TensorFlow入门——MNIST手写数字识别实战代码

我在该代码中使用了两个卷积层+池化层，最后连上两个全连接层。
第一层卷积使用32个5x5x1的卷积核，步长为1，边界处理方式为“SAME”（卷积的输入和输出保持相同尺寸），激发函数为Relu，然后接一个2x2的池化层，方式为最大化池化；
第二层卷积使用64个5x5x32的卷积核，步长为1，边界处理方式为“SAME”，激发函数为Relu， 后接一个2x2的池化层，方式为最大化池化；
第一层全连接层：使用1024个神经元，激发函数依然是Relu。
第二层全连接层：使用10个神经元，激发函数为softmax，用于输出结果

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

 #从网上下载mnist的4个安装包
mnist = input_data.read_data_sets('FLAGS.data_dir', one_hot=True)

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

def weight_variable(shape):
	initial = tf.truncated_normal(shape,stddev=0.1) # 产生随机变量     stddev标准差
    return tf.Variable(initial)

def bias_variable(shape):
    initial = tf.constant(0.1,shape=shape)
    return tf.Variable(initial)


#自定义卷积函数
def conv2d(x,W):
    # x 图片的所有信息     
    # stride [1,x_movement,y_movement]
    #Must have strides[0] = strides[3] = 1
    return tf.nn.conv2d(x,W,strides=[1,1,1,1],padding='SAME')

#自定义池化函数 
def max_pool_2x2(x):
    # stride [1,x_movement,y_movement]
    return tf.nn.max_pool(x,ksize=[1,2,2,1],strides=[1,2,2,1],padding='SAME')

#define placeholder for inputs to network
#设置占位符，尺寸为样本输入和输出的尺寸
xs = tf.placeholder(tf.float32,[None,784]) # 28*28
ys = tf.placeholder(tf.float32,[None,10])
keep_prob = tf.placeholder(tf.float32)
x_image = tf.reshape(xs,[-1,28,28,1]) 
#print(x_image.shape) #[n_samples,28,28,1]
#conv1 layer

#设置第一个卷积层和池化层
W_conv1 = weight_variable([5,5,1,32]) # patch 5*5， in size 1, out size 32
b_conv1 = bias_variable([32])    
h_conv1 = tf.nn.relu(conv2d(x_image,W_conv1) + b_conv1)    # output size 28*28*32
h_pool1 = max_pool_2x2(h_conv1)                         # output size 14*14*32
#conv2 layer

#设置第二个卷积层和池化层
W_conv2 = weight_variable([5,5,32,64]) # patch 5*5， in size 32, out size 64
b_conv2 = bias_variable([64])
h_conv2 = tf.nn.relu(conv2d(h_pool1,W_conv2) + b_conv2) # output size 14*14*64
h_pool2 = max_pool_2x2(h_conv2)                         # output size 7*7*64

#func1 layer
#设置第一个全连接层
W_fc1 = weight_variable([7*7*64,1024])
b_fc1 = bias_variable([1024])
# [n_samples,7,7,64] ->> [n_samples,7*7*64]
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
#设置第二个全连接层
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
#减小误差
cross_entropy = tf.reduce_mean(-tf.reduce_sum(ys*tf.log(prediction),reduction_indices=[1]))

#配置Adam优化器，学习速率为1e-4
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
sess = tf.Session()
#important step
sess.run(tf.initialize_all_variables())

for i in range(1000):
	#mnist.train.next_batch(100)是从训练集里一次提取100张图片数据来训练
    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:
        # 每50步输出一次准确率
        print(compute_accuracy(mnist.test.images,mnist.test.labels))