使用TensorFlow搭建模型之卷积神经网络详细解读(案例实战)

这个仅作为代码练手，读懂代码即可，熟悉使用TensorFlow网络搭建的流程。

代码演示：

from tensorflow.examples.tutorials.mnist import input_data
import tensorflow as tf

mnist = input_data.read_data_sets('mnistdata/',one_hot = True) #加载数据集
learning_rate = 1e-4 #学习率
epochs = 10000  #训练轮次
batch_size = 64 #每批训练的数量

x = tf.placeholder("float", shape=[None, 784]) #定义x，后面再喂入数据训练
y_ = tf.placeholder("float", shape=[None, 10])
#定义参数初始化函数
def weight_variable(shape):
    initial = tf.truncated_normal(shape,stddev=0.1)
    return tf.Variable(initial)
#定义偏置初始化函数
def bias_variable(shape):
    initial = tf.constant(0.1,shape=shape)
    return tf.Variable(initial)

#定义卷积函数，方便调用
def conv2d(x,W):
    return tf.nn.conv2d(x,W,strides=[1,1,1,1],padding='SAME')
#定义池化函数
def max_pool_2x2(x):
    return tf.nn.max_pool(x,ksize=[1,2,2,1],
                          strides=[1,2,2,1],padding='SAME')

x_image = tf.reshape(x,[-1,28,28,1])  #调整输入的形状，顺应网络输入结构
  
W_conv1 = weight_variable([5,5,1,32]) #相当于构建5*5的卷积核，数量为23
b_conv1 = bias_variable([32])

h_conv1 = tf.nn.relu(conv2d(x_image,W_conv1) + b_conv1) #卷积层运算，并使用relu激活函数
h_pool1 = max_pool_2x2(h_conv1) #池化层运算

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)

W_fc1 = weight_variable([7*7*64,1024]) #全连接层1，1024个神经元
b_fc1 = bias_variable([1024])

h_pool2_flat = tf.reshape(h_pool2,[-1,7*7*64]) #把feature_map平铺展开，这里即由(7,7,64)展开成(7*7*64,)
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat,W_fc1) + b_fc1) #全连接层1运算并使用relu激活函数

keep_prob = tf.placeholder('float')  #定义dropout比例
h_fc1_drop = tf.nn.dropout(h_fc1,keep_prob) #全连接层1进行dropout处理

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(learning_rate).minimize(cross_entropy) #Adam优化器
correct_prediction = tf.equal(tf.argmax(y_conv,1),tf.argmax(y_,1)) #统计预测正确的个数，返回的值是布尔型的
accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32)) #布尔型转为浮点，取平均即准确率

#创建会话
with tf.Session() as sess:
    sess.run(tf.initialize_all_variables())
    for i in range(epochs):
        batch = mnist.train.next_batch(batch_size)
        if i % 100 == 0:
            train_accuracy = accuracy.eval(feed_dict = {
                    x:batch[0],y_:batch[1],keep_prob:1.0})
            print('step {0:04d},training accuracy {1:.6f}'.format(i,train_accuracy))
        train_step.run(feed_dict={x:batch[0],y_:batch[1],keep_prob:0.5})
        
    print('test accuracy {0:.5f}'.format(accuracy.eval(feed_dict={
            x:mnist.test.images,y_:mnist.test.labels,keep_prob:1.0})))