-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdata_prepare.py
226 lines (178 loc) · 7.45 KB
/
data_prepare.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
222
223
224
225
226
import torch
import numpy as np
import random
from PIL import Image
from torch.utils.data import Dataset
import os
import os.path
import cv2
import torchvision
def make_dataset(image_list, labels):
if labels:
len_ = len(image_list)
images = [(image_list[i].strip(), labels[i, :]) for i in range(len_)]
else:
if len(image_list[0].split()) > 2:
images = [(val.split()[0], np.array([int(la) for la in val.split()[1:]])) for val in image_list]
else:
images = [(val.split()[0], int(val.split()[1])) for val in image_list]
return images
def rgb_loader(path):
with open(path, 'rb') as f:
with Image.open(f) as img:
return img.convert('RGB')
def l_loader(path):
with open(path, 'rb') as f:
with Image.open(f) as img:
return img.convert('L')
class ImageList(Dataset):
def __init__(self, image_list, labels=None, transform=None, target_transform=None, mode='RGB'):
imgs = make_dataset(image_list, labels)
if len(imgs) == 0:
raise(RuntimeError("Found 0 images in subfolders of: " + root + "\n"
"Supported image extensions are: " + ",".join(IMG_EXTENSIONS)))
self.imgs = imgs
self.transform = transform
self.target_transform = target_transform
if mode == 'RGB':
self.loader = rgb_loader
elif mode == 'L':
self.loader = l_loader
def __getitem__(self, index):
path, target = self.imgs[index]
img = self.loader(path)
if self.transform is not None:
img = self.transform(img)
if self.target_transform is not None:
target = self.target_transform(target)
return img, target
def __len__(self):
return len(self.imgs)
class ImageList_idx(Dataset):
def __init__(self, image_list, labels=None, transform=None, target_transform=None, mode='RGB'):
imgs = make_dataset(image_list, labels)
if len(imgs) == 0:
raise(RuntimeError("Found 0 images in subfolders of: " + root + "\n"
"Supported image extensions are: " + ",".join(IMG_EXTENSIONS)))
self.imgs = imgs
self.transform = transform
self.target_transform = target_transform
if mode == 'RGB':
self.loader = rgb_loader
elif mode == 'L':
self.loader = l_loader
def __getitem__(self, index):
path, target = self.imgs[index]
img = self.loader(path)
if self.transform is not None:
img = self.transform(img)
if self.target_transform is not None:
target = self.target_transform(target)
return img, target, index
def __len__(self):
return len(self.imgs)
def data_load(args):
## prepare data
dsets = {}
dset_loaders = {}
train_bs = args.batch_size
txt_tar = open(args.t_dset_path).readlines()
txt_test = open(args.test_dset_path).readlines()
if not args.da == 'uda':
label_map_s = {}
for i in range(len(args.src_classes)):
label_map_s[args.src_classes[i]] = i
new_tar = []
for i in range(len(txt_tar)):
rec = txt_tar[i]
reci = rec.strip().split(' ')
if int(reci[1]) in args.tar_classes:
if int(reci[1]) in args.src_classes:
line = reci[0] + ' ' + str(label_map_s[int(reci[1])]) + '\n'
new_tar.append(line)
else:
line = reci[0] + ' ' + str(len(label_map_s)) + '\n'
new_tar.append(line)
txt_tar = new_tar.copy()
txt_test = txt_tar.copy()
dsets["target"] = ImageList_idx(txt_tar, transform=image_train())
dset_loaders["target"] = DataLoader(dsets["target"], batch_size=train_bs, shuffle=True, num_workers=args.worker, drop_last=False)
dsets["test"] = ImageList_idx(txt_test, transform=image_test())
dset_loaders["test"] = DataLoader(dsets["test"], batch_size=train_bs*3, shuffle=False, num_workers=args.worker, drop_last=False)
return dset_loaders
def image_train(resize_size=256, crop_size=224):
return transforms.Compose([
transforms.Resize((resize_size, resize_size)),
transforms.RandomCrop(crop_size),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
torchvision.transforms.Normalize([0.485, 0.456, 0.406],
[0.229, 0.224, 0.225])
])
def image_test(resize_size=256, crop_size=224, alexnet=False):
if not alexnet:
normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225])
else:
normalize = Normalize(meanfile='./ilsvrc_2012_mean.npy')
return transforms.Compose([
transforms.Resize((resize_size, resize_size)),
transforms.CenterCrop(crop_size),
transforms.ToTensor(),
normalize
])
class InputList(object):
def __init__(self, scales):
self.scales = scales
def __call__(self, img):
# assert img.size[0] == self.scales[0], 'image shape should be equal to max scale'
# input_list = []
# for i in range(len(self.scales)):
# input_list.append(F.resize(img, self.scales[i]))
assert img.size()[1] == self.scales[0], 'image shape should be equal to max scale'
input_list = []
img = img[np.newaxis, :]
for i in range(len(self.scales)):
resized_img = F.interpolate(img, (self.scales[i], self.scales[i]), mode='bilinear', align_corners=True)
resized_img = torch.squeeze(resized_img)
input_list.append(resized_img)
return input_list
def image_list(resize_size=256, crop_size=224):
return transforms.Compose([
transforms.Resize((resize_size, resize_size)),
transforms.RandomCrop(crop_size),
transforms.RandomHorizontalFlip(),
transforms.ToTensor(),
torchvision.transforms.Normalize([0.485, 0.456, 0.406],
[0.229, 0.224, 0.225]),
InputList([224, 192, 160, 128])
])
def data_load_list(args):
## prepare data
dsets = {}
dset_loaders = {}
train_bs = args.batch_size
txt_tar = open(args.t_dset_path).readlines()
txt_test = open(args.test_dset_path).readlines()
if not args.da == 'uda':
label_map_s = {}
for i in range(len(args.src_classes)):
label_map_s[args.src_classes[i]] = i
new_tar = []
for i in range(len(txt_tar)):
rec = txt_tar[i]
reci = rec.strip().split(' ')
if int(reci[1]) in args.tar_classes:
if int(reci[1]) in args.src_classes:
line = reci[0] + ' ' + str(label_map_s[int(reci[1])]) + '\n'
new_tar.append(line)
else:
line = reci[0] + ' ' + str(len(label_map_s)) + '\n'
new_tar.append(line)
txt_tar = new_tar.copy()
txt_test = txt_tar.copy()
dsets["target"] = ImageList_idx(txt_tar, transform=image_list())
dset_loaders["target"] = DataLoader(dsets["target"], batch_size=train_bs, shuffle=True, num_workers=args.worker, drop_last=False)
dsets["test"] = ImageList_idx(txt_test, transform=image_test())
dset_loaders["test"] = DataLoader(dsets["test"], batch_size=train_bs*3, shuffle=False, num_workers=args.worker, drop_last=False)
return dset_loaders