mnist數(shù)據(jù)集獲取:可從Yann LeCun教授管網(wǎng)獲取;
tensorflow中可使用input_data.read_data_sets("/worker/mnistdata/", one_hot = True) 導(dǎo)入下載到本地的mnist數(shù)據(jù)集; "/worker/mnistdata/"為數(shù)據(jù)集存放的位置.

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data  #tensorflow已經(jīng)包含了mnist案例的數(shù)據(jù)
mnist = input_data.read_data_sets("/worker/mnistdata/", one_hot = True) #導(dǎo)入已經(jīng)下載好的數(shù)據(jù)集,"/worker/mnistdata/"為存放mnist數(shù)據(jù)集的文件夾

print(mnist.train.images.shape, mnist.train.labels.shape)
print(mnist.test.images.shape, mnist.test.labels.shape)
print(mnist.validation.images.shape, mnist.validation.labels.shape)

Extracting /worker/mnistdata/train-images-idx3-ubyte.gz
Extracting /worker/mnistdata/train-labels-idx1-ubyte.gz
Extracting /worker/mnistdata/t10k-images-idx3-ubyte.gz
Extracting /worker/mnistdata/t10k-labels-idx1-ubyte.gz
(55000, 784) (55000, 10)
(10000, 784) (10000, 10)
(5000, 784) (5000, 10)