | * Don't perform subtree raising.</pre> |
---|
94 | * |
---|
95 | * <pre> -L |
---|
96 | * Do not clean up after the tree has been built.</pre> |
---|
97 | * |
---|
98 | * <pre> -A |
---|
99 | * Laplace smoothing for predicted probabilities. (note: this option only affects initial tree; grafting process always uses laplace).</pre> |
---|
100 | * |
---|
101 | * <pre> -E |
---|
102 | * Relabel when grafting.</pre> |
---|
103 | * |
---|
104 | <!-- options-end --> |
---|
105 | * |
---|
106 | * @author Janice Boughton (jrbought@csse.monash.edu.au) |
---|
107 | * (based on J48.java written by Eibe Frank) |
---|
108 | * @version $Revision: 6088 $ |
---|
109 | */ |
---|
110 | public class J48graft |
---|
111 | extends AbstractClassifier |
---|
112 | implements OptionHandler, Drawable, Matchable, Sourcable, |
---|
113 | WeightedInstancesHandler, Summarizable, |
---|
114 | AdditionalMeasureProducer, TechnicalInformationHandler { |
---|
115 | |
---|
116 | /** for serialization */ |
---|
117 | static final long serialVersionUID = 8823716098042427799L; |
---|
118 | |
---|
119 | /** The decision tree */ |
---|
120 | private ClassifierTree m_root; |
---|
121 | |
---|
122 | /** Unpruned tree? */ |
---|
123 | private boolean m_unpruned = false; |
---|
124 | |
---|
125 | /** Confidence level */ |
---|
126 | private float m_CF = 0.25f; |
---|
127 | |
---|
128 | /** Minimum number of instances */ |
---|
129 | private int m_minNumObj = 2; |
---|
130 | |
---|
131 | /** Determines whether probabilities are smoothed using |
---|
132 | Laplace correction when predictions are generated */ |
---|
133 | private boolean m_useLaplace = false; |
---|
134 | |
---|
135 | /** Number of folds for reduced error pruning. */ |
---|
136 | private int m_numFolds = 3; |
---|
137 | |
---|
138 | /** Binary splits on nominal attributes? */ |
---|
139 | private boolean m_binarySplits = false; |
---|
140 | |
---|
141 | /** Subtree raising to be performed? */ |
---|
142 | private boolean m_subtreeRaising = true; |
---|
143 | |
---|
144 | /** Cleanup after the tree has been built. */ |
---|
145 | private boolean m_noCleanup = false; |
---|
146 | |
---|
147 | /** relabel instances when grafting */ |
---|
148 | private boolean m_relabel = false; |
---|
149 | |
---|
150 | /** |
---|
151 | * Returns a string describing classifier |
---|
152 | * @return a description suitable for |
---|
153 | * displaying in the explorer/experimenter gui |
---|
154 | */ |
---|
155 | public String globalInfo() { |
---|
156 | return "Class for generating a grafted (pruned or unpruned) C4.5 " |
---|
157 | + "decision tree. For more information, see\n\n" |
---|
158 | + getTechnicalInformation().toString(); |
---|
159 | } |
---|
160 | |
---|
161 | /** |
---|
162 | * Returns an instance of a TechnicalInformation object, containing |
---|
163 | * detailed information about the technical background of this class, |
---|
164 | * e.g., paper reference or book this class is based on. |
---|
165 | * |
---|
166 | * @return the technical information about this class |
---|
167 | */ |
---|
168 | public TechnicalInformation getTechnicalInformation() { |
---|
169 | TechnicalInformation result; |
---|
170 | |
---|
171 | result = new TechnicalInformation(Type.INPROCEEDINGS); |
---|
172 | result.setValue(Field.AUTHOR, "Geoff Webb"); |
---|
173 | result.setValue(Field.YEAR, "1999"); |
---|
174 | result.setValue(Field.TITLE, "Decision Tree Grafting From the All-Tests-But-One Partition"); |
---|
175 | result.setValue(Field.PUBLISHER, "Morgan Kaufmann"); |
---|
176 | result.setValue(Field.ADDRESS, "San Francisco, CA"); |
---|
177 | |
---|
178 | return result; |
---|
179 | } |
---|
180 | |
---|
181 | /** |
---|
182 | * Returns default capabilities of the classifier. |
---|
183 | * |
---|
184 | * @return the capabilities of this classifier |
---|
185 | */ |
---|
186 | public Capabilities getCapabilities() { |
---|
187 | Capabilities result; |
---|
188 | |
---|
189 | try { |
---|
190 | result = new C45PruneableClassifierTreeG(null, !m_unpruned, m_CF, m_subtreeRaising, m_relabel, !m_noCleanup).getCapabilities(); |
---|
191 | } |
---|
192 | catch (Exception e) { |
---|
193 | result = new Capabilities(this); |
---|
194 | result.disableAll(); |
---|
195 | } |
---|
196 | |
---|
197 | result.setOwner(this); |
---|
198 | |
---|
199 | return result; |
---|
200 | } |
---|
201 | |
---|
202 | /** |
---|
203 | * Generates the classifier. |
---|
204 | * |
---|
205 | * @param instances the data to train the classifier with |
---|
206 | * @throws Exception if classifier can't be built successfully |
---|
207 | */ |
---|
208 | public void buildClassifier(Instances instances) |
---|
209 | throws Exception { |
---|
210 | |
---|
211 | ModelSelection modSelection; |
---|
212 | |
---|
213 | if (m_binarySplits) |
---|
214 | modSelection = new BinC45ModelSelection(m_minNumObj, instances, true); |
---|
215 | else |
---|
216 | modSelection = new C45ModelSelection(m_minNumObj, instances, true); |
---|
217 | m_root = new C45PruneableClassifierTreeG(modSelection, |
---|
218 | !m_unpruned, m_CF, m_subtreeRaising, |
---|
219 | m_relabel, !m_noCleanup); |
---|
220 | m_root.buildClassifier(instances); |
---|
221 | |
---|
222 | if (m_binarySplits) { |
---|
223 | ((BinC45ModelSelection)modSelection).cleanup(); |
---|
224 | } else { |
---|
225 | ((C45ModelSelection)modSelection).cleanup(); |
---|
226 | } |
---|
227 | } |
---|
228 | |
---|
229 | /** |
---|
230 | * Classifies an instance. |
---|
231 | * |
---|
232 | * @param instance the instance to classify |
---|
233 | * @return the classification for the instance |
---|
234 | * @throws Exception if instance can't be classified successfully |
---|
235 | */ |
---|
236 | public double classifyInstance(Instance instance) throws Exception { |
---|
237 | |
---|
238 | return m_root.classifyInstance(instance); |
---|
239 | } |
---|
240 | |
---|
241 | /** |
---|
242 | * Returns class probabilities for an instance. |
---|
243 | * |
---|
244 | * @param instance the instance to calculate the class probabilities for |
---|
245 | * @return the class probabilities |
---|
246 | * @throws Exception if distribution can't be computed successfully |
---|
247 | */ |
---|
248 | public final double [] distributionForInstance(Instance instance) |
---|
249 | throws Exception { |
---|
250 | |
---|
251 | return m_root.distributionForInstance(instance, m_useLaplace); |
---|
252 | } |
---|
253 | |
---|
254 | /** |
---|
255 | * Returns the type of graph this classifier |
---|
256 | * represents. |
---|
257 | * @return Drawable.TREE |
---|
258 | */ |
---|
259 | public int graphType() { |
---|
260 | return Drawable.TREE; |
---|
261 | } |
---|
262 | |
---|
263 | /** |
---|
264 | * Returns graph describing the tree. |
---|
265 | * |
---|
266 | * @return the graph describing the tree |
---|
267 | * @throws Exception if graph can't be computed |
---|
268 | */ |
---|
269 | public String graph() throws Exception { |
---|
270 | |
---|
271 | return m_root.graph(); |
---|
272 | } |
---|
273 | |
---|
274 | /** |
---|
275 | * Returns tree in prefix order. |
---|
276 | * |
---|
277 | * @return the tree in prefix order |
---|
278 | * @throws Exception if something goes wrong |
---|
279 | */ |
---|
280 | public String prefix() throws Exception { |
---|
281 | |
---|
282 | return m_root.prefix(); |
---|
283 | } |
---|
284 | |
---|
285 | |
---|
286 | /** |
---|
287 | * Returns tree as an if-then statement. |
---|
288 | * |
---|
289 | * @param className the name of the Java class |
---|
290 | * @return the tree as a Java if-then type statement |
---|
291 | * @throws Exception if something goes wrong |
---|
292 | */ |
---|
293 | public String toSource(String className) throws Exception { |
---|
294 | |
---|
295 | StringBuffer [] source = m_root.toSource(className); |
---|
296 | return |
---|
297 | "class " + className + " {\n\n" |
---|
298 | +" public static double classify(Object [] i)\n" |
---|
299 | +" throws Exception {\n\n" |
---|
300 | +" double p = Double.NaN;\n" |
---|
301 | + source[0] // Assignment code |
---|
302 | +" return p;\n" |
---|
303 | +" }\n" |
---|
304 | + source[1] // Support code |
---|
305 | +"}\n"; |
---|
306 | } |
---|
307 | |
---|
308 | /** |
---|
309 | * Returns an enumeration describing the available options. |
---|
310 | * |
---|
311 | * Valid options are: <p> |
---|
312 | * |
---|
313 | * -U <br> |
---|
314 | * Use unpruned tree.<p> |
---|
315 | * |
---|
316 | * -C confidence <br> |
---|
317 | * Set confidence threshold for pruning. (Default: 0.25) <p> |
---|
318 | * |
---|
319 | * -M number <br> |
---|
320 | * Set minimum number of instances per leaf. (Default: 2) <p> |
---|
321 | * |
---|
322 | * -B <br> |
---|
323 | * Use binary splits for nominal attributes. <p> |
---|
324 | * |
---|
325 | * -S <br> |
---|
326 | * Don't perform subtree raising. <p> |
---|
327 | * |
---|
328 | * -L <br> |
---|
329 | * Do not clean up after the tree has been built. |
---|
330 | * |
---|
331 | * -A <br> |
---|
332 | * If set, Laplace smoothing is used for predicted probabilites. |
---|
333 | * (note: this option only affects initial tree; grafting process always uses laplace). <p> |
---|
334 | * |
---|
335 | * -E <br> |
---|
336 | * Allow relabelling when grafting. <p> |
---|
337 | * |
---|
338 | * @return an enumeration of all the available options. |
---|
339 | */ |
---|
340 | public Enumeration listOptions() { |
---|
341 | |
---|
342 | Vector newVector = new Vector(9); |
---|
343 | |
---|
344 | newVector. |
---|
345 | addElement(new Option("\tUse unpruned tree.", |
---|
346 | "U", 0, "-U")); |
---|
347 | newVector. |
---|
348 | addElement(new Option("\tSet confidence threshold for pruning.\n" + |
---|
349 | "\t(default 0.25)", |
---|
350 | "C", 1, "-C <pruning confidence>")); |
---|
351 | newVector. |
---|
352 | addElement(new Option("\tSet minimum number of instances per leaf.\n" + |
---|
353 | "\t(default 2)", |
---|
354 | "M", 1, "-M <minimum number of instances>")); |
---|
355 | newVector. |
---|
356 | addElement(new Option("\tUse binary splits only.", |
---|
357 | "B", 0, "-B")); |
---|
358 | newVector. |
---|
359 | addElement(new Option("\tDon't perform subtree raising.", |
---|
360 | "S", 0, "-S")); |
---|
361 | newVector. |
---|
362 | addElement(new Option("\tDo not clean up after the tree has been built.", |
---|
363 | "L", 0, "-L")); |
---|
364 | newVector. |
---|
365 | addElement(new Option("\tLaplace smoothing for predicted probabilities. (note: this option only affects initial tree; grafting process always uses laplace).", |
---|
366 | |
---|
367 | "A", 0, "-A")); |
---|
368 | newVector. |
---|
369 | addElement(new Option("\tRelabel when grafting.", |
---|
370 | "E", 0, "-E")); |
---|
371 | return newVector.elements(); |
---|
372 | } |
---|
373 | |
---|
374 | /** |
---|
375 | * Parses a given list of options. |
---|
376 | * |
---|
377 | <!-- options-start --> |
---|
378 | * Valid options are: <p/> |
---|
379 | * |
---|
380 | * <pre> -U |
---|
381 | * Use unpruned tree.</pre> |
---|
382 | * |
---|
383 | * <pre> -C <pruning confidence> |
---|
384 | * Set confidence threshold for pruning. |
---|
385 | * (default 0.25)</pre> |
---|
386 | * |
---|
387 | * <pre> -M <minimum number of instances> |
---|
388 | * Set minimum number of instances per leaf. |
---|
389 | * (default 2)</pre> |
---|
390 | * |
---|
391 | * <pre> -B |
---|
392 | * Use binary splits only.</pre> |
---|
393 | * |
---|
394 | * <pre> -S |
---|
395 | * Don't perform subtree raising.</pre> |
---|
396 | * |
---|
397 | * <pre> -L |
---|
398 | * Do not clean up after the tree has been built.</pre> |
---|
399 | * |
---|
400 | * <pre> -A |
---|
401 | * Laplace smoothing for predicted probabilities. (note: this option only affects initial tree; grafting process always uses laplace).</pre> |
---|
402 | * |
---|
403 | * <pre> -E |
---|
404 | * Relabel when grafting.</pre> |
---|
405 | * |
---|
406 | <!-- options-end --> |
---|
407 | * |
---|
408 | * @param options the list of options as an array of strings |
---|
409 | * @throws Exception if an option is not supported |
---|
410 | */ |
---|
411 | public void setOptions(String[] options) throws Exception { |
---|
412 | |
---|
413 | // Other options |
---|
414 | String minNumString = Utils.getOption('M', options); |
---|
415 | if (minNumString.length() != 0) { |
---|
416 | m_minNumObj = Integer.parseInt(minNumString); |
---|
417 | } else { |
---|
418 | m_minNumObj = 2; |
---|
419 | } |
---|
420 | m_binarySplits = Utils.getFlag('B', options); |
---|
421 | m_useLaplace = Utils.getFlag('A', options); |
---|
422 | |
---|
423 | // Pruning options |
---|
424 | m_unpruned = Utils.getFlag('U', options); |
---|
425 | m_subtreeRaising = !Utils.getFlag('S', options); |
---|
426 | m_noCleanup = Utils.getFlag('L', options); |
---|
427 | if ((m_unpruned) && (!m_subtreeRaising)) { |
---|
428 | throw new Exception("Subtree raising doesn't need to be unset for unpruned tree!"); |
---|
429 | } |
---|
430 | m_relabel = Utils.getFlag('E', options); |
---|
431 | String confidenceString = Utils.getOption('C', options); |
---|
432 | if (confidenceString.length() != 0) { |
---|
433 | if (m_unpruned) { |
---|
434 | throw new Exception("Doesn't make sense to change confidence for unpruned " |
---|
435 | +"tree!"); |
---|
436 | } else { |
---|
437 | m_CF = (new Float(confidenceString)).floatValue(); |
---|
438 | if ((m_CF <= 0) || (m_CF >= 1)) { |
---|
439 | throw new Exception("Confidence has to be greater than zero and smaller " + |
---|
440 | "than one!"); |
---|
441 | } |
---|
442 | } |
---|
443 | } else { |
---|
444 | m_CF = 0.25f; |
---|
445 | } |
---|
446 | } |
---|
447 | |
---|
448 | /** |
---|
449 | * Gets the current settings of the Classifier. |
---|
450 | * |
---|
451 | * @return an array of strings suitable for passing to setOptions |
---|
452 | */ |
---|
453 | public String [] getOptions() { |
---|
454 | |
---|
455 | String [] options = new String [10]; |
---|
456 | int current = 0; |
---|
457 | |
---|
458 | if (m_noCleanup) { |
---|
459 | options[current++] = "-L"; |
---|
460 | } |
---|
461 | if (m_unpruned) { |
---|
462 | options[current++] = "-U"; |
---|
463 | } else { |
---|
464 | if (!m_subtreeRaising) { |
---|
465 | options[current++] = "-S"; |
---|
466 | } |
---|
467 | options[current++] = "-C"; options[current++] = "" + m_CF; |
---|
468 | } |
---|
469 | if (m_binarySplits) { |
---|
470 | options[current++] = "-B"; |
---|
471 | } |
---|
472 | options[current++] = "-M"; options[current++] = "" + m_minNumObj; |
---|
473 | if (m_useLaplace) { |
---|
474 | options[current++] = "-A"; |
---|
475 | } |
---|
476 | |
---|
477 | if(m_relabel) { |
---|
478 | options[current++] = "-E"; |
---|
479 | } |
---|
480 | |
---|
481 | while (current < options.length) { |
---|
482 | options[current++] = ""; |
---|
483 | } |
---|
484 | return options; |
---|
485 | } |
---|
486 | |
---|
487 | /** |
---|
488 | * Returns the tip text for this property |
---|
489 | * @return tip text for this property suitable for |
---|
490 | * displaying in the explorer/experimenter gui |
---|
491 | */ |
---|
492 | public String useLaplaceTipText() { |
---|
493 | return "Whether counts at leaves are smoothed based on Laplace."; |
---|
494 | } |
---|
495 | |
---|
496 | /** |
---|
497 | * Get the value of useLaplace. |
---|
498 | * |
---|
499 | * @return Value of useLaplace. |
---|
500 | */ |
---|
501 | public boolean getUseLaplace() { |
---|
502 | |
---|
503 | return m_useLaplace; |
---|
504 | } |
---|
505 | |
---|
506 | /** |
---|
507 | * Set the value of useLaplace. |
---|
508 | * |
---|
509 | * @param newuseLaplace Value to assign to useLaplace. |
---|
510 | */ |
---|
511 | public void setUseLaplace(boolean newuseLaplace) { |
---|
512 | |
---|
513 | m_useLaplace = newuseLaplace; |
---|
514 | } |
---|
515 | |
---|
516 | /** |
---|
517 | * Returns a description of the classifier. |
---|
518 | * |
---|
519 | * @return a description of the classifier |
---|
520 | */ |
---|
521 | public String toString() { |
---|
522 | |
---|
523 | if (m_root == null) { |
---|
524 | return "No classifier built"; |
---|
525 | } |
---|
526 | if (m_unpruned) |
---|
527 | return "J48graft unpruned tree\n------------------\n" + m_root.toString(); |
---|
528 | else |
---|
529 | return "J48graft pruned tree\n------------------\n" + m_root.toString(); |
---|
530 | } |
---|
531 | |
---|
532 | /** |
---|
533 | * Returns a superconcise version of the model |
---|
534 | * |
---|
535 | * @return a summary of the model |
---|
536 | */ |
---|
537 | public String toSummaryString() { |
---|
538 | |
---|
539 | return "Number of leaves: " + m_root.numLeaves() + "\n" |
---|
540 | + "Size of the tree: " + m_root.numNodes() + "\n"; |
---|
541 | } |
---|
542 | |
---|
543 | /** |
---|
544 | * Returns the size of the tree |
---|
545 | * @return the size of the tree |
---|
546 | */ |
---|
547 | public double measureTreeSize() { |
---|
548 | return m_root.numNodes(); |
---|
549 | } |
---|
550 | |
---|
551 | /** |
---|
552 | * Returns the number of leaves |
---|
553 | * @return the number of leaves |
---|
554 | */ |
---|
555 | public double measureNumLeaves() { |
---|
556 | return m_root.numLeaves(); |
---|
557 | } |
---|
558 | |
---|
559 | /** |
---|
560 | * Returns the number of rules (same as number of leaves) |
---|
561 | * @return the number of rules |
---|
562 | */ |
---|
563 | public double measureNumRules() { |
---|
564 | return m_root.numLeaves(); |
---|
565 | } |
---|
566 | |
---|
567 | /** |
---|
568 | * Returns an enumeration of the additional measure names |
---|
569 | * @return an enumeration of the measure names |
---|
570 | */ |
---|
571 | public Enumeration enumerateMeasures() { |
---|
572 | Vector newVector = new Vector(3); |
---|
573 | newVector.addElement("measureTreeSize"); |
---|
574 | newVector.addElement("measureNumLeaves"); |
---|
575 | newVector.addElement("measureNumRules"); |
---|
576 | return newVector.elements(); |
---|
577 | } |
---|
578 | |
---|
579 | /** |
---|
580 | * Returns the value of the named measure |
---|
581 | * @param additionalMeasureName the name of the measure to query for its value |
---|
582 | * @return the value of the named measure |
---|
583 | * @throws IllegalArgumentException if the named measure is not supported |
---|
584 | */ |
---|
585 | public double getMeasure(String additionalMeasureName) { |
---|
586 | if (additionalMeasureName.compareTo("measureNumRules") == 0) { |
---|
587 | return measureNumRules(); |
---|
588 | } else if (additionalMeasureName.compareTo("measureTreeSize") == 0) { |
---|
589 | return measureTreeSize(); |
---|
590 | } else if (additionalMeasureName.compareTo("measureNumLeaves") == 0) { |
---|
591 | return measureNumLeaves(); |
---|
592 | } else { |
---|
593 | throw new IllegalArgumentException(additionalMeasureName |
---|
594 | + " not supported (j48)"); |
---|
595 | } |
---|
596 | } |
---|
597 | |
---|
598 | /** |
---|
599 | * Returns the tip text for this property |
---|
600 | * @return tip text for this property suitable for |
---|
601 | * displaying in the explorer/experimenter gui |
---|
602 | */ |
---|
603 | public String unprunedTipText() { |
---|
604 | return "Whether pruning is performed."; |
---|
605 | } |
---|
606 | |
---|
607 | /** |
---|
608 | * Get the value of unpruned. |
---|
609 | * |
---|
610 | * @return Value of unpruned. |
---|
611 | */ |
---|
612 | public boolean getUnpruned() { |
---|
613 | |
---|
614 | return m_unpruned; |
---|
615 | } |
---|
616 | |
---|
617 | /** |
---|
618 | * Set the value of unpruned. |
---|
619 | * @param v Value to assign to unpruned. |
---|
620 | */ |
---|
621 | public void setUnpruned(boolean v) { |
---|
622 | |
---|
623 | m_unpruned = v; |
---|
624 | } |
---|
625 | |
---|
626 | /** |
---|
627 | * Returns the tip text for this property |
---|
628 | * @return tip text for this property suitable for |
---|
629 | * displaying in the explorer/experimenter gui |
---|
630 | */ |
---|
631 | public String relabelTipText() { |
---|
632 | return "Whether relabelling is allowed during grafting."; |
---|
633 | } |
---|
634 | |
---|
635 | /** |
---|
636 | * Get the value of relabelling |
---|
637 | * |
---|
638 | * @return Value of relabelling. |
---|
639 | */ |
---|
640 | public boolean getRelabel() { |
---|
641 | |
---|
642 | return m_relabel; |
---|
643 | } |
---|
644 | |
---|
645 | /** |
---|
646 | * Set the value of relabelling. |
---|
647 | * |
---|
648 | * @param v Value to assign to relabelling flag. |
---|
649 | */ |
---|
650 | public void setRelabel(boolean v) { |
---|
651 | m_relabel = v; |
---|
652 | } |
---|
653 | |
---|
654 | /** |
---|
655 | * Returns the tip text for this property |
---|
656 | * @return tip text for this property suitable for |
---|
657 | * displaying in the explorer/experimenter gui |
---|
658 | */ |
---|
659 | public String confidenceFactorTipText() { |
---|
660 | return "The confidence factor used for pruning (smaller values incur " |
---|
661 | + "more pruning)."; |
---|
662 | } |
---|
663 | |
---|
664 | /** |
---|
665 | * Get the value of CF. |
---|
666 | * |
---|
667 | * @return Value of CF. |
---|
668 | */ |
---|
669 | public float getConfidenceFactor() { |
---|
670 | |
---|
671 | return m_CF; |
---|
672 | } |
---|
673 | |
---|
674 | /** |
---|
675 | * Set the value of CF. |
---|
676 | * |
---|
677 | * @param v Value to assign to CF. |
---|
678 | */ |
---|
679 | public void setConfidenceFactor(float v) { |
---|
680 | |
---|
681 | m_CF = v; |
---|
682 | } |
---|
683 | |
---|
684 | /** |
---|
685 | * Returns the tip text for this property |
---|
686 | * @return tip text for this property suitable for |
---|
687 | * displaying in the explorer/experimenter gui |
---|
688 | */ |
---|
689 | public String minNumObjTipText() { |
---|
690 | return "The minimum number of instances per leaf."; |
---|
691 | } |
---|
692 | |
---|
693 | /** |
---|
694 | * Get the value of minNumObj. |
---|
695 | * |
---|
696 | * @return Value of minNumObj. |
---|
697 | */ |
---|
698 | public int getMinNumObj() { |
---|
699 | |
---|
700 | return m_minNumObj; |
---|
701 | } |
---|
702 | |
---|
703 | /** |
---|
704 | * Set the value of minNumObj. |
---|
705 | * |
---|
706 | * @param v Value to assign to minNumObj. |
---|
707 | */ |
---|
708 | public void setMinNumObj(int v) { |
---|
709 | |
---|
710 | m_minNumObj = v; |
---|
711 | } |
---|
712 | |
---|
713 | /** |
---|
714 | * Returns the tip text for this property |
---|
715 | * @return tip text for this property suitable for |
---|
716 | * displaying in the explorer/experimenter gui |
---|
717 | */ |
---|
718 | public String binarySplitsTipText() { |
---|
719 | return "Whether to use binary splits on nominal attributes when " |
---|
720 | + "building the trees."; |
---|
721 | } |
---|
722 | |
---|
723 | /** |
---|
724 | * Get the value of binarySplits. |
---|
725 | * |
---|
726 | * @return Value of binarySplits. |
---|
727 | */ |
---|
728 | public boolean getBinarySplits() { |
---|
729 | |
---|
730 | return m_binarySplits; |
---|
731 | } |
---|
732 | |
---|
733 | /** |
---|
734 | * Set the value of binarySplits. |
---|
735 | * |
---|
736 | * @param v Value to assign to binarySplits. |
---|
737 | */ |
---|
738 | public void setBinarySplits(boolean v) { |
---|
739 | |
---|
740 | m_binarySplits = v; |
---|
741 | } |
---|
742 | |
---|
743 | /** |
---|
744 | * Returns the tip text for this property |
---|
745 | * @return tip text for this property suitable for |
---|
746 | * displaying in the explorer/experimenter gui |
---|
747 | */ |
---|
748 | public String subtreeRaisingTipText() { |
---|
749 | return "Whether to consider the subtree raising operation when pruning."; |
---|
750 | } |
---|
751 | |
---|
752 | /** |
---|
753 | * Get the value of subtreeRaising. |
---|
754 | * |
---|
755 | * @return Value of subtreeRaising. |
---|
756 | */ |
---|
757 | public boolean getSubtreeRaising() { |
---|
758 | |
---|
759 | return m_subtreeRaising; |
---|
760 | } |
---|
761 | |
---|
762 | /** |
---|
763 | * Set the value of subtreeRaising. |
---|
764 | * |
---|
765 | * @param v Value to assign to subtreeRaising. |
---|
766 | */ |
---|
767 | public void setSubtreeRaising(boolean v) { |
---|
768 | |
---|
769 | m_subtreeRaising = v; |
---|
770 | } |
---|
771 | |
---|
772 | /** |
---|
773 | * Returns the tip text for this property |
---|
774 | * @return tip text for this property suitable for |
---|
775 | * displaying in the explorer/experimenter gui |
---|
776 | */ |
---|
777 | public String saveInstanceDataTipText() { |
---|
778 | return "Whether to save the training data for visualization."; |
---|
779 | } |
---|
780 | |
---|
781 | /** |
---|
782 | * Check whether instance data is to be saved. |
---|
783 | * |
---|
784 | * @return true if instance data is saved |
---|
785 | */ |
---|
786 | public boolean getSaveInstanceData() { |
---|
787 | |
---|
788 | return m_noCleanup; |
---|
789 | } |
---|
790 | |
---|
791 | /** |
---|
792 | * Set whether instance data is to be saved. |
---|
793 | * @param v true if instance data is to be saved |
---|
794 | */ |
---|
795 | public void setSaveInstanceData(boolean v) { |
---|
796 | |
---|
797 | m_noCleanup = v; |
---|
798 | } |
---|
799 | |
---|
800 | /** |
---|
801 | * Returns the revision string. |
---|
802 | * |
---|
803 | * @return the revision |
---|
804 | */ |
---|
805 | public String getRevision() { |
---|
806 | return RevisionUtils.extract("$Revision: 6088 $"); |
---|
807 | } |
---|
808 | |
---|
809 | /** |
---|
810 | * Main method for testing this class |
---|
811 | * |
---|
812 | * @param argv the commandline options |
---|
813 | */ |
---|
814 | public static void main(String [] argv){ |
---|
815 | runClassifier(new J48graft(), argv); |
---|
816 | } |
---|
817 | } |
---|
818 | |
---|