-
Notifications
You must be signed in to change notification settings - Fork 1
Expand file tree
/
Copy pathSVC_Kombination_IEMO_MSP
More file actions
103 lines (81 loc) · 4.12 KB
/
SVC_Kombination_IEMO_MSP
File metadata and controls
103 lines (81 loc) · 4.12 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
SVC trainiert auf IEMO, getestet auf MSP:
precision recall f1-score support
A 0.23 0.66 0.34 746
H 0.23 0.33 0.27 1261
N 0.63 0.31 0.41 2293
S 0.30 0.02 0.03 807
accuracy 0.32 5107
macro avg 0.35 0.33 0.26 5107
weighted avg 0.42 0.32 0.31 5107
[[495 203 42 6]
[684 419 156 2]
[787 783 700 23]
[186 402 206 13]]
pred: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([2152, 1807, 1104, 44]))
test: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([ 746, 1261, 2293, 807]))
train: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([1090, 1615, 1695, 1075]))
SVC trainiert auf MSP, getested auf IEMO:
precision recall f1-score support
A 0.20 0.00 0.00 1091
H 0.36 0.15 0.21 1623
N 0.34 0.96 0.50 1690
S 0.40 0.00 0.00 1072
accuracy 0.34 5476
macro avg 0.32 0.28 0.18 5476
weighted avg 0.33 0.34 0.22 5476
[[ 1 313 777 0]
[ 2 243 1376 2]
[ 2 69 1618 1]
[ 0 49 1021 2]]
pred: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([ 5, 674, 4792, 5]))
test: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([1091, 1623, 1690, 1072]))
train: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([ 746, 1260, 2289, 811]))
Beide Datensätze zusammengefasst:
precision recall f1-score support
A 0.51 0.18 0.27 550
H 0.42 0.36 0.39 885
N 0.47 0.76 0.58 1201
S 0.51 0.25 0.33 571
accuracy 0.46 3207
macro avg 0.48 0.39 0.39 3207
weighted avg 0.47 0.46 0.43 3207
[[ 99 226 211 14]
[ 65 321 460 39]
[ 23 179 918 81]
[ 8 40 383 140]]
pred: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([ 195, 766, 1972, 274]))
test: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([ 550, 885, 1201, 571]))
train: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([1307, 2023, 2820, 1332]))
______________________________________________________________________________________________
SVC trainiert auf IEMO, getestet auf MSP mit Begrenzung der Klassen auf 754 Datenpunkte:
precision recall f1-score support
A 0.47 0.69 0.56 754
H 0.25 0.40 0.31 754
N 0.46 0.39 0.42 754
S 0.36 0.02 0.04 754
accuracy 0.38 3016
macro avg 0.38 0.38 0.33 3016
weighted avg 0.38 0.38 0.33 3016
[[517 193 39 5]
[334 304 113 3]
[ 78 363 296 17]
[179 366 195 14]]
pred: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([1108, 1226, 643, 39]))
test: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([754, 754, 754, 754]))
train: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([1103, 1636, 1708, 1084]))
DT trainiert auf MSP mit Begrenzung der Klassen auf 754 Datenpunkte, getested auf IEMO:
precision recall f1-score support
A 0.33 0.00 0.01 1103
H 0.38 0.32 0.35 1636
N 0.39 0.51 0.45 1708
S 0.19 0.33 0.24 1084
accuracy 0.32 5531
macro avg 0.32 0.29 0.26 5531
weighted avg 0.34 0.32 0.29 5531
[[ 4 486 195 418]
[ 5 531 514 586]
[ 3 286 879 540]
[ 0 89 640 355]]
pred: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([ 12, 1392, 2228, 1899]))
test: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([1103, 1636, 1708, 1084]))
train: (array(['A', 'H', 'N', 'S'], dtype='<U1'), array([754, 754, 754, 754]))