-
Notifications
You must be signed in to change notification settings - Fork 0
/
getDataset.py
162 lines (141 loc) · 6.33 KB
/
getDataset.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
import torch
import os, glob
import random, csv
from torch.utils.data import Dataset, DataLoader
from torchvision import transforms,datasets
from PIL import Image
import numpy
class MultipleApply:
"""Apply a list of transformations to an image and get multiple transformed images.
Args:
transforms (list or tuple): list of transformations
Example:
>>> transform1 = T.Compose([
... ResizeImage(256),
... T.RandomCrop(224)
... ])
>>> transform2 = T.Compose([
... ResizeImage(256),
... T.RandomCrop(224),
... ])
>>> multiply_transform = MultipleApply([transform1, transform2])
"""
def __init__(self, transforms):
self.transforms = transforms
def __call__(self, image):
return [t(image) for t in self.transforms]
class getDataset(Dataset):
def __init__(self, root, resize, mode,transform=None,s_t=None):
super(getDataset, self).__init__()
self.root = root
self.resize = resize
self.transform=transform
# print(self.transform)
self.name2label = {} # "sq...":0
for name in sorted(os.listdir(os.path.join(root))):
if not os.path.isdir(os.path.join(root, name)):
continue
self.name2label[name] = len(self.name2label.keys())
# image, label
self.images, self.labels = self.load_csv('images.csv')
#
#
source_image_path = []
source_image_label = []
target_image_label = []
target_image_path = []
if mode=='train': # 70%
self.images = self.images[:int(0.7 * len(self.images))]
self.labels = self.labels[:int(0.7 * len(self.labels))]
source_normal_counts=0
source_pneumonia_counts=0
if s_t=='source':
for i in range(len(self.images)):
if self.labels[i]<1 and source_normal_counts<5613:
source_image_path.append(self.images[i])
source_image_label.append(self.labels[i])
source_normal_counts+=1
if self.labels[i]==2 and source_pneumonia_counts<2306:
self.labels[i]=1
source_image_path.append(self.images[i])
source_image_label.append(self.labels[i])
source_pneumonia_counts+=1
self.images=source_image_path
self.labels=source_image_label
if s_t=='target':
target_normal_counts = 0
target_covid_counts = 0
for i in range(len(self.images)):
if self.labels[i] ==1 and target_covid_counts<258:
target_image_path.append(self.images[i])
target_image_label.append(self.labels[i])
target_covid_counts+=1
if self.labels[i] <1 and target_normal_counts<2541:
target_image_path.append(self.images[i])
target_image_label.append(self.labels[i])
target_normal_counts+=1
self.images = target_image_path
self.labels = target_image_label
if mode=='val':
self.images = self.images[int(0.7 * len(self.images)):int(0.85 * len(self.images))]
self.labels = self.labels[int(0.7 * len(self.labels)):int(0.85 * len(self.labels))]
for i in range(len(self.images)):
if self.labels[i]<2:
target_image_path.append(self.images[i])
target_image_label.append(self.labels[i])
self.images=target_image_path
self.labels=target_image_label
elif mode=='test':
self.images = self.images[int(0.85*len(self.images)):]
self.labels = self.labels[int(0.85*len(self.labels)):]
normal_counts=0
covid_counts=0
for i in range(len(self.images)):
if self.labels[i]<2:
if self.labels[i]==0 and normal_counts<885:
target_image_path.append(self.images[i])
target_image_label.append(self.labels[i])
normal_counts+=1
if self.labels[i]==1 and covid_counts<60:
target_image_path.append(self.images[i])
target_image_label.append(self.labels[i])
covid_counts+=1
self.images=target_image_path
self.labels=target_image_label
def load_csv(self, filename):
if not os.path.exists(os.path.join(self.root, filename)):
images = []
for name in self.name2label.keys():
images += glob.glob(os.path.join(self.root, name, '*.png'))
images += glob.glob(os.path.join(self.root, name, '*.jpg'))
images += glob.glob(os.path.join(self.root, name, '*.jpeg')) #加载路径下的所有.jpeg图片
print(len(images), images)
print(images)
random.shuffle(images)
with open(os.path.join(self.root, filename), mode='w', newline='') as f:
writer = csv.writer(f)
for img in images:
name = img.split(os.sep)[-2]
label = self.name2label[name]
writer.writerow([img, label])
print('writen into csv file:', filename)
# read from csv file
images, labels = [], []
with open(os.path.join(self.root, filename)) as f:
reader = csv.reader(f)
for row in reader:
img, label = row
label = int(label)
images.append(img)
labels.append(label)
assert len(images) == len(labels)
return images, labels
def __len__(self):
return len(self.images)
def __getitem__(self, idx):
img, label = self.images[idx], self.labels[idx]
img=Image.open(img).convert('RGB')
if self.transform is not None:
img=self.transform(img)
label = torch.tensor(label)
return img, label