In [1]:

    

from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets('MNIST_data', one_hot=True)


    

    




Successfully downloaded train-images-idx3-ubyte.gz 9912422 bytes.
Extracting MNIST_data/train-images-idx3-ubyte.gz
Successfully downloaded train-labels-idx1-ubyte.gz 28881 bytes.
Extracting MNIST_data/train-labels-idx1-ubyte.gz
Successfully downloaded t10k-images-idx3-ubyte.gz 1648877 bytes.
Extracting MNIST_data/t10k-images-idx3-ubyte.gz
Successfully downloaded t10k-labels-idx1-ubyte.gz 4542 bytes.
Extracting MNIST_data/t10k-labels-idx1-ubyte.gz

In [2]:

    

mnist


    

    
Out[2]:





Datasets(train=<tensorflow.contrib.learn.python.learn.datasets.mnist.DataSet object at 0x11b5cd908>, validation=<tensorflow.contrib.learn.python.learn.datasets.mnist.DataSet object at 0x120a63fd0>, test=<tensorflow.contrib.learn.python.learn.datasets.mnist.DataSet object at 0x120a692e8>)

In [3]:

    

import tensorflow as tf
sess = tf.InteractiveSession()


    

In [5]:

    

x = tf.placeholder(tf.float32, shape=[None, 784])
y_ = tf.placeholder(tf.float32, shape=[None, 10])


    

In [6]:

    

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


    

In [21]:

    

sess.run(tf.global_variables_initializer())


    

In [13]:

    

# [None, 784] x [784, 10] = [None, 10]
y = tf.matmul(x, W) + b


    

In [14]:

    

y


    

    
Out[14]:





<tf.Tensor 'add_2:0' shape=(?, 10) dtype=float32>

In [16]:

    

cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=y))


    

In [17]:

    

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


    

In [33]:

    

train_step


    

    
Out[33]:





<tf.Operation 'GradientDescent' type=NoOp>

In [22]:

    

for _ in range(1000):
    batch = mnist.train.next_batch(100)
    train_step.run(feed_dict={x: batch[0], y_: batch[1]})
    # sess.run(train_step, feed_dict={x: batch[0], y_: batch[1]})


    

In [26]:

    

y, tf.argmax(y, axis=0), tf.argmax(y, axis=1)


    

    
Out[26]:





(<tf.Tensor 'add_2:0' shape=(?, 10) dtype=float32>,
 <tf.Tensor 'ArgMax_2:0' shape=(10,) dtype=int64>,
 <tf.Tensor 'ArgMax_3:0' shape=(?,) dtype=int64>)

In [27]:

    

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


    

In [28]:

    

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


    

In [31]:

    

accuracy


    

    
Out[31]:





<tf.Tensor 'Mean_2:0' shape=() dtype=float32>

In [34]:

    

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


    

    




0.9175
0.9175

In [43]:

    

def weight_variable(shape):
    initial = tf.truncated_normal(shape, stddev=0.1)
    return tf.Variable(initial)

def bias_variable(shape):
    initial = tf.constant(0.1, shape=shape)
    return tf.Variable(initial)


    

In [46]:

    

a = tf.truncated_normal((3, 3), stddev=0.1)
print(a.eval())
b = tf.constant(0.1, shape=(3, 1))
print(b.eval())


    

    




[[-0.05010616 -0.04017609 -0.00569539]
 [ 0.13421093 -0.00210931 -0.08559476]
 [ 0.00119102 -0.03151929 -0.04067332]]
[[ 0.1]
 [ 0.1]
 [ 0.1]]

In [47]:

    

def conv2d(x, W):
    return tf.nn.conv2d(x, W, strides=[1, 1, 1, 1], padding='SAME')

def max_pool_2x2(x):
    return tf.nn.max_pool(x, ksize=[1, 2, 2, 1], strides=[1, 2, 2, 1], padding='SAME')


    

In [48]:

    

W_conv1 = weight_variable([5, 5, 1, 32])  # h, w, in_channel, out_channel
b_conv1 = bias_variable([32])


    

In [49]:

    

x


    

    
Out[49]:





<tf.Tensor 'Placeholder:0' shape=(?, 784) dtype=float32>

In [50]:

    

x_image = tf.reshape(x, [-1, 28, 28, 1])


    

In [52]:

    

x_image, W_conv1


    

    
Out[52]:





(<tf.Tensor 'Reshape_6:0' shape=(?, 28, 28, 1) dtype=float32>,
 <tf.Variable 'Variable_2:0' shape=(5, 5, 1, 32) dtype=float32_ref>)

In [53]:

    

h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)
h_pool1 = max_pool_2x2(h_conv1)


    

In [55]:

    

h_conv1, h_pool1


    

    
Out[55]:





(<tf.Tensor 'Relu:0' shape=(?, 28, 28, 32) dtype=float32>,
 <tf.Tensor 'MaxPool:0' shape=(?, 14, 14, 32) dtype=float32>)

In [56]:

    

W_conv2 = weight_variable([5, 5, 32, 64])
b_conv2 = bias_variable([64])

h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)
h_pool2 = max_pool_2x2(h_conv2)


    

In [57]:

    

h_conv2, h_pool2


    

    
Out[57]:





(<tf.Tensor 'Relu_1:0' shape=(?, 14, 14, 64) dtype=float32>,
 <tf.Tensor 'MaxPool_1:0' shape=(?, 7, 7, 64) dtype=float32>)

In [58]:

    

W_fc1 = weight_variable([7 * 7 * 64, 1024])
b_fc1 = bias_variable([1024])

h_pool2_flat = tf.reshape(h_pool2, [-1, 7 * 7 * 64])
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)


    

In [60]:

    

h_pool2_flat, h_fc1


    

    
Out[60]:





(<tf.Tensor 'Reshape_7:0' shape=(?, 3136) dtype=float32>,
 <tf.Tensor 'Relu_2:0' shape=(?, 1024) dtype=float32>)

In [61]:

    

keep_prob = tf.placeholder(tf.float32)
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)


    

In [63]:

    

h_fc1_drop


    

    
Out[63]:





<tf.Tensor 'dropout/mul:0' shape=(?, 1024) dtype=float32>

In [64]:

    

W_fc2 = weight_variable([1024, 10])
b_fc2 = bias_variable([10])

y_conv = tf.matmul(h_fc1_drop, W_fc2) + b_fc2


    

In [65]:

    

y_conv


    

    
Out[65]:





<tf.Tensor 'add_6:0' shape=(?, 10) dtype=float32>

In [66]:

    

cross_entropy = tf.reduce_mean(
    tf.nn.softmax_cross_entropy_with_logits(labels=y_, logits=y_conv))
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
correct_prediction = tf.equal(tf.argmax(y_conv, 1), tf.argmax(y_, 1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    for i in range(20000):
        batch = mnist.train.next_batch(50)
        if i % 100 == 0:
            train_accuracy = accuracy.eval(feed_dict={
                    x: batch[0], y_: batch[1], keep_prob: 1.0})
            print('step %d, training accuracy %g' % (i, train_accuracy))
        train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
    print('test_accuracy %g' % accuracy.eval(feed_dict={
                x: mnist.test_images, y_: mnist.test.labels, keep_prob: 1.0}))


    

    




step 0, training accuracy 0.08
step 100, training accuracy 0.96
step 200, training accuracy 0.94
step 300, training accuracy 0.92
step 400, training accuracy 0.88
step 500, training accuracy 0.88
step 600, training accuracy 0.96
step 700, training accuracy 0.88
step 800, training accuracy 1
step 900, training accuracy 0.98
step 1000, training accuracy 0.98






    




---------------------------------------------------------------------------
KeyboardInterrupt                         Traceback (most recent call last)
<ipython-input-66-512db485c63c> in <module>()
     13                     x: batch[0], y_: batch[1], keep_prob: 1.0})
     14             print('step %d, training accuracy %g' % (i, train_accuracy))
---> 15         train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})
     16     print('test_accuracy %g' % accuracy.eval(feed_dict={
     17                 x: mnist.test_images, y_: mnist.test.labels, keep_prob: 1.0}))

/Users/koichiro.mori/.pyenv/versions/anaconda3-4.2.0/lib/python3.5/site-packages/tensorflow/python/framework/ops.py in run(self, feed_dict, session)
   1742         none, the default session will be used.
   1743     """
-> 1744     _run_using_default_session(self, feed_dict, self.graph, session)
   1745 
   1746 

/Users/koichiro.mori/.pyenv/versions/anaconda3-4.2.0/lib/python3.5/site-packages/tensorflow/python/framework/ops.py in _run_using_default_session(operation, feed_dict, graph, session)
   4118                        "the operation's graph is different from the session's "
   4119                        "graph.")
-> 4120   session.run(operation, feed_dict)
   4121 
   4122 

/Users/koichiro.mori/.pyenv/versions/anaconda3-4.2.0/lib/python3.5/site-packages/tensorflow/python/client/session.py in run(self, fetches, feed_dict, options, run_metadata)
    893     try:
    894       result = self._run(None, fetches, feed_dict, options_ptr,
--> 895                          run_metadata_ptr)
    896       if run_metadata:
    897         proto_data = tf_session.TF_GetBuffer(run_metadata_ptr)

/Users/koichiro.mori/.pyenv/versions/anaconda3-4.2.0/lib/python3.5/site-packages/tensorflow/python/client/session.py in _run(self, handle, fetches, feed_dict, options, run_metadata)
   1122     if final_fetches or final_targets or (handle and feed_dict_tensor):
   1123       results = self._do_run(handle, final_targets, final_fetches,
-> 1124                              feed_dict_tensor, options, run_metadata)
   1125     else:
   1126       results = []

/Users/koichiro.mori/.pyenv/versions/anaconda3-4.2.0/lib/python3.5/site-packages/tensorflow/python/client/session.py in _do_run(self, handle, target_list, fetch_list, feed_dict, options, run_metadata)
   1319     if handle is None:
   1320       return self._do_call(_run_fn, self._session, feeds, fetches, targets,
-> 1321                            options, run_metadata)
   1322     else:
   1323       return self._do_call(_prun_fn, self._session, handle, feeds, fetches)

/Users/koichiro.mori/.pyenv/versions/anaconda3-4.2.0/lib/python3.5/site-packages/tensorflow/python/client/session.py in _do_call(self, fn, *args)
   1325   def _do_call(self, fn, *args):
   1326     try:
-> 1327       return fn(*args)
   1328     except errors.OpError as e:
   1329       message = compat.as_text(e.message)

/Users/koichiro.mori/.pyenv/versions/anaconda3-4.2.0/lib/python3.5/site-packages/tensorflow/python/client/session.py in _run_fn(session, feed_dict, fetch_list, target_list, options, run_metadata)
   1304           return tf_session.TF_Run(session, options,
   1305                                    feed_dict, fetch_list, target_list,
-> 1306                                    status, run_metadata)
   1307 
   1308     def _prun_fn(session, handle, feed_dict, fetch_list):

KeyboardInterrupt: 

In [ ]: