#针对训练集
import numpy as np
import struct
from PIL import Image
import os
data_le = 'somePath/train-images.idx3-ubyte' # 需要修改的路径
# It's 47040016B, but we should set to 47040000B
data_le_size = 47040016
data_le_size = str(data_le_size - 16) + 'B'
data_buf = open(data_le, 'rb').read()
magic, numImages, numRows, numColumns = struct.unpack_from(
'>IIII', data_buf, 0)
datas = struct.unpack_from(
'>' + data_le_size, data_buf, struct.calcsize('>IIII'))
datas = np.array(datas).astype(np.uint8).reshape(
numImages, 1, numRows, numColumns)
label_le = 'somePath/train-labels.idx1-ubyte' # 需要修改的路径
# It's 60008B, but we should set to 60000B
label_le_size = 60008
label_le_size = str(label_le_size - 8) + 'B'
label_buf = open(label_le, 'rb').read()
magic, numLabels = struct.unpack_from('>II', label_buf, 0)
labels = struct.unpack_from(
'>' + label_le_size, label_buf, struct.calcsize('>II'))
labels = np.array(labels).astype(np.int64)
datas_root = '/somePath/mnist_train' # 需要修改的路径
if not os.path.exists(datas_root):
os.mkdir(datas_root)
for i in range(10):
le_name = datas_root + os.sep + str(i)
if not os.path.exists(le_name):
os.mkdir(le_name)