Tensorflow 实现CNN

利用Tensorflow实现CNN，CNN的原理我就不再多说，直接上代码

#利用CNN实现一个mnist识别
import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data

#1到10的数字
mnist = input_data.read_data_sets('MNIST_data',one_hot = True)
#计算准确率
def compute_accuracy(v_xs,v_ys):
    #定义一个全局变量的准确值
    global prediction

    y_pre = sess.run(prediction,feed_dict = {xs:v_xs,keep_prob:1})
    correct_prediction = tf.equal(tf.argmax(y_pre,1),tf.argmax(v_ys,1))
    accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
    result = sess.run(accuracy,feed_dict = {xs:v_xs,ys:v_ys,keep_prob:1})

    return result
#定义一个权值的函数
def weight_variable(shape):
    #生成一个随机值
    initial = tf.truncated_normal(shape,stddev = 0.1)
    return tf.Variable(initial)
#定义一个偏置量
def bias_variable(shape):
    initial = tf.constant(0.1,shape = shape)
    return tf.Variable(initial)

def conv2d(x,W):
    #strides代表步长,在x方向和y方向都是1
    #conv2d这个函数的第一个输入为当前层的节点矩阵，即image，第二个参数提供了卷积层的权重，第三个参数为不同维度的步长
    return tf.nn.conv2d(x,W,strides = [1,1,1,1],padding ='SAME')

def max_pool_2x2(x):
    #池化层是为了进一步的减少参数量
    #ksize提供了过滤器的尺寸，stride提供了步长信息，padding提供了是否使用全0填充
    return tf.nn.max_pool(x,ksize=[1,2,2,1],strides=[1,2,2,1],padding = 'SAME')

#定义placeholder
xs = tf.placeholder(tf.float32,[None,784])#这里是28x28
ys = tf.placeholder(tf.float32,[None,10])
#keep_prob表示使用drop_out时保留的概率
keep_prob = tf.placeholder(tf.float32)

x_image = tf.reshape(xs,[-1,28,28,1])
#print(x_image,shape)

#第一个卷积层
W_conv1 = weight_variable([5,5,1,32])
b_conv1 = bias_variable([32])

#搭建,并进行非线性的处理

h_conv1 = tf.nn.relu(conv2d(x_image,W_conv1) + b_conv1)#size 28x28x32
h_pool1 = max_pool_2x2(h_conv1) #输出的维度为14x14x32


#第二个卷积层
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)#这里变为14x14x64
h_pool2 = max_pool_2x2(h_conv2)#变成7x7x64

#然后是全连接层
W_fc1 = weight_variable([7*7*64,1024])#隐藏层结点数为1024
b_fc1 = bias_variable([1024])
#将第二层pool后的数据变化为一维的数据
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)

#加一个dropout的处理
h_fc1_drop = tf.nn.dropout(h_fc1,keep_prob)

#第二层的全连接层
W_fc2 = weight_variable([1024,10])
b_fc2 = bias_variable([10])
#利用softmax计算预测值概率
prediction = tf.nn.softmax(tf.matmul(h_fc1,W_fc2) + b_fc2)


#计算输出误差
cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys*tf.log(prediction),
                                              reduction_indices = [1]))

train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)

sess = tf.Session()

sess.run(tf.global_variables_initializer())

for i in range(1000):
    batch_xs,batch_ys = mnist.train.next_batch(100)
    sess.run(train_step,feed_dict = {xs:batch_xs,ys:batch_ys,keep_prob:1})
    if i %50 == 0:
        print(compute_accuracy(mnist.test.images,mnist.test.labels))