CNN卷积神经网络

import tensorflow as tf
from tensorflow.examples.tutorials.mnist import input_data
# number 1 to 10 data
mnist = input_data.read_data_sets('MNIST_data', one_hot=True)

# def add_layer(inputs, in_size, out_size, activation_function=None,):
#     # add one more layer and return the output of this layer
#     Weights = tf.Variable(tf.random_normal([in_size, out_size]))
#     biases = tf.Variable(tf.zeros([1, out_size]) + 0.1,)
#     Wx_plus_b = tf.matmul(inputs, Weights) + biases
#     if activation_function is None:
#         outputs = Wx_plus_b
#     else:
#         outputs = activation_function(Wx_plus_b,)
#     return outputs

def compute_accuracy(v_xs, v_ys):
    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

#权值函数定义
def weight_variable(shape): # 定义Weight变量，输入shape，返回变量的参数
    initial = tf.truncated_normal(shape,stddev=0.1) # 正态分布，标准差为0.1，默认最大为1，最小为-1，均值为0
    return tf.Variable(initial)
#偏置值函数
def bias_variable(shape):#定义Weight变量，输入shape，返回变量的参数
    initial = tf.constant(0.1,shape=shape)
    return tf.Variable(initial)
#卷积函数定义
def conv2d(x,W):
    return tf.nn.conv2d(x,W,strides=[1, 1, 1, 1],padding='SAME') # 卷积遍历各方向步数为1，SAME：边缘外自动补0，遍历相乘
#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')


# define placeholder for inputs to network
xs = tf.placeholder(tf.float32, [None, 784]) # 28x28
ys = tf.placeholder(tf.float32, [None, 10])
keep_prob = tf.placeholder(tf.float32)
x_image = tf.reshape(xs,[-1,28,28,1])   # x_image又把xs reshape成了28*28*1的形状，因为是灰色图片，所以通道是1.作为训练时的input，-1代表图片数量不定

#第一次卷积加池化
W_conv1 = weight_variable([5,5,1,32])#patch 5x5
b_conv1 = bias_variable([32])
h_conv1 = tf.nn.relu(conv2d(x_image,W_conv1)+b_conv1)
h_pool1 = max_pool_2x2(h_conv1)

#第二次卷积加池化
W_conv2 = weight_variable([5,5,32,64])#patch 5x5
b_conv2 = bias_variable([64])
h_conv2 = tf.nn.relu(conv2d(h_pool1,W_conv2)+b_conv2)
h_pool2 = max_pool_2x2(h_conv2)


#func1 layer
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)
h_fc1_drop = tf.nn.dropout(h_fc1,keep_prob)

#func2 layer
W_fc2 = weight_variable([1024,10])
b_fc2 = bias_variable([10])
prediction = tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2)+b_fc2)


# the error between prediction and real data
#二次代价函数:预测值与真实值的误差
cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys * tf.log(prediction),
                                              reduction_indices=[1]))       # loss

#梯度下降法:数据太庞大,选用AdamOptimizer优化器
train_step = tf.train.AdamOptimizer(0.0004).minimize(cross_entropy)

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

for i in range(1000):
    batch_xs, batch_ys = mnist.train.next_batch(100)  # 分批进行学习，时间短，效果不一定比整套的差
    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))