卷积实现手写数字识别

import tensorflow as tf
import random
import matplotlib.pyplot as plt
from tensorflow.examples.tutorials.mnist import input_data

tf.set_random_seed(777)

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

alpha = 0.001
training_epoch = 15
batch_size = 100

X = tf.placeholder(tf.float32, [None, 784])
Y = tf.placeholder(tf.float32, [None, 10])

第一层卷积

W1 = tf.get_variable(‘W1’, shape=[784, 512], initializer=tf.contrib.layers.xavier_initializer())
b1 = tf.Variable(tf.random_normal([512]))
a1 = tf.nn.relu(tf.matmul(X, W1)+b1)

第二层卷积

W2 = tf.get_variable(‘W2’, shape=[512, 512], initializer=tf.contrib.layers.xavier_initializer())
b2 = tf.Variable(tf.random_normal([512]))
a2 = tf.nn.relu(tf.matmul(a1, W2)+b2)

第三层卷积

W3 = tf.get_variable(‘W3’, shape=[512, 512], initializer=tf.contrib.layers.xavier_initializer())
b3 = tf.Variable(tf.random_normal([512]))
a3 = tf.nn.relu(tf.matmul(a2, W3)+b3)

第四层卷积

W4 = tf.get_variable(‘W4’, shape=[512, 512], initializer=tf.contrib.layers.xavier_initializer())
b4 = tf.Variable(tf.random_normal([512]))
a4 = tf.nn.relu(tf.matmul(a3, W4)+b4)

第五层卷积

W5 = tf.get_variable(‘W5’, shape=[512, 10], initializer=tf.contrib.layers.xavier_initializer())
b5 = tf.Variable(tf.random_normal([10]))
h = tf.matmul(a4, W5)+b5

accuracy = tf.reduce_mean(tf.cast(tf.equal(tf.argmax(h, 1), tf.argmax(Y, 1)), tf.float32))

cost = -tf.reduce_mean(tf.reduce_sum(Y * tf.log(h), axis=1))

损失函数调库实现

cost = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(logits=h, labels=Y))

op = tf.train.AdamOptimizer(learning_rate=alpha).minimize(cost)

sess = tf.Session()
sess.run(tf.global_variables_initializer())

for i in range(training_epoch):
avg_cost = 0
m = int(mnist.train.num_examples/batch_size)
for j in range(m):
xdata, ydata = mnist.train.next_batch(batch_size)
cost_var, _ = sess.run([cost, op], feed_dict={X:xdata, Y:ydata})
avg_cost += cost_var/m
print(avg_cost)

print(sess.run(accuracy, feed_dict={X:mnist.test.images, Y:mnist.test.labels}))

r = random.randint(0, mnist.test.num_examples-1)

print(‘label’, sess.run(tf.argmax(mnist.test.labels[r:r+1], 1)))
print(‘predict’, sess.run(tf.argmax(h, 1), feed_dict={X:mnist.test.images[r:r+1]}))

plt.imshow(mnist.test.images[r:r+1].reshape(28, 28), cmap=‘Greys’, interpolation=‘nearest’)
plt.show()