TensorFlow教程06：MNIST的CNN实现——源码和运行结果

假定您已经安装好了TensorFlow，这里放了第二个MNIST实验的代码和参考结果，你可以直接运行验证。

源码

[python]
view plain
copy

print?

#!/usr/bin/python  
import tensorflow as tf  
import sys  
from tensorflow.examples.tutorials.mnist import input_data  
  
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)  
  
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')  
  
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)  
  
sess = tf.InteractiveSession()  
  
x = tf.placeholder("float", shape=[None, 784])  
y_ = tf.placeholder("float", shape=[None, 10])  
  
W = tf.Variable(tf.zeros([784,10]))  
b = tf.Variable(tf.zeros([10]))  
  
W_conv1 = weight_variable([5, 5, 1, 32])  
b_conv1 = bias_variable([32])  
  
x_image = tf.reshape(x, [-1, 28, 28, 1])  
  
h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)  
h_pool1 = max_pool_2x2(h_conv1)  
  
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)  
  
# Now image size is reduced to 7*7  
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)  
  
keep_prob = tf.placeholder("float")  
h_fc1_drop = tf.nn.dropout(h_fc1, keep_prob)  
  
W_fc2 = weight_variable([1024, 10])  
b_fc2 = bias_variable([10])  
y_conv=tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2) + b_fc2)  
  
  
cross_entropy = -tf.reduce_sum(y_*tf.log(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, "float"))  
sess.run(tf.initialize_all_variables())  
  
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 %.3f"%(i, train_accuracy)  
  train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})  
  
print "Training finished"  
  
print "test accuracy %.3f" % accuracy.eval(feed_dict={  
    x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0})  

#!/usr/bin/python
import tensorflow as tf
import sys
from tensorflow.examples.tutorials.mnist import input_data

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)

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')

mnist = input_data.read_data
b23d
_sets("MNIST_data/", one_hot=True)

sess = tf.InteractiveSession()

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

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

W_conv1 = weight_variable([5, 5, 1, 32])
b_conv1 = bias_variable([32])

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

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

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)

# Now image size is reduced to 7*7
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)

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

W_fc2 = weight_variable([1024, 10])
b_fc2 = bias_variable([10])
y_conv=tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2) + b_fc2)

cross_entropy = -tf.reduce_sum(y_*tf.log(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, "float"))
sess.run(tf.initialize_all_variables())

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 %.3f"%(i, train_accuracy)
train_step.run(feed_dict={x: batch[0], y_: batch[1], keep_prob: 0.5})

print "Training finished"

print "test accuracy %.3f" % accuracy.eval(feed_dict={
x: mnist.test.images, y_: mnist.test.labels, keep_prob: 1.0})

运行结果

[plain]
view plain
copy

print?

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  
step 0, training accuracy 0.140  
step 100, training accuracy 0.840  
step 200, training accuracy 0.900  
step 300, training accuracy 0.840  
step 400, training accuracy 0.980  
step 500, training accuracy 0.940  
...  
step 19900, training accuracy 1.000  
Training finished  
test accuracy 0.993  

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
step 0, training accuracy 0.140
step 100, training accuracy 0.840
step 200, training accuracy 0.900
step 300, training accuracy 0.840
step 400, training accuracy 0.980
step 500, training accuracy 0.940
...
step 19900, training accuracy 1.000
Training finished
test accuracy 0.993