-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathSmarthome_Loader_CV.py
221 lines (192 loc) · 6.61 KB
/
Smarthome_Loader_CV.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
220
221
import os
os.environ['KERAS_BACKEND'] = 'tensorflow'
from keras.utils import Sequence, to_categorical
import numpy as np
from random import sample, randint, shuffle
import glob
import cv2
import time
class DataLoader_video_train(Sequence):
def __init__(self, path1, version, batch_size = 4):
self.batch_size = batch_size
#
self.version = version
self.path = '/data/stars/user/rdai/smarthomes/'+self.version+'/'
self.files = [i.strip() for i in open(path1).readlines()]
#
self.stack_size = 64
#
self.num_classes = 19
self.stride = 2
def _name_to_int(self,name):
integer=0
if name=="Cutbread":
integer=1
elif name=="Drink.Frombottle":
integer=2
elif name=="Drink.Fromcan":
integer=3
elif name=="Drink.Fromcup":
integer=4
elif name=="Drink.Fromglass":
integer=5
elif name=="Eat.Attable":
integer=6
elif name=="Eat.Snack":
integer=7
elif name=="Enter":
integer=8
elif name=="Getup":
integer=9
elif name=="Leave":
integer=10
elif name=="Pour.Frombottle":
integer=11
elif name=="Pour.Fromcan":
integer=12
elif name=="Readbook":
integer=13
elif name=="Sitdown":
integer=14
elif name=="Takepills":
integer=15
elif name=="Uselaptop":
integer=16
elif name=="Usetablet":
integer=17
elif name=="Usetelephone":
integer=18
elif name=="Walk":
integer=19
return integer
def __len__(self):
return int(len(self.files) / self.batch_size)
def __getitem__(self, idx):
batch = self.files[idx * self.batch_size : (idx + 1) * self.batch_size]
batch = [os.path.splitext(i)[0] for i in batch]
x_train = [self._get_video(i) for i in batch]
x_train = np.array(x_train, np.float32)
#
x_train /= 127.5
x_train -= 1
#y_train = np.array([int(i[-3:]) for i in batch]) - 1
y_train = np.array([self._name_to_int(i.split('_')[0]) for i in batch]) - 1
y_train = to_categorical(y_train, num_classes = self.num_classes)
return x_train, y_train
def _get_video(self, vid_name):
images = glob.glob(self.path + vid_name + "/*")
images.sort()
files = []
if len(images) > (self.stack_size * self.stride):
start = randint(0, len(images) - self.stack_size * self.stride)
files.extend([images[i] for i in range(start, (start + self.stack_size * self.stride), self.stride)])
elif len(images) < self.stack_size:
files.extend(images)
while len(files) < self.stack_size:
files.extend(images)
files = files[:self.stack_size]
else:
start = randint(0, len(images) - self.stack_size)
files.extend([images[i] for i in range(start, (start + self.stack_size))])
files.sort()
arr = []
for i in files:
if os.path.isfile(i):
arr.append(cv2.resize(cv2.imread(i), (224, 224)))
else:
arr.append(arr[-1])
return arr
def on_epoch_end(self):
shuffle(self.files)
class DataLoader_video_test(Sequence):
def __init__(self, path1, version, batch_size = 4):
self.batch_size = batch_size
#
self.version = version
self.path = '/data/stars/user/rdai/smarthomes/'+self.version+'/'
self.files = [i.strip() for i in open(path1).readlines()]
#
self.stack_size = 64
#
self.num_classes = 19
self.stride = 2
def _name_to_int(self,name):
integer=0
if name=="Cutbread":
integer=1
elif name=="Drink.Frombottle":
integer=2
elif name=="Drink.Fromcan":
integer=3
elif name=="Drink.Fromcup":
integer=4
elif name=="Drink.Fromglass":
integer=5
elif name=="Eat.Attable":
integer=6
elif name=="Eat.Snack":
integer=7
elif name=="Enter":
integer=8
elif name=="Getup":
integer=9
elif name=="Leave":
integer=10
elif name=="Pour.Frombottle":
integer=11
elif name=="Pour.Fromcan":
integer=12
elif name=="Readbook":
integer=13
elif name=="Sitdown":
integer=14
elif name=="Takepills":
integer=15
elif name=="Uselaptop":
integer=16
elif name=="Usetablet":
integer=17
elif name=="Usetelephone":
integer=18
elif name=="Walk":
integer=19
return integer
def __len__(self):
return int(len(self.files) / self.batch_size)
def __getitem__(self, idx):
batch = self.files[idx * self.batch_size : (idx + 1) * self.batch_size]
batch = [os.path.splitext(i)[0] for i in batch]
x_train = [self._get_video(i) for i in batch]
x_train = np.array(x_train, np.float32)
#
x_train /= 127.5
x_train -= 1
#y_train = np.array([int(i[-3:]) for i in batch]) - 1
y_train = np.array([self._name_to_int(i.split('_')[0]) for i in batch]) - 1
y_train = to_categorical(y_train, num_classes = self.num_classes)
return x_train, y_train
def _get_video(self, vid_name):
images = glob.glob(self.path + vid_name + "/*")
images.sort()
files = []
if len(images) > (self.stack_size * self.stride):
start = randint(0, len(images) - self.stack_size * self.stride)
files.extend([images[i] for i in range(start, (start + self.stack_size * self.stride), self.stride)])
elif len(images) < self.stack_size:
files.extend(images)
while len(files) < self.stack_size:
files.extend(images)
files = files[:self.stack_size]
else:
start = randint(0, len(images) - self.stack_size)
files.extend([images[i] for i in range(start, (start + self.stack_size))])
files.sort()
arr = []
for i in files:
if os.path.isfile(i):
arr.append(cv2.resize(cv2.imread(i), (224, 224)))
else:
arr.append(arr[-1])
return arr
def on_epoch_end(self):
shuffle(self.files)