-
Notifications
You must be signed in to change notification settings - Fork 0
/
data.py
80 lines (52 loc) · 2.28 KB
/
data.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
import os
import numpy as np
from datasets import load_dataset
image_size = 512
save_location = 'datasets/'
def load_and_save_dataset():
# Get data from Hugging face
print('Downloading Dataset...')
master_dataset = load_dataset("rainerberger/Mri_segmentation")
train_dataset = master_dataset['train']
test_dataset = master_dataset['test']
train_total = len(train_dataset)
test_total = len(test_dataset)
# Split the data into Training data and Test data and save
# Processing training images
train_imgs = np.ndarray((train_total, image_size, image_size), dtype=np.uint8)
train_imgs_mask = np.ndarray((train_total, image_size, image_size), dtype=np.uint8)
for i in range(len(train_dataset)):
img = train_dataset[i]['image']
img_mask = train_dataset[i]['annotation']
train_imgs[i] = img
train_imgs_mask[i] = img_mask
if i%50 == 0:
print('Done: {0}/{1} images'.format(i, train_total))
print('Training images load complete.')
np.save(save_location + 'train_imgs.npy', train_imgs)
np.save(save_location + 'train_imgs_mask.npy', train_imgs_mask)
print('Saving to .npy files complete.')
# Processing testing images
test_imgs = np.ndarray((test_total, image_size, image_size), dtype=np.uint8)
test_imgs_mask = np.ndarray((test_total, ), dtype=np.int32)
for i in range(len(test_dataset)):
img = test_dataset[i]['image']
img_mask = test_dataset[i]['annotation']
test_imgs[i] = img
test_imgs_mask[i] = img_mask
if i%50 == 0:
print('Done: {0}/{1} images'.format(i, test_total))
print('Testing images load complete.')
np.save(save_location + 'test_imgs.npy', test_imgs)
np.save(save_location + 'test_imgs_mask.npy', test_imgs_mask)
print('Saving to .npy files complete.')
def load_train_data():
train_imgs = np.load(save_location + 'train_imgs.npy')
train_imgs_mask = np.load(save_location + 'train_imgs_mask.npy')
return train_imgs, train_imgs_mask
def load_test_data():
test_imgs = np.load(save_location + 'test_imgs.npy')
test_imgs_mask = np.load(save_location + 'test_imgs_mask.npy')
return test_imgs, test_imgs_mask
if __name__ == '__main__':
load_and_save_dataset()