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 | * MIWrapper.java |
---|
19 | * Copyright (C) 2005 University of Waikato, Hamilton, New Zealand |
---|
20 | * |
---|
21 | */ |
---|
22 | |
---|
23 | package weka.classifiers.mi; |
---|
24 | |
---|
25 | import weka.classifiers.SingleClassifierEnhancer; |
---|
26 | import weka.core.Capabilities; |
---|
27 | import weka.core.Instance; |
---|
28 | import weka.core.Instances; |
---|
29 | import weka.core.MultiInstanceCapabilitiesHandler; |
---|
30 | import weka.core.Option; |
---|
31 | import weka.core.OptionHandler; |
---|
32 | import weka.core.RevisionUtils; |
---|
33 | import weka.core.SelectedTag; |
---|
34 | import weka.core.Tag; |
---|
35 | import weka.core.TechnicalInformation; |
---|
36 | import weka.core.TechnicalInformationHandler; |
---|
37 | import weka.core.Utils; |
---|
38 | import weka.core.Capabilities.Capability; |
---|
39 | import weka.core.TechnicalInformation.Field; |
---|
40 | import weka.core.TechnicalInformation.Type; |
---|
41 | import weka.filters.Filter; |
---|
42 | import weka.filters.unsupervised.attribute.MultiInstanceToPropositional; |
---|
43 | |
---|
44 | import java.util.Enumeration; |
---|
45 | import java.util.Vector; |
---|
46 | |
---|
47 | /** |
---|
48 | <!-- globalinfo-start --> |
---|
49 | * A simple Wrapper method for applying standard propositional learners to multi-instance data.<br/> |
---|
50 | * <br/> |
---|
51 | * For more information see:<br/> |
---|
52 | * <br/> |
---|
53 | * E. T. Frank, X. Xu (2003). Applying propositional learning algorithms to multi-instance data. Department of Computer Science, University of Waikato, Hamilton, NZ. |
---|
54 | * <p/> |
---|
55 | <!-- globalinfo-end --> |
---|
56 | * |
---|
57 | <!-- technical-bibtex-start --> |
---|
58 | * BibTeX: |
---|
59 | * <pre> |
---|
60 | * @techreport{Frank2003, |
---|
61 | * address = {Department of Computer Science, University of Waikato, Hamilton, NZ}, |
---|
62 | * author = {E. T. Frank and X. Xu}, |
---|
63 | * institution = {University of Waikato}, |
---|
64 | * month = {06}, |
---|
65 | * title = {Applying propositional learning algorithms to multi-instance data}, |
---|
66 | * year = {2003} |
---|
67 | * } |
---|
68 | * </pre> |
---|
69 | * <p/> |
---|
70 | <!-- technical-bibtex-end --> |
---|
71 | * |
---|
72 | <!-- options-start --> |
---|
73 | * Valid options are: <p/> |
---|
74 | * |
---|
75 | * <pre> -P [1|2|3] |
---|
76 | * The method used in testing: |
---|
77 | * 1.arithmetic average |
---|
78 | * 2.geometric average |
---|
79 | * 3.max probability of positive bag. |
---|
80 | * (default: 1)</pre> |
---|
81 | * |
---|
82 | * <pre> -A [0|1|2|3] |
---|
83 | * The type of weight setting for each single-instance: |
---|
84 | * 0.keep the weight to be the same as the original value; |
---|
85 | * 1.weight = 1.0 |
---|
86 | * 2.weight = 1.0/Total number of single-instance in the |
---|
87 | * corresponding bag |
---|
88 | * 3. weight = Total number of single-instance / (Total |
---|
89 | * number of bags * Total number of single-instance |
---|
90 | * in the corresponding bag). |
---|
91 | * (default: 3)</pre> |
---|
92 | * |
---|
93 | * <pre> -D |
---|
94 | * If set, classifier is run in debug mode and |
---|
95 | * may output additional info to the console</pre> |
---|
96 | * |
---|
97 | * <pre> -W |
---|
98 | * Full name of base classifier. |
---|
99 | * (default: weka.classifiers.rules.ZeroR)</pre> |
---|
100 | * |
---|
101 | * <pre> |
---|
102 | * Options specific to classifier weka.classifiers.rules.ZeroR: |
---|
103 | * </pre> |
---|
104 | * |
---|
105 | * <pre> -D |
---|
106 | * If set, classifier is run in debug mode and |
---|
107 | * may output additional info to the console</pre> |
---|
108 | * |
---|
109 | <!-- options-end --> |
---|
110 | * |
---|
111 | * @author Eibe Frank (eibe@cs.waikato.ac.nz) |
---|
112 | * @author Xin Xu (xx5@cs.waikato.ac.nz) |
---|
113 | * @version $Revision: 1.5 $ |
---|
114 | */ |
---|
115 | public class MIWrapper |
---|
116 | extends SingleClassifierEnhancer |
---|
117 | implements MultiInstanceCapabilitiesHandler, OptionHandler, |
---|
118 | TechnicalInformationHandler { |
---|
119 | |
---|
120 | /** for serialization */ |
---|
121 | static final long serialVersionUID = -7707766152904315910L; |
---|
122 | |
---|
123 | /** The number of the class labels */ |
---|
124 | protected int m_NumClasses; |
---|
125 | |
---|
126 | /** arithmetic average */ |
---|
127 | public static final int TESTMETHOD_ARITHMETIC = 1; |
---|
128 | /** geometric average */ |
---|
129 | public static final int TESTMETHOD_GEOMETRIC = 2; |
---|
130 | /** max probability of positive bag */ |
---|
131 | public static final int TESTMETHOD_MAXPROB = 3; |
---|
132 | /** the test methods */ |
---|
133 | public static final Tag[] TAGS_TESTMETHOD = { |
---|
134 | new Tag(TESTMETHOD_ARITHMETIC, "arithmetic average"), |
---|
135 | new Tag(TESTMETHOD_GEOMETRIC, "geometric average"), |
---|
136 | new Tag(TESTMETHOD_MAXPROB, "max probability of positive bag") |
---|
137 | }; |
---|
138 | |
---|
139 | /** the test method */ |
---|
140 | protected int m_Method = TESTMETHOD_GEOMETRIC; |
---|
141 | |
---|
142 | /** Filter used to convert MI dataset into single-instance dataset */ |
---|
143 | protected MultiInstanceToPropositional m_ConvertToProp = new MultiInstanceToPropositional(); |
---|
144 | |
---|
145 | /** the single-instance weight setting method */ |
---|
146 | protected int m_WeightMethod = MultiInstanceToPropositional.WEIGHTMETHOD_INVERSE2; |
---|
147 | |
---|
148 | /** |
---|
149 | * Returns a string describing this filter |
---|
150 | * |
---|
151 | * @return a description of the filter suitable for |
---|
152 | * displaying in the explorer/experimenter gui |
---|
153 | */ |
---|
154 | public String globalInfo() { |
---|
155 | return |
---|
156 | "A simple Wrapper method for applying standard propositional learners " |
---|
157 | + "to multi-instance data.\n\n" |
---|
158 | + "For more information see:\n\n" |
---|
159 | + getTechnicalInformation().toString(); |
---|
160 | } |
---|
161 | |
---|
162 | /** |
---|
163 | * Returns an instance of a TechnicalInformation object, containing |
---|
164 | * detailed information about the technical background of this class, |
---|
165 | * e.g., paper reference or book this class is based on. |
---|
166 | * |
---|
167 | * @return the technical information about this class |
---|
168 | */ |
---|
169 | public TechnicalInformation getTechnicalInformation() { |
---|
170 | TechnicalInformation result; |
---|
171 | |
---|
172 | result = new TechnicalInformation(Type.TECHREPORT); |
---|
173 | result.setValue(Field.AUTHOR, "E. T. Frank and X. Xu"); |
---|
174 | result.setValue(Field.TITLE, "Applying propositional learning algorithms to multi-instance data"); |
---|
175 | result.setValue(Field.YEAR, "2003"); |
---|
176 | result.setValue(Field.MONTH, "06"); |
---|
177 | result.setValue(Field.INSTITUTION, "University of Waikato"); |
---|
178 | result.setValue(Field.ADDRESS, "Department of Computer Science, University of Waikato, Hamilton, NZ"); |
---|
179 | |
---|
180 | return result; |
---|
181 | } |
---|
182 | |
---|
183 | /** |
---|
184 | * Returns an enumeration describing the available options. |
---|
185 | * |
---|
186 | * @return an enumeration of all the available options. |
---|
187 | */ |
---|
188 | public Enumeration listOptions() { |
---|
189 | Vector result = new Vector(); |
---|
190 | |
---|
191 | result.addElement(new Option( |
---|
192 | "\tThe method used in testing:\n" |
---|
193 | + "\t1.arithmetic average\n" |
---|
194 | + "\t2.geometric average\n" |
---|
195 | + "\t3.max probability of positive bag.\n" |
---|
196 | + "\t(default: 1)", |
---|
197 | "P", 1, "-P [1|2|3]")); |
---|
198 | |
---|
199 | result.addElement(new Option( |
---|
200 | "\tThe type of weight setting for each single-instance:\n" |
---|
201 | + "\t0.keep the weight to be the same as the original value;\n" |
---|
202 | + "\t1.weight = 1.0\n" |
---|
203 | + "\t2.weight = 1.0/Total number of single-instance in the\n" |
---|
204 | + "\t\tcorresponding bag\n" |
---|
205 | + "\t3. weight = Total number of single-instance / (Total\n" |
---|
206 | + "\t\tnumber of bags * Total number of single-instance \n" |
---|
207 | + "\t\tin the corresponding bag).\n" |
---|
208 | + "\t(default: 3)", |
---|
209 | "A", 1, "-A [0|1|2|3]")); |
---|
210 | |
---|
211 | Enumeration enu = super.listOptions(); |
---|
212 | while (enu.hasMoreElements()) { |
---|
213 | result.addElement(enu.nextElement()); |
---|
214 | } |
---|
215 | |
---|
216 | return result.elements(); |
---|
217 | } |
---|
218 | |
---|
219 | |
---|
220 | /** |
---|
221 | * Parses a given list of options. <p/> |
---|
222 | * |
---|
223 | <!-- options-start --> |
---|
224 | * Valid options are: <p/> |
---|
225 | * |
---|
226 | * <pre> -P [1|2|3] |
---|
227 | * The method used in testing: |
---|
228 | * 1.arithmetic average |
---|
229 | * 2.geometric average |
---|
230 | * 3.max probability of positive bag. |
---|
231 | * (default: 1)</pre> |
---|
232 | * |
---|
233 | * <pre> -A [0|1|2|3] |
---|
234 | * The type of weight setting for each single-instance: |
---|
235 | * 0.keep the weight to be the same as the original value; |
---|
236 | * 1.weight = 1.0 |
---|
237 | * 2.weight = 1.0/Total number of single-instance in the |
---|
238 | * corresponding bag |
---|
239 | * 3. weight = Total number of single-instance / (Total |
---|
240 | * number of bags * Total number of single-instance |
---|
241 | * in the corresponding bag). |
---|
242 | * (default: 3)</pre> |
---|
243 | * |
---|
244 | * <pre> -D |
---|
245 | * If set, classifier is run in debug mode and |
---|
246 | * may output additional info to the console</pre> |
---|
247 | * |
---|
248 | * <pre> -W |
---|
249 | * Full name of base classifier. |
---|
250 | * (default: weka.classifiers.rules.ZeroR)</pre> |
---|
251 | * |
---|
252 | * <pre> |
---|
253 | * Options specific to classifier weka.classifiers.rules.ZeroR: |
---|
254 | * </pre> |
---|
255 | * |
---|
256 | * <pre> -D |
---|
257 | * If set, classifier is run in debug mode and |
---|
258 | * may output additional info to the console</pre> |
---|
259 | * |
---|
260 | <!-- options-end --> |
---|
261 | * |
---|
262 | * @param options the list of options as an array of strings |
---|
263 | * @throws Exception if an option is not supported |
---|
264 | */ |
---|
265 | public void setOptions(String[] options) throws Exception { |
---|
266 | |
---|
267 | setDebug(Utils.getFlag('D', options)); |
---|
268 | |
---|
269 | String methodString = Utils.getOption('P', options); |
---|
270 | if (methodString.length() != 0) { |
---|
271 | setMethod( |
---|
272 | new SelectedTag(Integer.parseInt(methodString), TAGS_TESTMETHOD)); |
---|
273 | } else { |
---|
274 | setMethod( |
---|
275 | new SelectedTag(TESTMETHOD_ARITHMETIC, TAGS_TESTMETHOD)); |
---|
276 | } |
---|
277 | |
---|
278 | String weightString = Utils.getOption('A', options); |
---|
279 | if (weightString.length() != 0) { |
---|
280 | setWeightMethod( |
---|
281 | new SelectedTag( |
---|
282 | Integer.parseInt(weightString), |
---|
283 | MultiInstanceToPropositional.TAGS_WEIGHTMETHOD)); |
---|
284 | } else { |
---|
285 | setWeightMethod( |
---|
286 | new SelectedTag( |
---|
287 | MultiInstanceToPropositional.WEIGHTMETHOD_INVERSE2, |
---|
288 | MultiInstanceToPropositional.TAGS_WEIGHTMETHOD)); |
---|
289 | } |
---|
290 | |
---|
291 | super.setOptions(options); |
---|
292 | } |
---|
293 | |
---|
294 | /** |
---|
295 | * Gets the current settings of the Classifier. |
---|
296 | * |
---|
297 | * @return an array of strings suitable for passing to setOptions |
---|
298 | */ |
---|
299 | public String[] getOptions() { |
---|
300 | Vector result; |
---|
301 | String[] options; |
---|
302 | int i; |
---|
303 | |
---|
304 | result = new Vector(); |
---|
305 | |
---|
306 | result.add("-P"); |
---|
307 | result.add("" + m_Method); |
---|
308 | |
---|
309 | result.add("-A"); |
---|
310 | result.add("" + m_WeightMethod); |
---|
311 | |
---|
312 | options = super.getOptions(); |
---|
313 | for (i = 0; i < options.length; i++) |
---|
314 | result.add(options[i]); |
---|
315 | |
---|
316 | return (String[]) result.toArray(new String[result.size()]); |
---|
317 | } |
---|
318 | |
---|
319 | /** |
---|
320 | * Returns the tip text for this property |
---|
321 | * |
---|
322 | * @return tip text for this property suitable for |
---|
323 | * displaying in the explorer/experimenter gui |
---|
324 | */ |
---|
325 | public String weightMethodTipText() { |
---|
326 | return "The method used for weighting the instances."; |
---|
327 | } |
---|
328 | |
---|
329 | /** |
---|
330 | * The new method for weighting the instances. |
---|
331 | * |
---|
332 | * @param method the new method |
---|
333 | */ |
---|
334 | public void setWeightMethod(SelectedTag method){ |
---|
335 | if (method.getTags() == MultiInstanceToPropositional.TAGS_WEIGHTMETHOD) |
---|
336 | m_WeightMethod = method.getSelectedTag().getID(); |
---|
337 | } |
---|
338 | |
---|
339 | /** |
---|
340 | * Returns the current weighting method for instances. |
---|
341 | * |
---|
342 | * @return the current weighting method |
---|
343 | */ |
---|
344 | public SelectedTag getWeightMethod(){ |
---|
345 | return new SelectedTag( |
---|
346 | m_WeightMethod, MultiInstanceToPropositional.TAGS_WEIGHTMETHOD); |
---|
347 | } |
---|
348 | |
---|
349 | /** |
---|
350 | * Returns the tip text for this property |
---|
351 | * |
---|
352 | * @return tip text for this property suitable for |
---|
353 | * displaying in the explorer/experimenter gui |
---|
354 | */ |
---|
355 | public String methodTipText() { |
---|
356 | return "The method used for testing."; |
---|
357 | } |
---|
358 | |
---|
359 | /** |
---|
360 | * Set the method used in testing. |
---|
361 | * |
---|
362 | * @param method the index of method to use. |
---|
363 | */ |
---|
364 | public void setMethod(SelectedTag method) { |
---|
365 | if (method.getTags() == TAGS_TESTMETHOD) |
---|
366 | m_Method = method.getSelectedTag().getID(); |
---|
367 | } |
---|
368 | |
---|
369 | /** |
---|
370 | * Get the method used in testing. |
---|
371 | * |
---|
372 | * @return the index of method used in testing. |
---|
373 | */ |
---|
374 | public SelectedTag getMethod() { |
---|
375 | return new SelectedTag(m_Method, TAGS_TESTMETHOD); |
---|
376 | } |
---|
377 | |
---|
378 | /** |
---|
379 | * Returns default capabilities of the classifier. |
---|
380 | * |
---|
381 | * @return the capabilities of this classifier |
---|
382 | */ |
---|
383 | public Capabilities getCapabilities() { |
---|
384 | Capabilities result = super.getCapabilities(); |
---|
385 | |
---|
386 | // class |
---|
387 | result.disableAllClasses(); |
---|
388 | result.disableAllClassDependencies(); |
---|
389 | if (super.getCapabilities().handles(Capability.NOMINAL_CLASS)) |
---|
390 | result.enable(Capability.NOMINAL_CLASS); |
---|
391 | if (super.getCapabilities().handles(Capability.BINARY_CLASS)) |
---|
392 | result.enable(Capability.BINARY_CLASS); |
---|
393 | result.enable(Capability.RELATIONAL_ATTRIBUTES); |
---|
394 | result.enable(Capability.MISSING_CLASS_VALUES); |
---|
395 | |
---|
396 | // other |
---|
397 | result.enable(Capability.ONLY_MULTIINSTANCE); |
---|
398 | |
---|
399 | return result; |
---|
400 | } |
---|
401 | |
---|
402 | /** |
---|
403 | * Returns the capabilities of this multi-instance classifier for the |
---|
404 | * relational data. |
---|
405 | * |
---|
406 | * @return the capabilities of this object |
---|
407 | * @see Capabilities |
---|
408 | */ |
---|
409 | public Capabilities getMultiInstanceCapabilities() { |
---|
410 | Capabilities result = super.getCapabilities(); |
---|
411 | |
---|
412 | // class |
---|
413 | result.disableAllClasses(); |
---|
414 | result.enable(Capability.NO_CLASS); |
---|
415 | |
---|
416 | return result; |
---|
417 | } |
---|
418 | |
---|
419 | /** |
---|
420 | * Builds the classifier |
---|
421 | * |
---|
422 | * @param data the training data to be used for generating the |
---|
423 | * boosted classifier. |
---|
424 | * @throws Exception if the classifier could not be built successfully |
---|
425 | */ |
---|
426 | public void buildClassifier(Instances data) throws Exception { |
---|
427 | |
---|
428 | // can classifier handle the data? |
---|
429 | getCapabilities().testWithFail(data); |
---|
430 | |
---|
431 | // remove instances with missing class |
---|
432 | Instances train = new Instances(data); |
---|
433 | train.deleteWithMissingClass(); |
---|
434 | |
---|
435 | if (m_Classifier == null) { |
---|
436 | throw new Exception("A base classifier has not been specified!"); |
---|
437 | } |
---|
438 | |
---|
439 | if (getDebug()) |
---|
440 | System.out.println("Start training ..."); |
---|
441 | m_NumClasses = train.numClasses(); |
---|
442 | |
---|
443 | //convert the training dataset into single-instance dataset |
---|
444 | m_ConvertToProp.setWeightMethod(getWeightMethod()); |
---|
445 | m_ConvertToProp.setInputFormat(train); |
---|
446 | train = Filter.useFilter(train, m_ConvertToProp); |
---|
447 | train.deleteAttributeAt(0); // remove the bag index attribute |
---|
448 | |
---|
449 | m_Classifier.buildClassifier(train); |
---|
450 | } |
---|
451 | |
---|
452 | /** |
---|
453 | * Computes the distribution for a given exemplar |
---|
454 | * |
---|
455 | * @param exmp the exemplar for which distribution is computed |
---|
456 | * @return the distribution |
---|
457 | * @throws Exception if the distribution can't be computed successfully |
---|
458 | */ |
---|
459 | public double[] distributionForInstance(Instance exmp) |
---|
460 | throws Exception { |
---|
461 | |
---|
462 | Instances testData = new Instances (exmp.dataset(),0); |
---|
463 | testData.add(exmp); |
---|
464 | |
---|
465 | // convert the training dataset into single-instance dataset |
---|
466 | m_ConvertToProp.setWeightMethod( |
---|
467 | new SelectedTag( |
---|
468 | MultiInstanceToPropositional.WEIGHTMETHOD_ORIGINAL, |
---|
469 | MultiInstanceToPropositional.TAGS_WEIGHTMETHOD)); |
---|
470 | testData = Filter.useFilter(testData, m_ConvertToProp); |
---|
471 | testData.deleteAttributeAt(0); //remove the bag index attribute |
---|
472 | |
---|
473 | // Compute the log-probability of the bag |
---|
474 | double [] distribution = new double[m_NumClasses]; |
---|
475 | double nI = (double)testData.numInstances(); |
---|
476 | double [] maxPr = new double [m_NumClasses]; |
---|
477 | |
---|
478 | for(int i=0; i<nI; i++){ |
---|
479 | double[] dist = m_Classifier.distributionForInstance(testData.instance(i)); |
---|
480 | for(int j=0; j<m_NumClasses; j++){ |
---|
481 | |
---|
482 | switch(m_Method){ |
---|
483 | case TESTMETHOD_ARITHMETIC: |
---|
484 | distribution[j] += dist[j]/nI; |
---|
485 | break; |
---|
486 | case TESTMETHOD_GEOMETRIC: |
---|
487 | // Avoid 0/1 probability |
---|
488 | if(dist[j]<0.001) |
---|
489 | dist[j] = 0.001; |
---|
490 | else if(dist[j]>0.999) |
---|
491 | dist[j] = 0.999; |
---|
492 | |
---|
493 | distribution[j] += Math.log(dist[j])/nI; |
---|
494 | break; |
---|
495 | case TESTMETHOD_MAXPROB: |
---|
496 | if (dist[j]>maxPr[j]) |
---|
497 | maxPr[j] = dist[j]; |
---|
498 | break; |
---|
499 | } |
---|
500 | } |
---|
501 | } |
---|
502 | |
---|
503 | if(m_Method == TESTMETHOD_GEOMETRIC) |
---|
504 | for(int j=0; j<m_NumClasses; j++) |
---|
505 | distribution[j] = Math.exp(distribution[j]); |
---|
506 | |
---|
507 | if(m_Method == TESTMETHOD_MAXPROB){ // for positive bag |
---|
508 | distribution[1] = maxPr[1]; |
---|
509 | distribution[0] = 1 - distribution[1]; |
---|
510 | } |
---|
511 | |
---|
512 | if (Utils.eq(Utils.sum(distribution), 0)) { |
---|
513 | for (int i = 0; i < distribution.length; i++) |
---|
514 | distribution[i] = 1.0 / (double) distribution.length; |
---|
515 | } |
---|
516 | else { |
---|
517 | Utils.normalize(distribution); |
---|
518 | } |
---|
519 | |
---|
520 | return distribution; |
---|
521 | } |
---|
522 | |
---|
523 | /** |
---|
524 | * Gets a string describing the classifier. |
---|
525 | * |
---|
526 | * @return a string describing the classifer built. |
---|
527 | */ |
---|
528 | public String toString() { |
---|
529 | return "MIWrapper with base classifier: \n"+m_Classifier.toString(); |
---|
530 | } |
---|
531 | |
---|
532 | /** |
---|
533 | * Returns the revision string. |
---|
534 | * |
---|
535 | * @return the revision |
---|
536 | */ |
---|
537 | public String getRevision() { |
---|
538 | return RevisionUtils.extract("$Revision: 1.5 $"); |
---|
539 | } |
---|
540 | |
---|
541 | /** |
---|
542 | * Main method for testing this class. |
---|
543 | * |
---|
544 | * @param argv should contain the command line arguments to the |
---|
545 | * scheme (see Evaluation) |
---|
546 | */ |
---|
547 | public static void main(String[] argv) { |
---|
548 | runClassifier(new MIWrapper(), argv); |
---|
549 | } |
---|
550 | } |
---|