YOLO/data_loader.py at master · JoeSu666/YOLO · GitHub

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
# coding:utf-8
# 数据迭代
from mxnet import image
from mxnet import nd

data_shape = 256
batch_size = 32
rgb_mean = nd.array([123, 117, 104])
rgb_std = nd.array([58.395, 57.12, 57.375])

def get_iterators(data_shape, batch_size):
    class_names = ["background",'pikachu']
    num_class = len(class_names)
    train_iter = image.ImageDetIter(
        batch_size=batch_size,
        data_shape=(3, data_shape, data_shape),
        path_imgrec='train.rec',
        path_imgidx='train.idx',
        shuffle=True,
        mean=True,
        std=True,
        rand_crop=1,
        min_object_covered=0.95,
        max_attempts=200)
    val_iter = image.ImageDetIter(
        batch_size=batch_size,
        data_shape=(3, data_shape, data_shape),
        path_imgrec='val.rec',
        shuffle=False,
        mean=True,
        std=True)
    return train_iter, val_iter, class_names, num_class

train_data, test_data, class_names, num_class = get_iterators(
    data_shape, batch_size)