1.开发环境
-Win10
-Visual Studio 2015 64位
-OpenCV3.1
2.MNIST手写数据库
MNIST数据集分为以下四部分:
(1)
train-images-idx3-ubyte
训练图像的集合,共有60000张,大小是28×28
(2)
train-labels-idx1-ubyte
对应于训练图像的标签集,为0~9
(3)
t10k-images-idx3-ubyte
测试图像的集合,共有10000张,大小是28×28
(4)
t10k-labels-idx1-ubyte
对应于测试图像的标签集,为0~9
TRAINING SET LABEL FILE (train-labels-idx1-ubyte):
[offset] [type] [value] [description]
0000 32 bit integer 0x00000801(2049) magic number (MSB first)
0004 32 bit integer 60000 number of items
0008 unsigned byte ?? label
0009 unsigned byte ?? label
........
xxxx unsigned byte ?? label
The labe