-
Notifications
You must be signed in to change notification settings - Fork 4
/
dataset_loader.py
54 lines (42 loc) · 1.47 KB
/
dataset_loader.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
import tensorflow as tf
import tensorflow_datasets as tfds
def preprocess_fn(features):
# [0 ~ 255] -> [0.0 ~ 1.0]
features['image'] = tf.image.convert_image_dtype(features['image'], dtype=tf.float32)
# [0.0 ~ 1.0] -> [-1.0 ~ 1.0]
features['image'] = (features['image'] - 0.5) * 2.0
return features
def get_mnist_by_name(batch_size, name='mnist'):
# will return [28, 28, 1] uint8 (0~255)
dataset = tfds.load(name=name, split=tfds.Split.ALL)
dataset = dataset.shuffle(70000 + 1)
dataset = dataset.map(lambda x: preprocess_fn(x))
dataset = dataset.batch(batch_size)
dataset = dataset.prefetch(tf.data.experimental.AUTOTUNE)
return dataset
def main():
# test mnist loaders
batch_size = 100
epochs = 1
name = 'mnist'
# name = 'fashion_mnist'
dataset = get_mnist_by_name(batch_size, name)
n_images = 0
with tf.Session() as sess:
for e in range(epochs):
iterator = dataset.make_one_shot_iterator()
next_elem = iterator.get_next()
while True:
try:
elem = sess.run(next_elem)
images = elem['image']
labels = elem['label']
n_images += images.shape[0]
# print()
except tf.errors.OutOfRangeError:
print('End of dataset')
break
print('{}'.format(n_images))
return
if __name__ == '__main__':
main()