-
Notifications
You must be signed in to change notification settings - Fork 8
/
Copy pathh5_merger.py
187 lines (174 loc) · 9.11 KB
/
h5_merger.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
import h5py
import numpy as np
import os
import shutil
from PIL import Image
import argparse
from tqdm import tqdm
import yaml
from glob import glob
folder_config_path = './folder_config.yml'
datasets_folder = './datasets/'
def merge_h5_file(args, name, split):
# Check input
if not name in ['database', 'queries']:
raise NotImplementedError('Name must be database or queries')
if not split in ['train', 'val', 'test']:
raise NotImplementedError('Split must be train or val or test')
# Check h5 names
read_path = []
database_indexes_list = [*args.database_indexes]
queries_indexes_list = [*args.queries_indexes]
if name == 'database':
if args.database_name == 'satellite' or args.database_name == 'foxtechmapping': # must contain satellite_i_thermalmapping_n
for i in range(len(queries_indexes_list)):
read_path.append(os.path.join(
datasets_folder, f'{args.database_name}_{database_indexes_list[0]}_{args.queries_name}_{queries_indexes_list[i]}/{split}_database.h5'))
elif args.database_name == 'thermalmapping':
for i in range(len(database_indexes_list)):
read_path.append(os.path.join(
datasets_folder, f'{args.database_name}_{database_indexes_list[i]}_{args.queries_name}_{queries_indexes_list[i]}/{split}_database.h5'))
else:
raise NotImplementedError()
save_path = os.path.join(datasets_folder, args.database_name + '_' + str(args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes), f'{split}_database.h5')
else:
if args.database_name == 'satellite' or args.database_name == 'foxtechmapping': # must contain satellite_i_thermalmapping_n
for i in range(len(queries_indexes_list)):
read_path.append(os.path.join(
datasets_folder, f'{args.database_name}_{database_indexes_list[0]}_{args.queries_name}_{queries_indexes_list[i]}/{split}_queries.h5'))
elif args.database_name == 'thermalmapping':
for i in range(len(queries_indexes_list)):
read_path.append(os.path.join(
datasets_folder, f'{args.database_name}_{database_indexes_list[i]}_{args.queries_name}_{queries_indexes_list[i]}/{split}_queries.h5'))
else:
raise NotImplementedError()
save_path = os.path.join(datasets_folder, args.database_name + '_' + str(args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes), f'{split}_queries.h5')
if os.path.isfile(save_path):
os.remove(save_path)
# Write h5
with h5py.File(save_path, "a") as hf:
start = False
for read_path_single in read_path:
with h5py.File(read_path_single, "r") as hf_single:
t = h5py.string_dtype(encoding="utf-8")
for i in tqdm(range(len(hf_single["image_data"]))):
img_np = np.expand_dims(hf_single["image_data"][i], axis=0)
img_size = np.expand_dims(hf_single["image_size"][i], axis=0)
img_name = np.expand_dims(hf_single["image_name"][i], axis=0)
if not start:
if args.compress:
hf.create_dataset(
"image_data",
data=img_np,
chunks=(1, 512, 512, 3),
maxshape=(None, 512, 512, 3),
compression="lzf",
)
hf.create_dataset(
"image_size",
data=img_size,
chunks=True,
maxshape=(None, 2),
compression="lzf",
)
hf.create_dataset(
"image_name",
data=img_name,
chunks=True,
maxshape=(None, ),
compression="lzf",
dtype=t
)
else:
hf.create_dataset(
"image_data",
data=img_np,
chunks=(1, 512, 512, 3),
maxshape=(None, 512, 512, 3),
)
hf.create_dataset(
"image_size",
data=img_size,
chunks=True,
maxshape=(None, 2)
)
hf.create_dataset(
"image_name",
data=img_name,
chunks=True,
maxshape=(None, ),
dtype=t
)
start = True
else:
hf["image_data"].resize(
hf["image_data"].shape[0] + img_np.shape[0], axis=0
)
hf["image_data"][-img_np.shape[0]:] = img_np
hf["image_size"].resize(
hf["image_size"].shape[0] + img_size.shape[0], axis=0
)
hf["image_size"][-img_size.shape[0]:] = hf_single["image_size"][i]
hf["image_name"].resize(
hf["image_name"].shape[0] + img_name.shape[0], axis=0
)
hf["image_name"][-img_name.shape[0]:] = hf_single["image_name"][i]
print("hdf5 file size: %d bytes" % os.path.getsize(save_path))
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument(
"--database_name",
type=str,
choices=['satellite', 'sirmionemapping',
'thermalmapping', 'foxtechmapping'],
help="The name of database map you want to use"
)
parser.add_argument(
"--database_indexes",
type=str,
default="0",
help="The index of database flight you want to use. For satellite map, 0 is bing."
)
parser.add_argument(
"--queries_name",
type=str,
choices=['satellite', 'sirmionemapping',
'thermalmapping', 'foxtechmapping'],
help="The name of queries map you want to use"
)
parser.add_argument(
"--queries_indexes",
type=str,
default="012345",
help="The index of queries flight you want to use. For satellite map, it is forced to be 0"
)
parser.add_argument("--compress", action="store_true")
parser.add_argument("--region_num", type=int, default=2, choices=[1, 2, 3])
args = parser.parse_args()
if args.database_name == 'satellite' and len(args.database_indexes) > 1:
raise ValueError("When creating satellite-thermal dataset, you can only choose 1 satellite map")
elif len(args.database_indexes) < 1 or len(args.queries_indexes) < 1:
raise ValueError("Indexes must contain more than 1 index")
if os.path.isdir(os.path.join(datasets_folder, args.database_name + '_' + str(args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes))):
rmpaths = glob(os.path.join(datasets_folder, args.database_name + '_' + str(
args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes), '*'))
for rmpath in rmpaths:
os.remove(rmpath)
else:
os.mkdir(os.path.join(datasets_folder, args.database_name + '_' +
str(args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes)))
if args.region_num >= 1:
merge_h5_file(args, name='database', split='train')
merge_h5_file(args, name='queries', split='train')
if args.region_num >= 2:
merge_h5_file(args, name='database', split='val')
merge_h5_file(args, name='queries', split='val')
if args.region_num <= 2:
# Not enough test data. Use val as test
os.symlink(os.path.abspath(os.path.join(datasets_folder, args.database_name + '_' + str(args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes), 'val_database.h5')),
os.path.join(datasets_folder, args.database_name + '_' + str(args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes), 'test_database.h5'))
os.symlink(os.path.abspath(os.path.join(datasets_folder, args.database_name + '_' + str(args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes), 'val_queries.h5')),
os.path.join(datasets_folder, args.database_name + '_' + str(args.database_indexes) + '_' + args.queries_name + '_' + str(args.queries_indexes), 'test_queries.h5'))
else:
merge_h5_file(args, name='database', split='test')
merge_h5_file(args, name='queries', split='test')