知识点:
代码:
# coding: utf-8
try:
import urllib.request
except ImportError:
raise ImportError('You should use Python 3.x')
import os.path
import gzip
import pickle
import os
import numpy as np
# 记录数据集位置
url_base = 'http://yann.lecun.com/exdb/mnist/'
key_file = {
'train_img': 'train-images-idx3-ubyte.gz',
'train_label': 'train-labels-idx1-ubyte.gz',
'test_img': 't10k-images-idx3-ubyte.gz',
'test_label': 't10k-labels-idx1-ubyte.gz'
}
# 将处理后的数据,保存到本地存储位置
dataset_dir = os.path.dirname(os.path.abspath(__file__))
save_file = dataset_dir + "/mnist.pkl"
train_num = 60000 # 训练集个数
test_num = 10000 # 测试集个数
img_dim = (1, 28, 28) # 每张图片的数据维度
img_size = 784 # 每张图片大小(28*28)
# 将单个具体的原始数据文件下载到本地
def _download(file_name):
file_path = dataset_dir + "/" + file_name
if os.path.exists(file_path):
return
print("Downloading " + file_name + " ... ")
urllib.request.urlretrieve(url_base + file_name, file_path)
print("Done")
# 下载4个数据文件
def download_mnist():
for v in key_file.values():
_download(v)
# 加载label数据,转化为Numpy数据格式
def _load_label(file_name):
file_path = dataset_dir + "/" + file_name
print("Converting " + file_name + " to NumPy Array ...")
with gzip.open(file_path, 'rb') as f:
labels = np.frombuffer(f.read(), np.uint8, offset=8)
print("Done")
return labels
# 加载图片数据,转化为numpy格式(n行,每行784个数字)
def _load_img(file_name):
file_path = dataset_dir + "/" + file_name
print("Converting " + file_name + " to NumPy Array ...")
with gzip.open(file_path, 'rb') as f:
data = np.frombuffer(f.read(), np.uint8, offset=16)
data = data.reshape(-1, img_size)
print("Done")
return data
# 将下载的数据文件转化为numpy格式
def _convert_numpy():
dataset = {}
dataset['train_img'] = _load_img(key_file['train_img'])
dataset['train_label'] = _load_label(key_file['train_label'])
dataset['test_img'] = _load_img(key_file['test_img'])
dataset['test_label'] = _load_label(key_file['test_label'])
return dataset
# 从目标地址下载数据文件,转化为numpy格式,并保存为pickle文件
def init_mnist():
download_mnist()
dataset = _convert_numpy()
print("Creating pickle file ...")
with open(save_file, 'wb') as f:
pickle.dump(dataset, f, -1)
print("Done!")
# 将标签数据,转化为one_hot格式(正确数字对应下标为1,其他为0)
# 例如,原标签为[2,3],转化后为:
# [[0,0,1,0,0,0,0,0,0,0,0],[0,0,0,1,0,0,0,0,0,0,0]]
def _change_one_hot_label(X):
T = np.zeros((X.size, 10))
for idx, row in enumerate(T):
row[X[idx]] = 1
return T
# 主函数,如果第一次调用,从目标地址下载数据文件并处理为numpy后存为本地pickle文件
# 下次调用直接读取pickle文件,并根据参数,处理为相应的格式
def load_mnist(normalize=True, flatten=True, one_hot_label=False):
"""读入MNIST数据集
Parameters
----------
normalize : 将图像的像素值正规化为0.0~1.0
one_hot_label :
one_hot_label为True的情况下,标签作为one-hot数组返回
one-hot数组是指[0,0,1,0,0,0,0,0,0,0]这样的数组
flatten : 是否将图像展开为一维数组
Returns
-------
(训练图像, 训练标签), (测试图像, 测试标签)
"""
if not os.path.exists(save_file):
init_mnist()
with open(save_file, 'rb') as f:
dataset = pickle.load(f)
if normalize:
for key in ('train_img', 'test_img'):
dataset[key] = dataset[key].astype(np.float32)
dataset[key] /= 255.0
if one_hot_label:
dataset['train_label'] = _change_one_hot_label(dataset['train_label'])
dataset['test_label'] = _change_one_hot_label(dataset['test_label'])
if not flatten:
for key in ('train_img', 'test_img'):
dataset[key] = dataset[key].reshape(-1, 1, 28, 28)
return (dataset['train_img'], dataset['train_label']), (dataset['test_img'], dataset['test_label'])
if __name__ == '__main__':
init_mnist()
运行结果,下载了训练和测试数据,并处理后存为.pkl格式
展示第一张训练图片:
# coding: utf-8
import numpy as np
from MNISTData import load_mnist #MNISTData是上一段代码的文件名
from PIL import Image
def img_show(img):
pil_img = Image.fromarray(np.uint8(img))
pil_img.show()
# 加载测试集合训练集
(x_train, t_train), (x_test, t_test) = load_mnist(flatten=True, normalize=False)
img = x_train[0] # 第一张图片
label = t_train[0] # 第一张图片的标签
print(label) # 5
#
print(img.shape) # (784,)
img = img.reshape(28, 28) # 把图像的形状变为原来的尺寸
print(img.shape) # (28, 28)
#
img_show(img)
运行结果:
5
(784,)
(28, 28)
图片展示:
声明:以上代码主要来自[日]斋藤康毅 所著《深度学习入门-基于Python的理论与实现》