-
Notifications
You must be signed in to change notification settings - Fork 0
/
train_save_model.yml
125 lines (119 loc) · 3.58 KB
/
train_save_model.yml
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
# The energy input measurement features
energy_params:
- Power1
- Power2
- Power3
- PowerReac1
- PowerReac2
- PowerReac3
- PowerApp1
- PowerApp2
- PowerApp3
# Segmentation window size and other properties as required
segment_window: 60
n_per_seg: 15
overlap_rate: 0.8
# Associating a class to files in the data folder
class_file_association:
on-ref:
- machine_ON_ref_no-error_1.csv
- machine_ON_ref_no-error_2.csv
- machine_ON_ref_no-error_3.csv
- machine_ON_ref_no-error_4.csv
- machine_ON_ref_no-error_5.csv
- machine_ON_ref_no-error_6.csv
- machine_ON_ref_no-error_7.csv
- machine_ON_ref_no-error_8.csv
- machine_ON_ref_no-error_9.csv
- machine_ON_ref_no-error_10.csv
- machine_ON_ref_no-error_11.csv
- machine_ON_ref_no-error_12.csv
- machine_ON_ref_no-error_13.csv
on-noref-error:
- machine_ON_no-ref_start-error_1.csv
- machine_ON_no-ref_start-error_2.csv
- machine_ON_no-ref_start-error_3.csv
- machine_ON_no-ref_start-error_4.csv
- machine_ON_no-ref_1.csv
- machine_ON_no-ref_2.csv
overtravel-x:
- machine_ON_ref_overtravel-error_x_neg_1.csv
- machine_ON_ref_overtravel-error_x_pos_1.csv
- machine_ON_no-ref_overtravel-error_x_neg_1.csv
- machine_ON_no-ref_overtravel-error_x_pos_1.csv
- machine_ON_ref_overtravel-error_x_neg_axes-extreme_1.csv
- machine_ON_ref_overtravel-error_x_neg_axes-extreme_2.csv
- machine_ON_ref_overtravel-error_x_pos_axes-extreme_1.csv
overtravel-y:
- machine_ON_ref_overtravel-error_y_neg_axes-extreme_1.csv
- machine_ON_ref_overtravel-error_y_neg_1.csv
- machine_ON_ref_overtravel-error_y_pos_1.csv
- machine_ON_ref_overtravel-error_y_pos_axes-extreme_1.csv
overtravel-z:
- machine_ON_ref_overtravel-error_z_neg_1.csv
- machine_ON_ref_overtravel-error_z_neg_axes-extreme_1.csv
- machine_ON_ref_overtravel-error_z_pos_1.csv
- machine_ON_ref_overtravel-error_z_pos_axes-extreme_1.csv
# Training flags
train_flags:
combine_overtravel: !!python/bool true
# Associating labels to a class
class_label_associations:
on-ref: 0
on-noref-error: 1
overtravel-x: 2
overtravel-y: 3
overtravel-z: 4
# The scripts to be trained
multi-class_models:
LogisticRegression:
max_iter: !!python/long 1000000000
multi_class: multinomial
n_jobs: 4
C: 0.1 # Needed for convergence
tol: !!python/float 0.1
class_weight: balanced
DecisionTreeClassifier:
min_samples_split: 100
class_weight: balanced
min_samples_leaf: 1
max_depth: null
KNeighborsClassifier:
n_neighbors: 5
SVC:
kernel: linear
tol: !!python/float 0.001
class_weight: balanced
BaggingClassifier:
n_estimators: 10
RandomForestClassifier:
class_weight: balanced
max_depth: 15
min_samples_leaf: 1
min_samples_split: 100
n_estimators: 100
# The anomaly detection scripts
anomaly_detection_models:
MahalanobisDistanceClassifier:
model_parameters:
threshold_level: 3
rowvar: !!python/bool false
PCA:
n_components: 140
svd_solver: full
KDEAnomalyDetector:
model_parameters:
quantile_threshold: !!python/float 0.02
kernel: gaussian
PCA:
n_components: 140
svd_solver: full
IsolationForestClassifier:
model_parameters:
n_estimators: 1000
bootstrap: !!python/bool true
contamination: !!python/float 0.05
PCA:
# Remove below if PCA is not required
n_components: 140
svd_solver: full