/
SVM_results_12_4_19.txt
322 lines (287 loc) · 13.3 KB
/
SVM_results_12_4_19.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
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
11/22 version of data (feat_df.csv)
Models run 12/3/19
Output from SVM_models.py
Baseline, MDB data only/75/25 split, kernel= linear, 6 instrument classes
features= ['centroid', 'rms', 'zeroCrossings', 'crest', 'flux', 'mfcc1', 'mfcc2', 'mfcc3', 'mfcc4', 'mfcc5']
[[ 0 61 0 0 0 0]
[ 0 484 0 0 0 0]
[ 0 284 0 0 0 0]
[ 0 25 0 0 0 0]
[ 0 802 0 0 0 0]
[ 0 30 0 0 0 0]]
C:\Users\Laney\AppData\Local\Programs\Python\Python37-32\lib\site-packages\sklearn\metrics\classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.
'precision', 'predicted', average, warn_for)
precision recall f1-score support
CY 0.00 0.00 0.00 61
HH 0.29 1.00 0.45 484
KD 0.00 0.00 0.00 284
OT 0.00 0.00 0.00 25
SD 0.00 0.00 0.00 802
TT 0.00 0.00 0.00 30
accuracy 0.29 1686
macro avg 0.05 0.17 0.07 1686
weighted avg 0.08 0.29 0.13 1686
############################################################
Baseline, MDB data only, 75/25 split, kernel= sigmoid
features= ['centroid', 'rms', 'zeroCrossings', 'crest', 'flux', 'mfcc1', 'mfcc2', 'mfcc3', 'mfcc4', 'mfcc5']
[[ 13 32 0 0 16 0]
[140 250 0 0 92 2]
[ 76 145 0 0 63 0]
[ 15 9 0 0 1 0]
[260 358 0 0 178 6]
[ 10 11 0 0 9 0]]
C:\Users\Laney\AppData\Local\Programs\Python\Python37-32\lib\site-packages\sklearn\metrics\classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.
'precision', 'predicted', average, warn_for)
precision recall f1-score support
CY 0.03 0.21 0.05 61
HH 0.31 0.52 0.39 484
KD 0.00 0.00 0.00 284
OT 0.00 0.00 0.00 25
SD 0.50 0.22 0.31 802
TT 0.00 0.00 0.00 30
accuracy 0.26 1686
macro avg 0.14 0.16 0.12 1686
weighted avg 0.33 0.26 0.26 1686
############################################################
Baseline, MDB data only, 75/25 split, kernel= poly, deg=2
features= ['centroid', 'rms', 'zeroCrossings', 'crest', 'flux', 'mfcc1', 'mfcc2', 'mfcc3', 'mfcc4', 'mfcc5']
[[ 0 61 0 0 0 0]
[ 0 380 0 0 104 0]
[ 0 239 0 0 45 0]
[ 0 22 0 0 3 0]
[ 0 659 0 0 143 0]
[ 0 30 0 0 0 0]]
C:\Users\Laney\AppData\Local\Programs\Python\Python37-32\lib\site-packages\sklearn\metrics\classification.py:1437: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples.
'precision', 'predicted', average, warn_for)
precision recall f1-score support
CY 0.00 0.00 0.00 61
HH 0.27 0.79 0.41 484
KD 0.00 0.00 0.00 284
OT 0.00 0.00 0.00 25
SD 0.48 0.18 0.26 802
TT 0.00 0.00 0.00 30
accuracy 0.31 1686
macro avg 0.13 0.16 0.11 1686
weighted avg 0.31 0.31 0.24 1686
############################################################
Baseline, MDB data only, 10 fold cross validation
fScores= [0.13 0.21 0.3 0.27 0.3 0.24 0.15 0.56 0.17 0.27]
average F score for MDB cross validation= 0.26
confusion matrices for each fold of MDB cross validation= [array([[ 1, 3, 0, 0, 2, 0],
[ 18, 108, 3, 0, 14, 0],
[ 19, 87, 7, 0, 6, 0],
[ 2, 17, 3, 0, 3, 0],
[ 57, 275, 31, 0, 20, 0],
[ 1, 19, 4, 0, 3, 0]], dtype=int64), array([[ 1, 4, 0, 0, 0, 0],
[ 11, 107, 9, 0, 20, 0],
[ 13, 48, 5, 0, 12, 0],
[ 5, 12, 4, 0, 6, 0],
[ 37, 198, 13, 0, 29, 2],
[ 0, 0, 0, 0, 0, 0]], dtype=int64), array([[ 26, 20, 49, 3, 40, 0],
[ 19, 177, 189, 6, 281, 0],
[ 30, 24, 68, 2, 77, 0],
[ 0, 0, 0, 0, 0, 0],
[ 67, 85, 164, 4, 158, 1],
[ 5, 4, 4, 3, 6, 0]], dtype=int64), array([[29, 30, 3, 0, 16, 0],
[46, 65, 12, 1, 55, 0],
[47, 57, 13, 0, 55, 0],
[ 0, 0, 0, 0, 0, 0],
[46, 46, 5, 0, 66, 0],
[ 5, 4, 1, 0, 3, 0]], dtype=int64), array([[ 0, 1, 0, 0, 0, 0],
[16, 69, 3, 0, 28, 1],
[ 7, 49, 0, 1, 12, 0],
[ 0, 0, 0, 0, 0, 0],
[ 8, 31, 0, 0, 15, 0],
[ 0, 5, 0, 0, 0, 0]], dtype=int64), array([[ 11, 91, 9, 135, 0],
[ 10, 79, 10, 128, 2],
[ 2, 44, 4, 40, 0],
[ 9, 144, 12, 156, 0],
[ 0, 0, 0, 1, 0]], dtype=int64), array([[ 4, 12, 1, 0, 8, 0],
[10, 25, 1, 1, 11, 0],
[16, 37, 0, 0, 16, 0],
[ 0, 2, 0, 0, 0, 0],
[ 4, 18, 1, 0, 9, 1],
[ 3, 0, 0, 0, 1, 0]], dtype=int64), array([[ 0, 0, 0, 0, 0],
[ 8, 69, 2, 24, 2],
[ 1, 13, 2, 2, 0],
[ 0, 14, 0, 6, 1],
[ 0, 0, 0, 0, 0]], dtype=int64), array([[19, 48, 3, 21, 0],
[22, 36, 1, 4, 1],
[37, 61, 4, 8, 1],
[16, 28, 1, 9, 0],
[ 0, 0, 0, 0, 0]], dtype=int64), array([[ 1, 3, 0, 0, 1, 0],
[34, 37, 22, 2, 17, 5],
[21, 24, 9, 4, 28, 3],
[ 0, 0, 0, 0, 0, 0],
[50, 41, 7, 4, 32, 3],
[ 0, 0, 0, 0, 0, 0]], dtype=int64)]
############################################################
MDB data only, 3 instrument classes, baseline set of 10 features, sigmoid kernel
precision recall f1-score support
HH 0.34 0.89 0.49 133
KD 0.00 0.00 0.00 61
SD 0.49 0.12 0.19 206
accuracy 0.36 400
macro avg 0.28 0.34 0.23 400
weighted avg 0.36 0.36 0.26 400
MDB data only, 3 instrument classes, baseline set of 10 features, sigmoid kernel
F= 0.26
confusion matrix=
[[118 0 15]
[ 50 0 11]
[181 0 25]]
############################################################
All Data (ESNT+MDB), baseline set of 10 features, sigmoid kernel
precision recall f1-score support
HH 0.38 0.66 0.48 524
KD 0.25 0.08 0.12 299
SD 0.34 0.21 0.26 492
accuracy 0.36 1315
macro avg 0.32 0.32 0.29 1315
weighted avg 0.34 0.36 0.32 1315
F= 0.32
confusion matrix=
[[347 30 147]
[215 24 60]
[346 41 105]]
##############################################################
ENST data only, baseline set of 10 features, sigmoid kernel
precision recall f1-score support
HH 0.28 0.49 0.35 836
KD 0.34 0.46 0.39 1027
SD 0.27 0.00 0.01 1034
accuracy 0.31 2897
macro avg 0.30 0.32 0.25 2897
weighted avg 0.30 0.31 0.24 2897
F= 0.24
confusion matrix=
[[410 423 3]
[545 477 5]
[529 502 3]]
############################################################
MDB data only, 3 instrument classes, set of 9 features from logistic reg, sigmoid kernel
precision recall f1-score support
HH 0.29 0.30 0.29 133
KD 0.10 0.15 0.12 61
SD 0.53 0.45 0.48 206
accuracy 0.35 400
macro avg 0.31 0.30 0.30 400
weighted avg 0.38 0.35 0.37 400
F= 0.37
confusion matrix=
[[40 40 53]
[23 9 29]
[76 38 92]]
############################################################
All Data (ESNT+MDB), set of 9 features from logistic reg, sigmoid kernel
precision recall f1-score support
HH 0.47 0.22 0.30 524
KD 0.19 0.46 0.27 299
SD 0.33 0.24 0.28 492
accuracy 0.28 1315
macro avg 0.33 0.31 0.28 1315
weighted avg 0.35 0.28 0.28 1315
F= 0.28
confusion matrix=
[[113 261 150]
[ 66 138 95]
[ 61 313 118]]
###############################################################
ENST data only, set of 9 features from logistic reg, sigmoid kernel
precision recall f1-score support
HH 0.25 0.24 0.25 836
KD 0.35 0.34 0.34 1027
SD 0.32 0.34 0.33 1034
accuracy 0.31 2897
macro avg 0.31 0.31 0.31 2897
weighted avg 0.31 0.31 0.31 2897
F= 0.31
confusion matrix= [[203 314 319]
[254 348 425]
[351 334 349]]
#############################################################
MDB only, 3 instrument classes, set of 13 features from logistic reg and/or dist. plots, sigmoid kernel
precision recall f1-score support
HH 0.29 0.52 0.37 133
KD 0.07 0.07 0.07 61
SD 0.49 0.24 0.32 206
accuracy 0.31 400
macro avg 0.28 0.28 0.25 400
weighted avg 0.36 0.31 0.30 400
F= 0.3
confusion matrix= [[ 69 30 34]
[ 39 4 18]
[133 23 50]]
#############################################################
All Data (ESNT+MDB), set of 13 features from logistic reg and/or dist. plots, sigmoid kernel
precision recall f1-score support
HH 0.40 0.60 0.48 524
KD 0.20 0.31 0.24 299
SD 0.22 0.03 0.05 492
accuracy 0.32 1315
macro avg 0.27 0.31 0.26 1315
weighted avg 0.29 0.32 0.27 1315
F= 0.27
confusion matrix= [[313 179 32]
[186 93 20]
[284 193 15]]
###############################################################
ENST data only, set of 13 features from logistic reg and/or dist. plots, sigmoid kernel
precision recall f1-score support
HH 0.28 0.55 0.37 836
KD 0.37 0.29 0.32 1027
SD 0.44 0.19 0.26 1034
accuracy 0.33 2897
macro avg 0.36 0.34 0.32 2897
weighted avg 0.37 0.33 0.32 2897
ENST data only, optimal set of 13 features, sigmoid kernel
F= 0.32
confusion matrix=
[[463 247 126]
[606 293 128]
[584 254 196]]
###############################################################
MDB only, set of 13 features from logistic reg and/or dist. plots, poly kernel deg 4
precision recall f1-score support
HH 0.36 0.83 0.50 133
KD 0.16 0.05 0.07 61
SD 0.66 0.24 0.35 206
accuracy 0.41 400
macro avg 0.39 0.37 0.31 400
weighted avg 0.49 0.41 0.36 400
ENST data only, set of 13 features from logistic reg and/or dist. plots, poly kernel deg 4
F= 0.36
confusion matrix=
[[111 9 13]
[ 46 3 12]
[150 7 49]]
###############################################################
MDB+ENST, set of 13 features from logistic reg and/or dist. plots, poly kernel deg 4
precision recall f1-score support
HH 0.37 0.03 0.06 524
KD 0.25 0.25 0.25 299
SD 0.38 0.74 0.50 492
accuracy 0.35 1315
macro avg 0.33 0.34 0.27 1315
weighted avg 0.34 0.35 0.27 1315
All Data (ESNT+MDB), set of 13 features from logistic reg and/or dist. plots, poly kernel deg 4
F= 0.27
confusion matrix=
[[ 18 119 387]
[ 9 76 214]
[ 22 108 362]]
###############################################################
ENST data only, set of 13 features from logistic reg and/or dist. plots, poly kernel deg 4
precision recall f1-score support
HH 0.35 0.80 0.49 133
KD 0.00 0.00 0.00 61
SD 0.56 0.26 0.36 206
accuracy 0.40 400
macro avg 0.30 0.35 0.28 400
weighted avg 0.40 0.40 0.35 400
F= 0.35
confusion matrix=
[[106 0 27]
[ 45 0 16]
[151 1 54]]
###############################################################