-
Notifications
You must be signed in to change notification settings - Fork 15
/
diagnosis_test.py
185 lines (166 loc) · 8.57 KB
/
diagnosis_test.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
from scipy.io import loadmat
import numpy as np
import os
from keras.models import Sequential
from keras.layers import Dense
from keras.models import load_model
import preprocess_for_diagnosis
def capture(path):
file = loadmat(path)
return file
def slice(data):
x_test = data['x1']
x_test = x_test[:,0:896]
# for i in range(np.size(x_test,0)):
# 增加[-1,1]归一化处理
# x_test[i,:] = 2*(x_test[i,:]- min(x_test[i,:]))/(max(x_test[i,:])-min(x_test[i,:]))-1
# x_test[i,:] = (x_test[i,:]- np.mean(x_test[i,:]))/(max(x_test[i,:])-min(x_test[i,:]))
return x_test
def load_gdata(g_path, target):
filenames = os.listdir(g_path)
# gan模式只从小类(故障)文件中读取数据
filename = [i for i in filenames if (target in i) and ('.mat' in i)]
file_path = os.path.join(g_path, filename[0])
file = loadmat(file_path)
file_keys = file.keys()
for key in file_keys:
if 'DE' in key:
generated_data = file[key]
generated_data = generated_data[:,0:896]
generated_data = np.asarray(generated_data).reshape(-1,896)
return generated_data
model = load_model('diagnosis_CWRU_model/none-order-1-best.h5')
# 测试原始数据训练集的准确率
# imbalance_ratio = 100
# over_sampling = 'none' #'GAN', 'SMOTE', 'ADASYN','RANDOM', 'sampling_method'. 默认'none'
# path = r'data/0HP'
# length = 896 # 大于2周期
# normalization='minmax' #'minmax', 'mean'
# number = 1000
# generated = 'generated_data/ORDER_minmax_ratio10'
# rate = [0.5,0.25,0.25] # 测试集验证集划分比例
# sampling = 'order'
# num = int(number*rate[0])//imbalance_ratio
# list= ['B007','B014','B021','IR007','IR014','IR021','OR007','OR014','OR021']
# x_train, y_train, x_valid, y_valid, x_test, y_test = preprocess_for_diagnosis.prepro(d_path=path,
# gan_data=generated,
# length=length,
# number=number,
# normalization=normalization,
# rate=rate,
# sampling=sampling,
# over_sampling=over_sampling,
# imbalance_ratio=imbalance_ratio,
# )
# x_train = x_train[:,:,np.newaxis]
# start = 0
# for i in range(9):
# x_trainpiece = x_train[start:start+num,:,:]
# y_trainpiece = y_train[start:start+num,:]
# start += num
# score = model.evaluate(x=x_trainpiece, y=y_trainpiece, verbose=0)
# print("{} : {}".format(list[i],score[1]))
#测试某个.mat文件准确率
# path = r"samples\WGAN-GP\ORDER\ratio-50\orderIR021-05-17-20_22\signals\0061.mat"
# data = capture(path)
# x_train = slice(data)
# num = x_train.shape[0]
# y_train = np.ones((num,10), dtype=int)
# # y_train = [1,0,0,0,0,0,0,0,0,0]*y_train
# # y_train = [0,1,0,0,0,0,0,0,0,0]*y_train
# # y_train = [0,0,1,0,0,0,0,0,0,0]*y_train
# # y_train = [0,0,0,1,0,0,0,0,0,0]*y_train
# # y_train = [0,0,0,0,1,0,0,0,0,0]*y_train
# y_train = [0,0,0,0,0,1,0,0,0,0]*y_train
# # y_train = [0,0,0,0,0,0,1,0,0,0]*y_train
# # y_train = [0,0,0,0,0,0,0,1,0,0]*y_train
# # y_train = [0,0,0,0,0,0,0,0,1,0]*y_train
# score = model.evaluate(x=x_train, y=y_train, verbose=0)
# # print("测试集上的损失：", score[0])
# print(score[1])
# 展示9类GAN生成样本的分类准确率
y_label = [[1,0,0,0,0,0,0,0,0,0],
[0,1,0,0,0,0,0,0,0,0],
[0,0,1,0,0,0,0,0,0,0],
[0,0,0,1,0,0,0,0,0,0],
[0,0,0,0,1,0,0,0,0,0],
[0,0,0,0,0,1,0,0,0,0],
[0,0,0,0,0,0,1,0,0,0],
[0,0,0,0,0,0,0,1,0,0],
[0,0,0,0,0,0,0,0,1,0]]
list= ['B007','B014','B021','IR007','IR014','IR021','OR007','OR014','OR021']
path = 'generated_data/ORDER_minmax_ratio50'
for i in range(len(list)):
target = list[i]
x_train = load_gdata(g_path=path,target=target)
x_train = x_train[:,:,np.newaxis]
num = x_train.shape[0]
y_train = np.ones((num,10), dtype=int)
y_train = y_label[i]*y_train
# print('label', model.predict(x_train))
score = model.evaluate(x=x_train, y=y_train, verbose=0)
# print("测试集上的损失：", score[0])
print("{} : {}".format(target,score[1]))
#展示imlearn生成样本的分类准确率
# imbalance_ratio = 10
# over_sampling = 'RANDOM' #'GAN', 'SMOTE', 'ADASYN','RANDOM', 'sampling_method'. 默认'none'
# path = r'data/0HP'
# length = 896 # 大于2周期
# normalization='minmax' #'minmax', 'mean'
# number = 1000
# generated = 'generated_data/ORDER_minmax_ratio10'
# rate = [0.5,0.25,0.25] # 测试集验证集划分比例
# num = int(number*rate[0])
# minor_num = num//imbalance_ratio
# sampling = 'order'
# display_n = 100
# list= ['B007','B014','B021','IR007','IR014','IR021','OR007','OR014','OR021']
# x_train, y_train, x_valid, y_valid, x_test, y_test = preprocess_for_diagnosis.prepro(d_path=path,
# gan_data=generated,
# length=length,
# number=number,
# normalization=normalization,
# rate=rate,
# sampling=sampling,
# over_sampling=over_sampling,
# imbalance_ratio=imbalance_ratio)
# x_train = x_train[:,:,np.newaxis]
# start = 9*minor_num+num
# g_num = num-minor_num
# for i in range(9):
# x_trainpiece = x_train[start:start+g_num,:,:]
# y_trainpiece = y_train[start:start+g_num,:]
# start += g_num
# score = model.evaluate(x=x_trainpiece, y=y_trainpiece, verbose=0)
# print("{} : {}".format(list[i],score[1]))
#计算训练好的CNN的测试准确率
# model.load_model(r'diagnosis_CWRU_model\ADASYN\order-50-0.71-best.hdf5')
# classification = '10' #'binary' '10'
# if classification == 'binary':
# path = r'data/0HP Binary Classification' # 10分类
# else:
# path = r'data/0HP'
# generated = 'generated_data/ORDER_minmax_ratio10'
# length = 896
# number = 1000 # 每类样本的数量/大类样本的数量
# normalization='minmax' # 最大最小值归一化'minmax', 均值归一化'mean'
# rate = [0.5,0.25,0.25] # 测试集验证集划分比例
# sampling = 'order'
# over_sampling = 'none' #'GAN', 'SMOTE', 'ADASYN','RANDOM', 'sampling_method'. 默认'none'
# imbalance_ratio = 1
# batch_size = 256
# epochs = 600
# BatchNorm = True
# x_train, y_train, x_valid, y_valid, x_test, y_test = preprocess_for_diagnosis.prepro(d_path=path,
# gan_data=generated,
# length=length,
# number=number,
# normalization=normalization,
# rate=rate,
# sampling=sampling,
# over_sampling=over_sampling,
# imbalance_ratio=imbalance_ratio,
# )
# x_train, x_test = x_train[:,:,np.newaxis], x_test[:,:,np.newaxis]
# score = model.evaluate(x=x_test, y=y_test, verbose=0)
# print(score[0])