-
Notifications
You must be signed in to change notification settings - Fork 0
/
ar.txt
269 lines (208 loc) · 11.5 KB
/
ar.txt
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
*****NEW RESULT*************************
***ar-B.arff***
best classifier: weka.classifiers.lazy.LWL
arguments: [-K, -1, -A, weka.core.neighboursearch.LinearNNSearch, -W, weka.classifiers.trees.REPTree, --, -M, 38, -V, 3.3612402329737156E-4, -L, -1, -P]
attribute search: null
attribute search arguments: []
attribute evaluation: null
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.13986013986013987
training time on evaluation dataset: 0.001 seconds
You can use the chosen classifier in your own code as follows:
Classifier classifier = AbstractClassifier.forName("weka.classifiers.lazy.LWL", new String[]{"-K", "-1", "-A", "weka.core.neighboursearch.LinearNNSearch", "-W", "weka.classifiers.trees.REPTree", "--", "-M", "38", "-V", "3.3612402329737156E-4", "-L", "-1", "-P"});
classifier.buildClassifier(instances);
Correctly Classified Instances 123 86.014 %
Incorrectly Classified Instances 20 13.986 %
Kappa statistic 0.3128
Mean absolute error 0.2453
Root mean squared error 0.3412
Relative absolute error 89.8199 %
Root relative squared error 92.8718 %
Total Number of Instances 143
=== Confusion Matrix ===
a b <-- classified as
117 3 | a = FALSE
17 6 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.975 0.739 0.873 0.975 0.921 0.357 0.757 0.933 FALSE
0.261 0.025 0.667 0.261 0.375 0.357 0.757 0.517 TRUE
Weighted Avg. 0.860 0.624 0.840 0.860 0.833 0.357 0.757 0.866
Temporary run directories:
/tmp/autoweka6605315650889576295/
For better performance, try giving Auto-WEKA more time.
Tried 976 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
**ar_A.arff**
best classifier: weka.classifiers.bayes.BayesNet
arguments: [-Q, weka.classifiers.bayes.net.search.local.TAN]
attribute search: weka.attributeSelection.GreedyStepwise
attribute search arguments: [-B, -N, 147]
attribute evaluation: weka.attributeSelection.CfsSubsetEval
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.0912280701754386
training time on evaluation dataset: 0.053 seconds
You can use the chosen classifier in your own code as follows:
AttributeSelection as = new AttributeSelection();
ASSearch asSearch = ASSearch.forName("weka.attributeSelection.GreedyStepwise", new String[]{"-B", "-N", "147"});
as.setSearch(asSearch);
ASEvaluation asEval = ASEvaluation.forName("weka.attributeSelection.CfsSubsetEval", new String[]{});
as.setEvaluator(asEval);
as.SelectAttributes(instances);
instances = as.reduceDimensionality(instances);
Classifier classifier = AbstractClassifier.forName("weka.classifiers.bayes.BayesNet", new String[]{"-Q", "weka.classifiers.bayes.net.search.local.TAN"});
classifier.buildClassifier(instances);
Correctly Classified Instances 259 90.8772 %
Incorrectly Classified Instances 26 9.1228 %
Kappa statistic 0.4757
Mean absolute error 0.1448
Root mean squared error 0.2686
Relative absolute error 63.5721 %
Root relative squared error 79.8995 %
Total Number of Instances 285
=== Confusion Matrix ===
a b <-- classified as
245 3 | a = FALSE
23 14 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.988 0.622 0.914 0.988 0.950 0.520 0.809 0.947 FALSE
0.378 0.012 0.824 0.378 0.519 0.520 0.809 0.528 TRUE
Weighted Avg. 0.909 0.542 0.902 0.909 0.894 0.520 0.809 0.893
Temporary run directories:
/tmp/autoweka4248923110847910918/
For better performance, try giving Auto-WEKA more time.
Tried 547 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
**ar_A_O1.arff**
best classifier: weka.classifiers.functions.SMO
arguments: [-C, 0.6680673876996616, -N, 2, -M, -K, weka.classifiers.functions.supportVector.RBFKernel -G 0.2582956076700777]
attribute search: weka.attributeSelection.BestFirst
attribute search arguments: [-D, 2, -N, 9]
attribute evaluation: weka.attributeSelection.CfsSubsetEval
attribute evaluation arguments: [-M]
metric: errorRate
estimated errorRate: 0.018145161290322582
training time on evaluation dataset: 0.492 seconds
You can use the chosen classifier in your own code as follows:
AttributeSelection as = new AttributeSelection();
ASSearch asSearch = ASSearch.forName("weka.attributeSelection.BestFirst", new String[]{"-D", "2", "-N", "9"});
as.setSearch(asSearch);
ASEvaluation asEval = ASEvaluation.forName("weka.attributeSelection.CfsSubsetEval", new String[]{"-M"});
as.setEvaluator(asEval);
as.SelectAttributes(instances);
instances = as.reduceDimensionality(instances);
Classifier classifier = AbstractClassifier.forName("weka.classifiers.functions.SMO", new String[]{"-C", "0.6680673876996616", "-N", "2", "-M", "-K", "weka.classifiers.functions.supportVector.RBFKernel -G 0.2582956076700777"});
classifier.buildClassifier(instances);
Correctly Classified Instances 487 98.1855 %
Incorrectly Classified Instances 9 1.8145 %
Kappa statistic 0.9637
Mean absolute error 0.0395
Root mean squared error 0.1398
Relative absolute error 7.898 %
Root relative squared error 27.9515 %
Total Number of Instances 496
=== Confusion Matrix ===
a b <-- classified as
239 9 | a = FALSE
0 248 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.964 0.000 1.000 0.964 0.982 0.964 0.989 0.992 FALSE
1.000 0.036 0.965 1.000 0.982 0.964 0.989 0.987 TRUE
Weighted Avg. 0.982 0.018 0.982 0.982 0.982 0.964 0.989 0.989
Temporary run directories:
/tmp/autoweka5494895306790478269/
For better performance, try giving Auto-WEKA more time.
Tried 403 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
**ar_A_S1.arff**
best classifier: weka.classifiers.meta.Vote
arguments: [-R, MIN, -S, 1, -B, weka.classifiers.rules.OneR -B 3, -B, weka.classifiers.functions.SimpleLogistic -S -W 0 -A]
attribute search: weka.attributeSelection.BestFirst
attribute search arguments: [-D, 2, -N, 4]
attribute evaluation: weka.attributeSelection.CfsSubsetEval
attribute evaluation arguments: [-M, -L]
metric: errorRate
estimated errorRate: 0.07862903225806452
training time on evaluation dataset: 0.145 seconds
You can use the chosen classifier in your own code as follows:
AttributeSelection as = new AttributeSelection();
ASSearch asSearch = ASSearch.forName("weka.attributeSelection.BestFirst", new String[]{"-D", "2", "-N", "4"});
as.setSearch(asSearch);
ASEvaluation asEval = ASEvaluation.forName("weka.attributeSelection.CfsSubsetEval", new String[]{"-M", "-L"});
as.setEvaluator(asEval);
as.SelectAttributes(instances);
instances = as.reduceDimensionality(instances);
Classifier classifier = AbstractClassifier.forName("weka.classifiers.meta.Vote", new String[]{"-R", "MIN", "-S", "1", "-B", "weka.classifiers.rules.OneR -B 3", "-B", "weka.classifiers.functions.SimpleLogistic -S -W 0 -A"});
classifier.buildClassifier(instances);
Correctly Classified Instances 457 92.1371 %
Incorrectly Classified Instances 39 7.8629 %
Kappa statistic 0.8427
Mean absolute error 0.0786
Root mean squared error 0.2804
Relative absolute error 15.7258 %
Root relative squared error 56.0817 %
Total Number of Instances 496
=== Confusion Matrix ===
a b <-- classified as
239 9 | a = FALSE
30 218 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.964 0.121 0.888 0.964 0.925 0.846 0.921 0.874 FALSE
0.879 0.036 0.960 0.879 0.918 0.846 0.921 0.905 TRUE
Weighted Avg. 0.921 0.079 0.924 0.921 0.921 0.846 0.921 0.890
Temporary run directories:
/tmp/autoweka2231559370811610385/
For better performance, try giving Auto-WEKA more time.
Tried 390 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
**ar_A_U1.arff**
best classifier: weka.classifiers.rules.PART
arguments: [-N, 4, -M, 22, -R, -B]
attribute search: weka.attributeSelection.GreedyStepwise
attribute search arguments: [-B, -R]
attribute evaluation: weka.attributeSelection.CfsSubsetEval
attribute evaluation arguments: [-M]
metric: errorRate
estimated errorRate: 0.21621621621621623
training time on evaluation dataset: 0.014 seconds
You can use the chosen classifier in your own code as follows:
AttributeSelection as = new AttributeSelection();
ASSearch asSearch = ASSearch.forName("weka.attributeSelection.GreedyStepwise", new String[]{"-B", "-R"});
as.setSearch(asSearch);
ASEvaluation asEval = ASEvaluation.forName("weka.attributeSelection.CfsSubsetEval", new String[]{"-M"});
as.setEvaluator(asEval);
as.SelectAttributes(instances);
instances = as.reduceDimensionality(instances);
Classifier classifier = AbstractClassifier.forName("weka.classifiers.rules.PART", new String[]{"-N", "4", "-M", "22", "-R", "-B"});
classifier.buildClassifier(instances);
Correctly Classified Instances 58 78.3784 %
Incorrectly Classified Instances 16 21.6216 %
Kappa statistic 0.5676
Mean absolute error 0.3552
Root mean squared error 0.4118
Relative absolute error 71.0304 %
Root relative squared error 82.3652 %
Total Number of Instances 74
=== Confusion Matrix ===
a b <-- classified as
27 10 | a = FALSE
6 31 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.730 0.162 0.818 0.730 0.771 0.571 0.784 0.732 FALSE
0.838 0.270 0.756 0.838 0.795 0.571 0.784 0.715 TRUE
Weighted Avg. 0.784 0.216 0.787 0.784 0.783 0.571 0.784 0.723
Temporary run directories:
/tmp/autoweka3759300770611324689/
For better performance, try giving Auto-WEKA more time.
Tried 691 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************