In [1]:

    

from tensorflow.examples.tutorials.mnist import input_data


    

In [2]:

    

mnist = input_data.read_data_sets("MNIST_data/",one_hot = True)


    

    




Extracting MNIST_data/train-images-idx3-ubyte.gz
Extracting MNIST_data/train-labels-idx1-ubyte.gz
Extracting MNIST_data/t10k-images-idx3-ubyte.gz
Extracting MNIST_data/t10k-labels-idx1-ubyte.gz

In [3]:

    

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


    

    




(55000, 784) (55000, 10)

In [4]:

    

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


    

    




(10000, 784) (10000, 10)

In [5]:

    

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


    

    




(5000, 784) (5000, 10)

In [6]:

    

mnist


    

    
Out[6]:





Datasets(train=<tensorflow.contrib.learn.python.learn.datasets.mnist.DataSet object at 0x7fdb1d2d46a0>, validation=<tensorflow.contrib.learn.python.learn.datasets.mnist.DataSet object at 0x7fdb1d2d44e0>, test=<tensorflow.contrib.learn.python.learn.datasets.mnist.DataSet object at 0x7fdb18a060b8>)

In [7]:

    

import tensorflow as tf


    

In [8]:

    

sess = tf.InteractiveSession()
x = tf.placeholder(tf.float32,[None,784])


    

In [9]:

    

W = tf.Variable(tf.zeros([784,10]))
b = tf.Variable(tf.zeros([10]))


    

In [10]:

    

y = tf.nn.softmax(tf.matmul(x,W)+b)


    

In [11]:

    

y_=tf.placeholder(tf.float32,[None,10])


    

In [12]:

    

cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_*tf.log(y),reduction_indices=[1]))


    

In [13]:

    

train_step = tf.train.GradientDescentOptimizer(0.5).minimize(cross_entropy)


    

In [14]:

    

tf.global_variables_initializer().run()


    

In [15]:

    

for i in range(1000):
    batch_xs,batch_ys = mnist.train.next_batch(100)
    train_step.run({x: batch_xs,y_:batch_ys})


    

In [16]:

    

correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(y_,1))


    

In [17]:

    

accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))


    

In [18]:

    

print(accuracy.eval({x:mnist.test.images,y_:mnist.test.labels}))


    

    




0.9206

In [ ]: