-
Notifications
You must be signed in to change notification settings - Fork 23
/
base_dataset.py
executable file
·173 lines (142 loc) · 6.29 KB
/
base_dataset.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
import random
import torch
import torch.utils.data as data
import torchvision.transforms as transforms
from PIL import Image
class BaseDataset(data.Dataset):
def __init__(self):
super(BaseDataset, self).__init__()
def name(self):
return 'BaseDataset'
def initialize(self, opt):
pass
def get_transform(opt):
transform_list = []
if opt.resize_or_crop == 'resize_and_crop':
osize = [opt.loadSize, opt.loadSize]
transform_list.append(transforms.Scale(osize, Image.BICUBIC))
transform_list.append(transforms.RandomCrop(opt.fineSize))
elif opt.resize_or_crop == 'crop':
transform_list.append(transforms.RandomCrop(opt.fineSize))
elif opt.resize_or_crop == 'scale_width':
transform_list.append(transforms.Lambda(
lambda img: __scale_width(img, opt.fineSize)))
elif opt.resize_or_crop == 'scale_width_and_crop':
transform_list.append(transforms.Lambda(
lambda img: __scale_width(img, opt.loadSize)))
transform_list.append(transforms.RandomCrop(opt.fineSize))
elif opt.resize_or_crop == 'none':
transform_list = []
if opt.isTrain and not opt.no_flip:
transform_list.append(transforms.RandomHorizontalFlip())
transform_list.append(transforms.RandomVerticalFlip())
transform_list += [transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5),
(0.5, 0.5, 0.5))]
return transforms.Compose(transform_list)
def transform_multi(opt, A, B, C, Bases, Shapes, Colors, blur_Shapes, blur_Colors):
"""Transformer for multi fusion, pretrain dataset
"""
if not opt.resize_or_crop == 'none':
raise ValueError(
"Only support none mode for resize_or_crop on base_gray_color dataset")
assert(isinstance(Bases, list))
assert(isinstance(Shapes, list))
assert(isinstance(Colors, list))
assert(isinstance(blur_Shapes, list))
assert(isinstance(blur_Colors, list))
A = transforms.ToTensor()(A)
B = transforms.ToTensor()(B)
C = transforms.ToTensor()(C)
A = transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(A)
B = transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(B)
C = transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(C)
Bases = list(map(lambda b: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(b)), Bases))
Shapes = list(map(lambda s: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(s)), Shapes))
Colors = list(map(lambda c: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(c)), Colors))
blur_Shapes = list(map(lambda bs: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(bs)), blur_Shapes))
blur_Colors = list(map(lambda bc: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(bc)), blur_Colors))
Bases = torch.cat(Bases)
Shapes = torch.cat(Shapes)
Colors = torch.cat(Colors)
blur_Shapes = torch.cat(blur_Shapes)
blur_Colors = torch.cat(blur_Colors)
return A, B, C, Bases, Shapes, Colors, blur_Shapes, blur_Colors
def transform_few_with_label(opt, A, B, C, label, Bases, Shapes, Colors, blur_Shapes, blur_Colors):
if not opt.resize_or_crop == 'none':
raise ValueError(
"Only support none mode for resize_or_crop on base_gray_color dataset")
assert(isinstance(Bases, list))
assert(isinstance(Shapes, list))
assert(isinstance(Colors, list))
assert(isinstance(blur_Shapes, list))
assert(isinstance(blur_Colors, list))
A = transforms.ToTensor()(A)
B = transforms.ToTensor()(B)
C = transforms.ToTensor()(C)
A = transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(A)
B = transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(B)
C = transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(C)
if label == 0.0:
C_l = torch.zeros_like(C)
else:
C_l = C
label = torch.tensor(label)
Bases = list(map(lambda b: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(b)), Bases))
Shapes = list(map(lambda s: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(s)), Shapes))
Colors = list(map(lambda c: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(c)), Colors))
blur_Shapes = list(map(lambda bs: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(bs)), blur_Shapes))
blur_Colors = list(map(lambda bc: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(bc)), blur_Colors))
rand_idx = random.randrange(opt.nencode)
if label == 0.0:
B_G = Shapes[rand_idx]
C_G = Colors[rand_idx]
else:
B_G = B
C_G = C
Bases = torch.cat(Bases)
Shapes = torch.cat(Shapes)
Colors = torch.cat(Colors)
blur_Shapes = torch.cat(blur_Shapes)
blur_Colors = torch.cat(blur_Colors)
return A, B, B_G, C, C_G, C_l, label, Bases, Shapes, Colors, blur_Shapes, blur_Colors
def __scale_width(img, target_width):
ow, oh = img.size
if (ow == target_width):
return img
w = target_width
h = int(target_width * oh / ow)
return img.resize((w, h), Image.BICUBIC)
def transform_grid(Shapes, Colors):
"""Transform Shpaes/Colors to grid
0 1 2 3 -> 0 1
2 3
Arguments:
Shapes {PIL Images} -- [4*W*H*C]
Colors {PIL Images} -- [4*W*H*C]
Returns:
[grid tensors] -- [C*2W*2H]
"""
assert(isinstance(Shapes, list))
assert(isinstance(Colors, list))
Shapes = list(map(lambda s: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(s)), Shapes))
shapes_row1 = torch.cat((Shapes[0], Shapes[1]), 1)
shapes_row2 = torch.cat((Shapes[2], Shapes[3]), 1)
Shapes_grid = torch.cat((shapes_row1, shapes_row2), 2)
Colors = list(map(lambda c: transforms.Normalize(
(0.5, 0.5, 0.5), (0.5, 0.5, 0.5))(transforms.ToTensor()(c)), Colors))
colors_row1 = torch.cat((Colors[0], Colors[1]), 1)
colors_row2 = torch.cat((Colors[2], Colors[3]), 1)
Colors_grid = torch.cat((colors_row1, colors_row2), 2)
return Shapes_grid, Colors_grid