forked from potterhsu/SVHNClassifier-PyTorch
-
Notifications
You must be signed in to change notification settings - Fork 0
/
dataset.py
60 lines (45 loc) · 1.82 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
56
57
58
59
60
import lmdb
import numpy as np
import torch.utils.data as data
from PIL import Image
from . import example_pb2
class Dataset(data.Dataset):
def __init__(self, path_to_lmdb_dir, transform):
self._path_to_lmdb_dir = path_to_lmdb_dir
self._reader = lmdb.open(path_to_lmdb_dir, lock=False)
with self._reader.begin() as txn:
self._length = txn.stat()['entries']
self._keys = self._keys = [key for key, _ in txn.cursor()]
self._transform = transform
def __len__(self):
return self._length
def __getitem__(self, index):
if isinstance(index, slice):
result = []
for i in range(index.start, index.stop):
path = self._keys[i]
with self._reader.begin() as txn:
value = txn.get(path)
example = example_pb2.Example()
example.ParseFromString(value)
image = np.frombuffer(example.image, dtype=np.uint8)
image = image.reshape([64, 64, 3])
image = Image.fromarray(image)
image = self._transform(image)
length = example.length
digits = example.digits
result.append((image, length, digits, path))
return result
else:
path = self._keys[index]
with self._reader.begin() as txn:
value = txn.get(path)
example = example_pb2.Example()
example.ParseFromString(value)
image = np.frombuffer(example.image, dtype=np.uint8)
image = image.reshape([64, 64, 3])
image = Image.fromarray(image)
image = self._transform(image)
length = example.length
digits = example.digits
return image, length, digits, path