-
Notifications
You must be signed in to change notification settings - Fork 1
Expand file tree
/
Copy pathVergleiche_DT_IEMOvsMSP
More file actions
304 lines (236 loc) · 9.93 KB
/
Vergleiche_DT_IEMOvsMSP
File metadata and controls
304 lines (236 loc) · 9.93 KB
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
IEMO vs MSP
IEMO classes: angry(0), excited(2), neutral(6), sad(8)
merged excited(2) and happy(5):
Accuracy: 0.5355421686746988
precision recall f1-score support
0.0 0.59 0.60 0.60 327
2.0 0.50 0.44 0.47 502
6.0 0.50 0.57 0.54 537
8.0 0.60 0.56 0.58 294
accuracy 0.53 1660
macro avg 0.55 0.54 0.54 1660
weighted avg 0.53 0.53 0.53 1660
[[194 69 53 11]
[ 88 219 156 39]
[ 34 134 307 62]
[ 11 21 97 165]]
pred: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([327, 443, 613, 277]))
test: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([327, 502, 537, 294]))
train: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([ 776, 1134, 1171, 790]))
no merge:
Accuracy: 0.5408507765023632
precision recall f1-score support
0.0 0.56 0.61 0.59 339
2.0 0.42 0.38 0.40 328
6.0 0.55 0.55 0.55 497
8.0 0.62 0.61 0.62 317
..............................................................
MSP classes: angry(A), happy(H), neutral(N), sad(S)
removed Preperation data (P-folder):
Accuracy: 0.5071059431524548
precision recall f1-score support
A 0.36 0.25 0.30 220
H 0.42 0.42 0.42 358
N 0.57 0.69 0.63 718
S 0.49 0.31 0.38 252
accuracy 0.51 1548
macro avg 0.46 0.42 0.43 1548
weighted avg 0.49 0.51 0.49 1548
[[ 55 78 82 5]
[ 31 153 161 13]
[ 52 105 499 62]
[ 12 26 136 78]]
pred: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([150, 362, 878, 158]))
test: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([220, 358, 718, 252]))
train: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([ 534, 914, 1595, 567]))
include Prep data:
Accuracy: 0.4931623931623932
precision recall f1-score support
A 0.32 0.25 0.28 233
H 0.48 0.49 0.49 779
N 0.54 0.66 0.60 1051
S 0.23 0.06 0.10 277
--------------------------------------------------------------
--------------------------------------------------------------
compare two classes:
angry vs happy
IEMO (excited instead of happy)
Accuracy: 0.6599378881987578
precision recall f1-score support
0.0 0.68 0.64 0.66 335
2.0 0.64 0.68 0.66 309
accuracy 0.66 644
macro avg 0.66 0.66 0.66 644
weighted avg 0.66 0.66 0.66 644
[[216 119]
[100 209]]
pred: (array([0., 2.]), array([316, 328]))
test: (array([0., 2.]), array([335, 309]))
train: (array([0., 2.]), array([768, 732]))
IEMO (excited merged with happy)
Accuracy: 0.6776155717761557
precision recall f1-score support
0.0 0.58 0.67 0.62 326
2.0 0.76 0.68 0.72 496
accuracy 0.68 822
macro avg 0.67 0.68 0.67 822
weighted avg 0.69 0.68 0.68 822
[[220 106]
[159 337]]
pred: (array([0., 2.]), array([379, 443]))
test: (array([0., 2.]), array([326, 496]))
train: (array([0., 2.]), array([ 777, 1140]))
MSP
Accuracy: 0.65625
precision recall f1-score support
A 0.53 0.48 0.51 222
H 0.72 0.76 0.74 386
accuracy 0.66 608
macro avg 0.62 0.62 0.62 608
weighted avg 0.65 0.66 0.65 608
[[107 115]
[ 94 292]]
pred: (array(['A', 'H'], dtype='<U1'), array([201, 407]))
test: (array(['A', 'H'], dtype='<U1'), array([222, 386]))
train: (array(['A', 'H'], dtype='<U1'), array([532, 886]))
______________________________________________________________
angry vs neutral
IEMO
Accuracy: 0.8080568720379147
precision recall f1-score support
0.0 0.76 0.74 0.75 327
6.0 0.84 0.85 0.84 517
accuracy 0.81 844
macro avg 0.80 0.80 0.80 844
weighted avg 0.81 0.81 0.81 844
[[243 84]
[ 78 439]]
pred: (array([0., 6.]), array([321, 523]))
test: (array([0., 6.]), array([327, 517]))
train: (array([0., 6.]), array([ 776, 1191]))
MSP
Accuracy: 0.745928338762215
precision recall f1-score support
A 0.48 0.45 0.46 225
N 0.83 0.84 0.83 696
accuracy 0.75 921
macro avg 0.65 0.65 0.65 921
weighted avg 0.74 0.75 0.74 921
[[101 124]
[110 586]]
pred: (array(['A', 'N'], dtype='<U1'), array([211, 710]))
test: (array(['A', 'N'], dtype='<U1'), array([225, 696]))
train: (array(['A', 'N'], dtype='<U1'), array([ 529, 1617]))
______________________________________________________________
angry vs sad:
IEMO
Accuracy: 0.8797564687975646
precision recall f1-score support
0.0 0.88 0.88 0.88 330
8.0 0.88 0.87 0.88 327
accuracy 0.88 657
macro avg 0.88 0.88 0.88 657
weighted avg 0.88 0.88 0.88 657
[[292 38]
[ 41 286]]
pred: (array([0., 8.]), array([333, 324]))
test: (array([0., 8.]), array([330, 327]))
train: (array([0., 8.]), array([773, 757]))
MSP
Accuracy: 0.7627118644067796
precision recall f1-score support
A 0.81 0.71 0.75 241
S 0.73 0.82 0.77 231
accuracy 0.76 472
macro avg 0.77 0.76 0.76 472
weighted avg 0.77 0.76 0.76 472
[[170 71]
[ 41 190]]
pred: (array(['A', 'S'], dtype='<U1'), array([211, 261]))
test: (array(['A', 'S'], dtype='<U1'), array([241, 231]))
train: (array(['A', 'S'], dtype='<U1'), array([513, 588]))
______________________________________________________________
______________________________________________________________
happy vs neutral:
IEMO (merged)
Accuracy: 0.6683266932270916
precision recall f1-score support
2.0 0.67 0.65 0.66 502
6.0 0.66 0.68 0.67 502
accuracy 0.67 1004
macro avg 0.67 0.67 0.67 1004
weighted avg 0.67 0.67 0.67 1004
[[328 174]
[159 343]]
pred: (array([2., 6.]), array([487, 517]))
test: (array([2., 6.]), array([502, 502]))
train: (array([2., 6.]), array([1134, 1206]))
MSP
Accuracy: 0.6579925650557621
precision recall f1-score support
H 0.51 0.45 0.48 375
N 0.72 0.77 0.75 701
accuracy 0.66 1076
macro avg 0.62 0.61 0.61 1076
weighted avg 0.65 0.66 0.65 1076
[[169 206]
[162 539]]
pred: (array(['H', 'N'], dtype='<U1'), array([331, 745]))
test: (array(['H', 'N'], dtype='<U1'), array([375, 701]))
train: (array(['H', 'N'], dtype='<U1'), array([ 897, 1612]))
______________________________________________________________
happy vs sad:
IEMO (merged)
Accuracy: 0.8026960784313726
precision recall f1-score support
2.0 0.84 0.82 0.83 488
8.0 0.75 0.77 0.76 328
accuracy 0.80 816
macro avg 0.79 0.80 0.80 816
weighted avg 0.80 0.80 0.80 816
[[402 86]
[ 75 253]]
pred: (array([2., 8.]), array([477, 339]))
test: (array([2., 8.]), array([488, 328]))
train: (array([2., 8.]), array([1148, 756]))
MSP
Accuracy: 0.7054140127388535
precision recall f1-score support
H 0.76 0.76 0.76 387
S 0.61 0.62 0.62 241
accuracy 0.71 628
macro avg 0.69 0.69 0.69 628
weighted avg 0.71 0.71 0.71 628
[[293 94]
[ 91 150]]
pred: (array(['H', 'S'], dtype='<U1'), array([384, 244]))
test: (array(['H', 'S'], dtype='<U1'), array([387, 241]))
train: (array(['H', 'S'], dtype='<U1'), array([885, 578]))
______________________________________________________________
neutral vs sad:
IEMO
Accuracy: 0.7315035799522673
precision recall f1-score support
6.0 0.78 0.79 0.79 521
8.0 0.65 0.63 0.64 317
accuracy 0.73 838
macro avg 0.71 0.71 0.71 838
weighted avg 0.73 0.73 0.73 838
[[413 108]
[117 200]]
pred: (array([6., 8.]), array([530, 308]))
test: (array([6., 8.]), array([521, 317]))
train: (array([6., 8.]), array([1187, 767]))
MSP
Accuracy: 0.7244680851063829
precision recall f1-score support
N 0.79 0.85 0.82 693
S 0.47 0.36 0.41 247
accuracy 0.72 940
macro avg 0.63 0.61 0.62 940
weighted avg 0.71 0.72 0.71 940
[[591 102]
[157 90]]
pred: (array(['N', 'S'], dtype='<U1'), array([748, 192]))
test: (array(['N', 'S'], dtype='<U1'), array([693, 247]))
train: (array(['N', 'S'], dtype='<U1'), array([1620, 572]))