-
Notifications
You must be signed in to change notification settings - Fork 0
/
syn.txt
261 lines (201 loc) · 11 KB
/
syn.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
*****NEW RESULT*************************
***Syn_B.arff***
best classifier: weka.classifiers.meta.AdaBoostM1
arguments: [-P, 100, -I, 33, -Q, -S, 1, -W, weka.classifiers.trees.DecisionStump, --]
attribute search: weka.attributeSelection.BestFirst
attribute search arguments: [-D, 0, -N, 10]
attribute evaluation: weka.attributeSelection.CfsSubsetEval
attribute evaluation arguments: [-M]
metric: errorRate
estimated errorRate: 0.19718309859154928
training time on evaluation dataset: 0.19 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", "0", "-N", "10"});
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.meta.AdaBoostM1", new String[]{"-P", "100", "-I", "33", "-Q", "-S", "1", "-W", "weka.classifiers.trees.DecisionStump", "--"});
classifier.buildClassifier(instances);
Correctly Classified Instances 171 80.2817 %
Incorrectly Classified Instances 42 19.7183 %
Kappa statistic 0.517
Mean absolute error 0.2664
Root mean squared error 0.366
Relative absolute error 60.7465 %
Root relative squared error 78.2059 %
Total Number of Instances 213
=== Confusion Matrix ===
a b <-- classified as
132 12 | a = FALSE
30 39 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.917 0.435 0.815 0.917 0.863 0.528 0.876 0.934 FALSE
0.565 0.083 0.765 0.565 0.650 0.528 0.876 0.787 TRUE
Weighted Avg. 0.803 0.321 0.799 0.803 0.794 0.528 0.876 0.886
Temporary run directories:
/tmp/autoweka8180312502107488512/
For better performance, try giving Auto-WEKA more time.
Tried 548 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
***Syn_A.arff***
best classifier: weka.classifiers.functions.MultilayerPerceptron
arguments: [-L, 0.620304021165965, -M, 0.5434011542858376, -H, i, -C, -R, -D, -S, 1]
attribute search: weka.attributeSelection.GreedyStepwise
attribute search arguments: [-R]
attribute evaluation: weka.attributeSelection.CfsSubsetEval
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.17772511848341233
training time on evaluation dataset: 0.963 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[]{"-R"});
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.functions.MultilayerPerceptron", new String[]{"-L", "0.620304021165965", "-M", "0.5434011542858376", "-H", "i", "-C", "-R", "-D", "-S", "1"});
classifier.buildClassifier(instances);
Correctly Classified Instances 347 82.2275 %
Incorrectly Classified Instances 75 17.7725 %
Kappa statistic 0.3499
Mean absolute error 0.2672
Root mean squared error 0.3664
Relative absolute error 77.5847 %
Root relative squared error 88.3867 %
Total Number of Instances 422
=== Confusion Matrix ===
a b <-- classified as
318 11 | a = FALSE
64 29 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.967 0.688 0.832 0.967 0.895 0.394 0.772 0.908 FALSE
0.312 0.033 0.725 0.312 0.436 0.394 0.772 0.561 TRUE
Weighted Avg. 0.822 0.544 0.809 0.822 0.793 0.394 0.772 0.832
Temporary run directories:
/tmp/autoweka9062503481717770915/
For better performance, try giving Auto-WEKA more time.
Tried 331 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
***Syn_A_O1.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: 0.02127659574468085
training time on evaluation dataset: 0.12 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 590 89.6657 %
Incorrectly Classified Instances 68 10.3343 %
Kappa statistic 0.7933
Mean absolute error 0.3022
Root mean squared error 0.3373
Relative absolute error 60.4384 %
Root relative squared error 67.4587 %
Total Number of Instances 658
=== Confusion Matrix ===
a b <-- classified as
314 15 | a = FALSE
53 276 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.954 0.161 0.856 0.954 0.902 0.799 0.969 0.971 FALSE
0.839 0.046 0.948 0.839 0.890 0.799 0.969 0.970 TRUE
Weighted Avg. 0.897 0.103 0.902 0.897 0.896 0.799 0.969 0.970
Temporary run directories:
/tmp/autoweka3596922266898421272/
For better performance, try giving Auto-WEKA more time.
Tried 319 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
***Syn_A_S1.arff***
best classifier: weka.classifiers.trees.RandomForest
arguments: [-I, 10, -K, 0, -depth, 0]
attribute search: null
attribute search arguments: []
attribute evaluation: null
attribute evaluation arguments: []
metric: errorRate
estimated errorRate: 0.028875379939209727
training time on evaluation dataset: 0.113 seconds
You can use the chosen classifier in your own code as follows:
Classifier classifier = AbstractClassifier.forName("weka.classifiers.trees.RandomForest", new String[]{"-I", "10", "-K", "0", "-depth", "0"});
classifier.buildClassifier(instances);
Correctly Classified Instances 642 97.5684 %
Incorrectly Classified Instances 16 2.4316 %
Kappa statistic 0.9514
Mean absolute error 0.1078
Root mean squared error 0.178
Relative absolute error 21.5542 %
Root relative squared error 35.5922 %
Total Number of Instances 658
=== Confusion Matrix ===
a b <-- classified as
321 8 | a = FALSE
8 321 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.976 0.024 0.976 0.976 0.976 0.951 0.998 0.998 FALSE
0.976 0.024 0.976 0.976 0.976 0.951 0.998 0.998 TRUE
Weighted Avg. 0.976 0.024 0.976 0.976 0.976 0.951 0.998 0.998
Temporary run directories:
/tmp/autoweka8838425426889866877/
For better performance, try giving Auto-WEKA more time.
Tried 298 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************
*****NEW RESULT*************************
***Syn_A_U1.arff***
best classifier: weka.classifiers.lazy.LWL
arguments: [-A, weka.core.neighboursearch.LinearNNSearch, -W, weka.classifiers.trees.DecisionStump, --]
attribute search: weka.attributeSelection.BestFirst
attribute search arguments: [-D, 0, -N, 5]
attribute evaluation: weka.attributeSelection.CfsSubsetEval
attribute evaluation arguments: [-M]
metric: errorRate
estimated errorRate: 0.2849462365591398
training time on evaluation dataset: 0.0 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", "0", "-N", "5"});
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.lazy.LWL", new String[]{"-A", "weka.core.neighboursearch.LinearNNSearch", "-W", "weka.classifiers.trees.DecisionStump", "--"});
classifier.buildClassifier(instances);
Correctly Classified Instances 133 71.5054 %
Incorrectly Classified Instances 53 28.4946 %
Kappa statistic 0.4301
Mean absolute error 0.3935
Root mean squared error 0.4413
Relative absolute error 78.7078 %
Root relative squared error 88.2546 %
Total Number of Instances 186
=== Confusion Matrix ===
a b <-- classified as
78 15 | a = FALSE
38 55 | b = TRUE
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure MCC ROC Area PRC Area Class
0.839 0.409 0.672 0.839 0.746 0.444 0.792 0.785 FALSE
0.591 0.161 0.786 0.591 0.675 0.444 0.792 0.821 TRUE
Weighted Avg. 0.715 0.285 0.729 0.715 0.711 0.444 0.792 0.803
Temporary run directories:
/tmp/autoweka7612245354958617689/
For better performance, try giving Auto-WEKA more time.
Tried 811 configurations; to get good results reliably you may need to allow for trying thousands of configurations.
*********************************************