import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data",one_hot=True)
# 每个批次的大小
batch_size = 100
# 计算一共有多少个批次
n_batch = mnist.train.num_examples // batch_size
# 初始化权值
def weight_variable(shape):
# 生成一个截断的正态分布
initial = tf.truncated_normal(shape, stddev=0.1)
return tf.Variable(initial)
# 初始化偏置
def bias_variable(shape):
initial = tf.constant(0.1, shape=shape)
return tf.Variable(initial)
# 卷积层
def conv2d(x,W):
# x input tensor of shape [batch, in_height, in_width, in_channels]
# W filter / kernel tensor of shape [filter_height, filter_width, in_channels, out_channels]
# in_channels代表输入通道数,out_channels代表输出方向数
# strides[0] = strides[1] = 0
# strides[1]代表x方向步长, strides[2]代表y方向步长
# Padding: "SAME'和"VALID'
# 'SAME'在外面补零
return tf.nn.conv2d(x, W, strides=[1,1,1,1], padding = 'SAME')
# 池化层
def max_pool_2x2(x):
# ksize [1,x,y,1]
return tf.nn.max_pool(x, ksize=[1,2,2,1],strides=[1,2,2,1],padding='SAME')
# 定义两个占位符
x = tf.placeholder(tf.float32,[None,28*28])
y = tf.placeholder(tf.float32,[None,10])
# 改变x的值为4D向量[batch_size,in_height,in_width,in_channels]
x_image = tf.reshape(x,[-1, 28, 28, 1])
# 初始化第一个卷积层的权值和偏置
W_conv1 = weight_variable([5,5,1,32]) # 5*5采样窗口，32个卷积核从1个平面提取数据
b_conv1 = bias_variable([32]) # 每个卷积核，一个偏置
# 把x_image和权值向量进行卷积，再加上偏置，然后应用于激活函数
h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)
# 进行池化操作
h_pool1 = max_pool_2x2(h_conv1)
# 初始化第二个卷积层的全值和偏置
W_conv2 = weight_variable([5,5,32,64]) # 5*5采样窗口，64个卷积核从32个平面抽取特征
b_conv2 = weight_variable([64]) # 每个卷积核，一个偏置
# 把h_pool1和权值向量进行卷积，再加上偏置，然后应用于激活函数
h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)
# 进行池化操作
h_pool2 = max_pool_2x2(h_conv2)
# 28*28的图片，第一次卷积还是28*28，第一池化变为14*14
# 第二次卷积还是14*14，第二次池化变为7*7
# 通过上面操作，得到64*7*7的平面
# 初始化第一个全连接层的权值
W_fcl = weight_variable([64*7*7,1024]) # 上一张有7*7*64的输入,1024个神经元
b_fcl = bias_variable([1024])
# 将池化后的图片扁平化为一维
h_pool2_flact = tf.reshape(h_pool2,[-1,7*7*64])
# 求第一个全连接层的输出
h_fcl = tf.nn.relu(tf.matmul(h_pool2_flact, W_fcl) + b_fcl)
# keep_prob 用来表示神经元的输出概率
keep_prob = tf.placeholder(tf.float32)
h_fcl_drop =tf.nn.dropout(h_fcl, keep_prob)
# 初始化第二个全连接层 1024个输入，10个输出
W_fc2 = weight_variable([1024,10])
b_fc2 = bias_variable([10])
# 计算输出
prediction = tf.nn.softmax(tf.matmul(h_fcl_drop,W_fc2) + b_fc2)
# 交叉熵代价函数
cross_entropy = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y, logits=prediction))
# 使用优化器进行优化
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
correct_prediction = tf.equal(tf.argmax(prediction,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 epoch in range(21):
for batch in range(n_batch):
batch_xs,batch_ys = mnist.train.next_batch(batch_size)
sess.run(train_step, feed_dict={x:batch_xs,y:batch_ys,keep_prob:0.7})
if batch%100 ==0:
print(str(batch)+"/"+str(n_batch))
acc = sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels,keep_prob:1.0})
print("Iter " + str(epoch) +" Test Accuracy " + str(acc))
