-
Notifications
You must be signed in to change notification settings - Fork 1
/
DatasetGenerator_Imbalance.py
132 lines (88 loc) · 4.67 KB
/
DatasetGenerator_Imbalance.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
import os
import numpy as np
from PIL import Image
import torch
from torch.utils.data import Dataset
# --------------------------------------------------------------------------------
class DatasetGeneratorforTraining (Dataset):
#--------------------------------------------------------------------------------
def __init__ (self, pathImageDirectory, pathDatasetFile, transform):
self.listImagePaths = []
self.listImageLabels = []
self.transform = transform
#---- Open file, get image paths and labels
fileDescriptor = open(pathDatasetFile, "r")
#---- get into the loop
line = True
while line:
line = fileDescriptor.readline()
#--- if not empty
if line:
lineItems = line.split()
imagePath = os.path.join(pathImageDirectory, lineItems[0])
imageLabel = lineItems[1:]
imageLabel = [int(i) for i in imageLabel]
if lineItems[0].split('/')[0] == 'chest14':
self.listImagePaths.append(imagePath)
self.listImageLabels.append(imageLabel)
elif lineItems[0].split('/')[0] == 'MON' and str(imageLabel) == '[0]':
for i in range(1384):
self.listImagePaths.append(imagePath)
self.listImageLabels.append(imageLabel)
elif lineItems[0].split('/')[0] == 'China' and str(imageLabel) == '[0]':
for i in range(343):
self.listImagePaths.append(imagePath)
self.listImageLabels.append(imageLabel)
elif lineItems[0].split('/')[0] == 'MON' and str(imageLabel) == '[1]':
for i in range(2800):
self.listImagePaths.append(imagePath)
self.listImageLabels.append(imageLabel)
elif lineItems[0].split('/')[0] == 'China' and str(imageLabel) == '[1]':
for i in range(500):
self.listImagePaths.append(imagePath)
self.listImageLabels.append(imageLabel)
else: pass
fileDescriptor.close()
#--------------------------------------------------------------------------------
def __getitem__(self, index):
imagePath = self.listImagePaths[index]
imageData = Image.open(imagePath).convert('RGB')
imageLabel= torch.FloatTensor(self.listImageLabels[index])
if self.transform != None: imageData = self.transform(imageData)
return imageData, imageLabel
#--------------------------------------------------------------------------------
def __len__(self):
return len(self.listImagePaths)
#--------------------------------------------------------------------------------
class TestDatasetGenerator (Dataset):
#--------------------------------------------------------------------------------
def __init__ (self, pathImageDirectory, pathDatasetFile, transform):
self.listImagePaths = []
self.listImageLabels = []
self.transform = transform
#---- Open file, get image paths and labels
fileDescriptor = open(pathDatasetFile, "r")
#---- get into the loop
line = True
while line:
line = fileDescriptor.readline()
#--- if not empty
if line:
lineItems = line.split()
imagePath = os.path.join(pathImageDirectory, lineItems[0])
imageLabel = lineItems[1:]
imageLabel = [int(i) for i in imageLabel]
self.listImagePaths.append(imagePath)
self.listImageLabels.append(imageLabel)
fileDescriptor.close()
#--------------------------------------------------------------------------------
def __getitem__(self, index):
imagePath = self.listImagePaths[index]
imageData = Image.open(imagePath).convert('RGB')
imageLabel= torch.FloatTensor(self.listImageLabels[index])
if self.transform != None: imageData = self.transform(imageData)
return imageData, imageLabel
#--------------------------------------------------------------------------------
def __len__(self):
return len(self.listImagePaths)
#--------------------------------------------------------------------------------