-
Notifications
You must be signed in to change notification settings - Fork 0
/
dataset.py
55 lines (43 loc) · 1.89 KB
/
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
import numpy as np
import torch, os, cv2
class BasicDataset(torch.utils.data.Dataset):
def __init__(self, folder, img_h, img_w, num_classes=1, transform=None):
self.imgs_dir = f"{folder}/imgs/"
self.masks_dir = f"{folder}/masks/"
self.img_h = img_h
self.img_w = img_w
self.transform = transform
self.num_classes = num_classes
self.ids = [os.path.splitext(file)[0] for file in os.listdir(self.imgs_dir) if not file.startswith(".")]
print(f"Creating dataset with {len(self.ids)} examples")
def __len__(self):
return len(self.ids)
@classmethod
def pre_process(self, img):
if len(img.shape) == 2:
img = np.expand_dims(img, axis=2)
img = img.transpose((2, 0, 1))
if img.max() > 1:
img = img / 255
return img
def __getitem__(self, i):
idx = self.ids[i]
mask_file = f"{self.masks_dir}{idx}.png"
img_file = f"{self.imgs_dir}{idx}.png"
img = np.array(cv2.cvtColor(cv2.imread(img_file, cv2.IMREAD_COLOR), cv2.COLOR_BGR2RGB))
mask = np.array(cv2.imread(mask_file, cv2.IMREAD_GRAYSCALE))
img = cv2.resize(img, (self.img_w, self.img_h), interpolation=cv2.INTER_LINEAR)
mask = cv2.resize(mask, (self.img_w, self.img_h), interpolation=cv2.INTER_NEAREST)
if self.transform is not None:
augmented = self.transform(**{"image": img, "mask": mask})
img, mask = augmented["image"], augmented["mask"]
# HWC to CHW
img = self.pre_process(img)
img_t = torch.from_numpy(img).type(torch.FloatTensor)
if self.num_classes > 1:
mask_t = torch.from_numpy(mask).type(torch.LongTensor)
else:
mask[mask != 0] = 1
mask = self.pre_process(mask)
mask_t = torch.from_numpy(mask).type(torch.FloatTensor)
return img_t, mask_t