forked from mims-harvard/Raincoat
-
Notifications
You must be signed in to change notification settings - Fork 1
/
data_model_configs.py
219 lines (185 loc) · 6.31 KB
/
data_model_configs.py
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
def get_dataset_class(dataset_name):
"""Return the algorithm class with the given name."""
if dataset_name not in globals():
raise NotImplementedError("Dataset not found: {}".format(dataset_name))
return globals()[dataset_name]
class HAR():
def __init__(self):
super(HAR, self)
self.scenarios = [("2", "11"), ("6", "23"),("7", "13"),("9", "18"),("12", "16"),\
("13", "19"), ("18", "21"), ("20", "6"),("23", "13"),("24", "12")]
# self.scenarios = [("24", "12")]
self.class_names = ['walk', 'upstairs', 'downstairs', 'sit', 'stand', 'lie']
self.sequence_len = 128
self.shuffle = True
self.drop_last = True
self.normalize = True
# model configs
self.input_channels = 9
self.kernel_size = 5
self.stride = 1
self.dropout = 0.5
self.num_classes = 6
self.fourier_modes = 64
self.out_dim = 256 #192
# CNN and RESNET features
self.mid_channels = 64
self.final_out_channels = 128
self.features_len = 1
# TCN features
self.tcn_layers = [75, 150]
self.tcn_final_out_channles = self.tcn_layers[-1]
self.tcn_kernel_size = 17
self.tcn_dropout = 0.0
# lstm features
self.lstm_hid = 128
self.lstm_n_layers = 1
self.lstm_bid = False
# discriminator
self.disc_hid_dim = 64
self.hidden_dim = 500
self.DSKN_disc_hid = 128
class EEG():
def __init__(self):
super(EEG, self).__init__()
# data parameters
self.num_classes = 5
self.class_names = ['W', 'N1', 'N2', 'N3', 'REM']
self.sequence_len = 3000
self.scenarios = [("0", "11"), ("2", "5"), ("12", "5"), ("7", "18"), ("16", "1"), ("9", "14"),\
("4", "12"),("10", "7"),("6", "3"),("8", "10")]
self.shuffle = True
self.drop_last = True
self.normalize = True
# model configs
self.input_channels = 1
self.kernel_size = 25
self.stride = 6
self.dropout = 0.2
# features
self.mid_channels = 32
self.final_out_channels = 128
self.features_len = 1
self.fourier_modes = 300
self.out_dim = 256
# TCN features
self.tcn_layers = [32,64]
self.tcn_final_out_channles = self.tcn_layers[-1]
self.tcn_kernel_size = 15 # 25
self.tcn_dropout = 0.0
# lstm features
self.lstm_hid = 128
self.lstm_n_layers = 1
self.lstm_bid = False
# discriminator
self.DSKN_disc_hid = 128
self.hidden_dim = 500
self.disc_hid_dim = 100
class WISDM(object):
def __init__(self):
super(WISDM, self).__init__()
self.class_names = ['walk', 'jog', 'sit', 'stand', 'upstairs', 'downstairs']
self.sequence_len = 128
# Closed Set DA
'''self.scenarios = [("2", "32"), ("4", "15"),("7", "30"),('12','7'), ('12','19'),('18','20'),\
('20','30'), ("21", "31"),("25", "29"), ('26','2')]'''
self.scenarios = [("1", "0"), ("10", "11"), ("22", "17"), ("27", "15")]
self.num_classes = 6
self.shuffle = True
self.drop_last = False
self.normalize = True
# model configs
self.input_channels = 3
self.kernel_size = 5
self.stride = 1
self.dropout = 0.5
self.num_classes = 6
self.width = 64 # for FNN
self.fourier_modes = 64
# features
self.mid_channels = 64
self.final_out_channels = 128
self.out_dim = 192
self.features_len = 1
# TCN features
self.tcn_layers = [75,150,300]
self.tcn_final_out_channles = self.tcn_layers[-1]
self.tcn_kernel_size = 17
self.tcn_dropout = 0.0
# lstm features
self.lstm_hid = 128
self.lstm_n_layers = 1
self.lstm_bid = False
# discriminator
self.disc_hid_dim = 64
self.DSKN_disc_hid = 128
self.hidden_dim = 500
class HHAR_SA(object): ## HHAR dataset, SAMSUNG device.
def __init__(self):
super(HHAR_SA, self).__init__()
self.sequence_len = 128
# self.scenarios = [("0", "2")]
self.scenarios = [("0", "2"), ("1", "6"),("2", "4"),("4", "0"),("4", "5"),\
("5", "1"),("5", "2"),("7", "2"),("7", "5"),("8", "4")]
self.class_names = ['bike', 'sit', 'stand', 'walk', 'stairs_up', 'stairs_down']
self.num_classes = 6
self.shuffle = True
self.drop_last = True
self.normalize = True
self.fourier_modes = 32
# model configs
self.input_channels = 3
self.kernel_size = 5
self.stride = 1
self.dropout = 0.5
# features
self.mid_channels = 64 * 2
self.final_out_channels = 128
self.features_len = 1
self.out_dim = 128
# TCN features
self.tcn_layers = [75,150]
self.tcn_final_out_channles = self.tcn_layers[-1]
self.tcn_kernel_size = 17
self.tcn_dropout = 0.0
# lstm features
self.lstm_hid = 128
self.lstm_n_layers = 1
self.lstm_bid = False
# discriminator
self.disc_hid_dim = 64
self.DSKN_disc_hid = 128
self.hidden_dim = 500
class Boiler(object):
def __init__(self):
super(Boiler, self).__init__()
self.class_names = ['0','1']
self.sequence_len = 6
self.scenarios = [("1", "2"),("1", "3"),("2", "3")]
self.num_classes = 2
self.sequence_len = 6
self.shuffle = True
self.drop_last = True
self.normalize = True
# model configs
self.input_channels = 20
self.kernel_size = 5
self.stride = 1
self.dropout = 0.2
# features
self.mid_channels = 32
self.final_out_channels = 64
self.features_len = 1
# TCN features
self.tcn_layers = [32,64]
self.tcn_final_out_channles = self.tcn_layers[-1]
self.tcn_kernel_size = 15# 25
self.tcn_dropout = 0.0
# lstm features
self.lstm_hid = 128
self.lstm_n_layers = 1
self.lstm_bid = False
# discriminator
self.DSKN_disc_hid = 128
self.hidden_dim = 500
self.disc_hid_dim = 100