-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathpbean.txt
246 lines (186 loc) · 9.87 KB
/
pbean.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
*****NEW RESULT*************************
***Pbean_B.arff***
best classifier: weka.classifiers.functions.SMO
arguments: [-C, 0.8930273415347204, -N, 0, -M, -K, weka.classifiers.functions.supportVector.Puk -S 0.8887234503578151 -O 0.5921472864175]
attribute search: null
attribute search arguments: []
attribute evaluation: null
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.038461538461538464
training time on evaluation dataset: 0.073 seconds
You can use the chosen classifier in your own code as follows:
Classifier classifier = AbstractClassifier.forName("weka.classifiers.functions.SMO", new String[]{"-C", "0.8930273415347204", "-N", "0", "-M", "-K", "weka.classifiers.functions.supportVector.Puk -S 0.8887234503578151 -O 0.5921472864175"});
classifier.buildClassifier(instances);
Correctly Classified Instances 25 96.1538 %
Incorrectly Classified Instances 1 3.8462 %
Kappa statistic 0.7797
Mean absolute error 0.0702
Root mean squared error 0.1933
Relative absolute error 31.1483 %
Root relative squared error 60.2775 %
Total Number of Instances 26
=== Confusion Matrix ===
a b <-- classified as
23 0 | a = FALSE
1 2 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
1.000 0.333 0.958 1.000 0.979 0.799 0.986 0.998 FALSE
0.667 0.000 1.000 0.667 0.800 0.799 0.986 0.917 TRUE
Weighted Avg. 0.962 0.295 0.963 0.962 0.958 0.799 0.986 0.989
Temporary run directories:
/tmp/autoweka6787825712646441939/
For better performance, try giving Auto-WEKA more time.
Tried 752 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
***Pbean_A.arff***
best classifier: weka.classifiers.lazy.IBk
arguments: [-E, -K, 14, -X, -F]
attribute search: weka.attributeSelection.GreedyStepwise
attribute search arguments: [-B, -N, 271]
attribute evaluation: weka.attributeSelection.CfsSubsetEval
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.23529411764705882
training time on evaluation dataset: 0.001 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", "271"});
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.lazy.IBk", new String[]{"-E", "-K", "14", "-X", "-F"});
classifier.buildClassifier(instances);
Correctly Classified Instances 39 76.4706 %
Incorrectly Classified Instances 12 23.5294 %
Kappa statistic 0.53
Mean absolute error 0.352
Root mean squared error 0.4193
Relative absolute error 70.6414 %
Root relative squared error 84.0043 %
Total Number of Instances 51
=== Confusion Matrix ===
a b <-- classified as
20 7 | a = TRUE
5 19 | b = FALSE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.741 0.208 0.800 0.741 0.769 0.532 0.812 0.849 TRUE
0.792 0.259 0.731 0.792 0.760 0.532 0.812 0.754 FALSE
Weighted Avg. 0.765 0.232 0.767 0.765 0.765 0.532 0.812 0.805
Temporary run directories:
/tmp/autoweka7069296824417443602/
For better performance, try giving Auto-WEKA more time.
Tried 532 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
***Pbean_A_O1.arff***
best classifier: weka.classifiers.lazy.LWL
arguments: [-K, -1, -A, weka.core.neighboursearch.LinearNNSearch, -W, weka.classifiers.rules.JRip, --, -N, 1.2611076313688434, -O, 5]
attribute search: null
attribute search arguments: []
attribute evaluation: null
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.2037037037037037
training time on evaluation dataset: 0.0 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.rules.JRip", "--", "-N", "1.2611076313688434", "-O", "5"});
classifier.buildClassifier(instances);
Correctly Classified Instances 43 79.6296 %
Incorrectly Classified Instances 11 20.3704 %
Kappa statistic 0.5926
Mean absolute error 0.2435
Root mean squared error 0.3532
Relative absolute error 48.6908 %
Root relative squared error 70.6459 %
Total Number of Instances 54
=== Confusion Matrix ===
a b <-- classified as
16 11 | a = TRUE
0 27 | b = FALSE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.593 0.000 1.000 0.593 0.744 0.649 0.914 0.908 TRUE
1.000 0.407 0.711 1.000 0.831 0.649 0.914 0.894 FALSE
Weighted Avg. 0.796 0.204 0.855 0.796 0.787 0.649 0.914 0.901
Temporary run directories:
/tmp/autoweka6419256368198990266/
For better performance, try giving Auto-WEKA more time.
Tried 580 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
***Pbean_A_S1.arff***
best classifier: weka.classifiers.meta.RandomSubSpace
arguments: [-I, 36, -P, 0.2757331204049802, -S, 1, -W, weka.classifiers.lazy.IBk, --, -E, -K, 10, -X]
attribute search: null
attribute search arguments: []
attribute evaluation: null
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.24074074074074073
training time on evaluation dataset: 0.102 seconds
You can use the chosen classifier in your own code as follows:
Classifier classifier = AbstractClassifier.forName("weka.classifiers.meta.RandomSubSpace", new String[]{"-I", "36", "-P", "0.2757331204049802", "-S", "1", "-W", "weka.classifiers.lazy.IBk", "--", "-E", "-K", "10", "-X"});
classifier.buildClassifier(instances);
Correctly Classified Instances 41 75.9259 %
Incorrectly Classified Instances 13 24.0741 %
Kappa statistic 0.5185
Mean absolute error 0.2952
Root mean squared error 0.3908
Relative absolute error 59.0398 %
Root relative squared error 78.1506 %
Total Number of Instances 54
=== Confusion Matrix ===
a b <-- classified as
21 6 | a = TRUE
7 20 | b = FALSE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.778 0.259 0.750 0.778 0.764 0.519 0.857 0.856 TRUE
0.741 0.222 0.769 0.741 0.755 0.519 0.857 0.826 FALSE
Weighted Avg. 0.759 0.241 0.760 0.759 0.759 0.519 0.857 0.841
Temporary run directories:
/tmp/autoweka7389772957988353950/
For better performance, try giving Auto-WEKA more time.
Tried 740 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
***Pbean_A_U1.arff***
best classifier: weka.classifiers.trees.DecisionStump
arguments: []
attribute search: null
attribute search arguments: []
attribute evaluation: null
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.20833333333333334
training time on evaluation dataset: 0.0 seconds
You can use the chosen classifier in your own code as follows:
Classifier classifier = AbstractClassifier.forName("weka.classifiers.trees.DecisionStump", new String[]{});
classifier.buildClassifier(instances);
Correctly Classified Instances 38 79.1667 %
Incorrectly Classified Instances 10 20.8333 %
Kappa statistic 0.5833
Mean absolute error 0.3287
Root mean squared error 0.4054
Relative absolute error 65.7343 %
Root relative squared error 81.0767 %
Total Number of Instances 48
=== Confusion Matrix ===
a b <-- classified as
18 6 | a = TRUE
4 20 | b = FALSE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.750 0.167 0.818 0.750 0.783 0.585 0.792 0.739 TRUE
0.833 0.250 0.769 0.833 0.800 0.585 0.792 0.724 FALSE
Weighted Avg. 0.792 0.208 0.794 0.792 0.791 0.585 0.792 0.731
Temporary run directories:
/tmp/autoweka7735617244047540885/
For better performance, try giving Auto-WEKA more time.
*********************************************