-
Notifications
You must be signed in to change notification settings - Fork 1
/
Copy pathdata_loader_1d.py
198 lines (142 loc) · 6 KB
/
data_loader_1d.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
from torchvision import datasets, transforms
import torch
import numpy as np
from scipy.fftpack import fft
import scipy.io as scio
import math
def wgn(x, snr):
Ps = np.sum(abs(x)**2,axis=1)/len(x)
Pn = Ps/(10**((snr/10)))
row,columns=x.shape
Pn = np.repeat(Pn.reshape(-1,1),columns, axis=1)
noise = np.random.randn(row,columns) * np.sqrt(Pn)
signal_add_noise = x + noise
return signal_add_noise
def zscore(Z):
Zmax, Zmin = Z.max(axis=1), Z.min(axis=1)
Z = (Z - Zmin.reshape(-1,1)) / (Zmax.reshape(-1,1) - Zmin.reshape(-1,1))
return Z
def min_max(Z):
Zmin = Z.min(axis=1)
Z = np.log(Z - Zmin.reshape(-1, 1) + 1)
return Z
def load_training(notargetlist,root_path,dir1,dir2,dir3,src_list, fft1, class_num , batch_size, kwargs):
src_list=torch.from_numpy(src_list)
data = scio.loadmat(root_path)
if fft1==True:
train_fea_1=zscore((min_max(abs(fft(data[dir1]))[:, 0:512])))
if fft1==False:
train_fea_1 = zscore(data[dir1])
long = len(notargetlist)
if long != 0:
b = np.linspace(0, 799, 800).reshape(-1, 1)
removelist = []
for i in range(long):
removelist = np.append(removelist, (notargetlist[i] - 1) * 800 + b)
train_fea_1 = np.delete(train_fea_1, removelist, axis=0)
train_label_1 = torch.zeros((800 * class_num,2))
for i in range(800 * class_num):
train_label_1[i][0] = i // 800
train_label_1[i][1] =0
data = scio.loadmat(root_path)
if fft1 == True:
train_fea_2= zscore((min_max(abs(fft(data[dir2]))[:, 0:512])))
if fft1 == False:
train_fea_2= zscore(data[dir2])
long = len(notargetlist)
if long != 0:
b = np.linspace(0, 799, 800).reshape(-1, 1)
removelist = []
for i in range(long):
removelist = np.append(removelist, (notargetlist[i] - 1) * 800 + b)
train_fea_2 = np.delete(train_fea_2, removelist, axis=0)
train_label_2= torch.zeros((800 * class_num, 2))
for i in range(800 * class_num):
train_label_2[i][0] = i // 800
train_label_2[i][1] = 1
data = scio.loadmat(root_path)
if fft1 == True:
train_fea_3= zscore((min_max(abs(fft(data[dir3]))[:, 0:512])))
if fft1 == False:
train_fea_3= zscore(data[dir3])
long = len(notargetlist)
if long != 0:
b = np.linspace(0, 799, 800).reshape(-1, 1)
removelist = []
for i in range(long):
removelist = np.append(removelist, (notargetlist[i] - 1) * 800 + b)
train_fea_3 = np.delete(train_fea_3, removelist, axis=0)
train_label_3= torch.zeros((800 * class_num, 2))
for i in range(800 * class_num):
train_label_3[i][0] = i // 800
train_label_3[i][1] =2
train_fea=np.vstack((train_fea_1,train_fea_2,train_fea_3))
train_label=torch.cat([train_label_1,train_label_2,train_label_3], dim=0)
train_label = train_label.long()
train_fea=torch.from_numpy(train_fea)
train_fea= torch.tensor(train_fea, dtype=torch.float32)
data = torch.utils.data.TensorDataset(train_fea, train_label)
train_loader = torch.utils.data.DataLoader(data, batch_size=batch_size, shuffle=True, drop_last=True, **kwargs)
return train_loader
def load_testing(root_path, dir, fft1, class_num, batch_size, kwargs):
data = scio.loadmat(root_path)
if fft1 == True:
train_fea = zscore((min_max(abs(fft(data[dir]))[:, 0:512])))
if fft1 == False:
train_fea = zscore(data[dir])
train_label = torch.zeros((200 * class_num))
for i in range(200 * class_num):
train_label[i] = i // 200
print(train_fea.shape)
train_label = train_label.long()
train_fea = torch.from_numpy(train_fea)
train_fea = torch.tensor(train_fea, dtype=torch.float32)
data = torch.utils.data.TensorDataset(train_fea, train_label)
train_loader = torch.utils.data.DataLoader(data, batch_size=batch_size, shuffle=False, drop_last=False, **kwargs)
return train_loader
def load_testing_known(notargetlist, root_path, dir, fft1, class_num, batch_size, kwargs):
data = scio.loadmat(root_path)
if fft1 == True:
train_fea = zscore((min_max(abs(fft(data[dir]))[:, 0:512])))
if fft1 == False:
train_fea = zscore(data[dir])
train_label = torch.zeros((200 * class_num))
for i in range(200 * class_num):
train_label[i] = i // 200
#
long = len(notargetlist)
if long!=0:
b = np.linspace(0, 199, 200).reshape(-1, 1)
removelist = []
for i in range(long):
removelist = np.append(removelist, (notargetlist[i] - 1) * 200 + b)
train_fea = np.delete(train_fea, removelist, axis=0)
train_label = train_label.long()
train_fea = torch.from_numpy(train_fea)
train_fea = torch.tensor(train_fea, dtype=torch.float32)
data = torch.utils.data.TensorDataset(train_fea, train_label)
test_loader = torch.utils.data.DataLoader(data, batch_size=batch_size, shuffle=False, drop_last=False, **kwargs)
return test_loader
def load_testing_unknown(notargetlist, root_path, dir, fft1, class_num, batch_size, kwargs):
data = scio.loadmat(root_path)
if fft1 == True:
train_fea = zscore((min_max(abs(fft(data[dir]))[:, 0:512])))
if fft1 == False:
train_fea = zscore(data[dir])
train_label = torch.zeros((200 * class_num))
for i in range(200 * class_num):
train_label[i] = len(notargetlist)
#
long = len(notargetlist)
if long!=0:
b = np.linspace(0, 199, 200).reshape(-1, 1)
removelist = []
for i in range(long):
removelist = np.append(removelist, (notargetlist[i] - 1) * 200 + b)
train_fea = np.delete(train_fea, removelist, axis=0)
train_label = train_label.long()
train_fea = torch.from_numpy(train_fea)
train_fea = torch.tensor(train_fea, dtype=torch.float32)
data = torch.utils.data.TensorDataset(train_fea, train_label)
test_loader = torch.utils.data.DataLoader(data, batch_size=batch_size, shuffle=False, drop_last=False, **kwargs)
return test_loader