tensorflow实现简单的卷积网络

import tensorflow as tf
import gc

################导入input_data用于自动下载和安装MNIST数据集###########
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("F:\ZXY\python\MNIST_data/", one_hot = True)

############创建一个交互式Session#######
sess = tf.InteractiveSession()

def weight_variable(shape):
    initial = tf.truncated_normal(shape, stddev = 0.1)
    return tf.Variable(initial)

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

##############2维卷积函数################
def conv2d(x, W):
    return tf.nn.conv2d(x, W, strides = [1, 1, 1 ,1], padding = 'SAME')

##############最大池化###################
def max_pool_2x2(x):
    return tf.nn.max_pool(x, ksize = [1, 2, 2, 1], strides = [1, 2, 2, 1], padding = 'SAME')

#定义输入，并转化为28 * 28的图片
x = tf.placeholder(tf.float32, [None, 784])
y_ = tf.placeholder(tf.float32, [None, 10])
x_image = tf.reshape(x,[-1,28,28,1])#-1:样本数量不固定 28*28大小，单通道

##############第一个卷积层###############
W_conv1 = weight_variable([5,5,1,32])#w是5*5大小，单通道，32个卷积核
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2个卷积层###############
W_conv2 = weight_variable([5,5,32,64])#w是5*5大小，单通道，32个卷积核
b_conv2 = bias_variable([64])
h_conv2 = tf.nn.relu(conv2d(h_pool1, W_conv2) + b_conv2)
h_pool2 = max_pool_2x2(h_conv2)

#################全连接层###############
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)


#############dropout层#################
keep_prob = tf.placeholder(tf.float32)
h_fc1_drop = tf.nn.dropout(h_fc1,keep_prob)

##############第2层全连接层##################
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)


###############损失函数 + 优化器###########
cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_ * tf.log(y_conv),reduction_indices = [1])) #使用交叉熵验证输出和真实值的差别
train_step = tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)   #使用Adam优化损失函数

#############模型准确率###################
correct_prediction = tf.equal(tf.argmax(y_conv,1),tf.argmax(y_,1))
accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

####################训练过程#################
tf.global_variables_initializer().run()
for i in range(10000):
    batch = mnist.train.next_batch(50)
    if i%300 == 0:
        train_accuracy = accuracy.eval(feed_dict = {x:batch[0], y_:batch[1], keep_prob:1.0})
        print("step:%d, training accuracy %g"%(i, train_accuracy))
    train_step.run(feed_dict = {x:batch[0],y_:batch[1], keep_prob:0.5})
    
    

print("test accuracy %g " %accuracy.eval(feed_dict={x:mnist.test.images, y_:mnist.test.labels, keep_prob:1.0}))

结果：

Extracting F:\ZXY\python\MNIST_data/train-images-idx3-ubyte.gz
Extracting F:\ZXY\python\MNIST_data/train-labels-idx1-ubyte.gz
Extracting F:\ZXY\python\MNIST_data/t10k-images-idx3-ubyte.gz
Extracting F:\ZXY\python\MNIST_data/t10k-labels-idx1-ubyte.gz
step:0, training accuracy 0.02
step:300, training accuracy 0.88
step:600, training accuracy 0.98
step:900, training accuracy 0.96
step:1200, training accuracy 0.98
step:1500, training accuracy 0.98
step:1800, training accuracy 0.94
step:2100, training accuracy 0.96
step:2400, training accuracy 1
step:2700, training accuracy 1
step:3000, training accuracy 0.96
step:3300, training accuracy 0.98
step:3600, training accuracy 0.92
step:3900, training accuracy 0.98
step:4200, training accuracy 0.98
step:4500, training accuracy 0.98
step:4800, training accuracy 0.98
step:5100, training accuracy 0.98
step:5400, training accuracy 1
step:5700, training accuracy 1
step:6000, training accuracy 1
step:6300, training accuracy 1
step:6600, training accuracy 1
step:6900, training accuracy 0.98
step:7200, training accuracy 1
step:7500, training accuracy 1
step:7800, training accuracy 1
step:8100, training accuracy 1
step:8400, training accuracy 1
step:8700, training accuracy 1
step:9000, training accuracy 1
step:9300, training accuracy 1
step:9600, training accuracy 0.98
step:9900, training accuracy 1
test accuracy 0.9911