Tensorflow用CNN实现Mnist手写数据集分类

本文内容只是方便自己下次学习，如有侵权，请联系我进行删除。内容主要是《Tensorflow实战》中第五章用CNN实现Mnist手写数据集分类

一、加载数据

from tensorflow.examples.tutorials.mnist import input_data  
import tensorflow as tf  
mnist = input_data.read_data_sets("MNSIT_data/", one_hot=True)  
sess = tf.InteractiveSession() 

二、初始化参数

def weight_variable(shape):#初始化权重，利用一些随机噪声打破完全对称
    initial=tf.truncated_normal(shape,stddev=0.1)#截断的正太分布，标准差是0.1
    return tf.Variable(initial)

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

三、定义网络层

def conv2d(x,w):#setting conv
    return tf.nn.conv2d(x,w,strides=[1,1,1,1],padding='SAME')
# tf.nn.conv2d 卷积函数  
    # 参数 value 输入图像             四维[图片数量, 图片高度, 图片宽度, 图像通道数]  
    # 参数 ksize 池化窗口             四维[卷积核的高度，卷积核的宽度，图像通道数，卷积核个数]  
    # 参数 strides 卷积核移动量     四维[图片数量, 图片高度, 图片宽度, 图像通道数]，一般不对图片数量和图像通道数进行池化，所以都是1  
    # 参数 padding 边缘处理方式     SAME和VALID,SAME就是可以在外围补0再卷积，VALID不能补0  
def max_pool_2x2(x):
    return tf.nn.max_pool(x,ksize=[1,2,2,1],strides=[1,2,2,1],padding='SAME')

x=tf.placeholder(tf.float32,[None,784])
y_=tf.placeholder(tf.float32,[None,10])
x_image=tf.reshape(x,[-1,28,28,1])

W_conv1 = weight_variable([5, 5, 1, 32])  
b_conv1 = bias_variable([32]) #channel=num
h_conv1 = tf.nn.relu(conv2d(x_image, W_conv1) + b_conv1)  
h_pool1 = max_pool_2x2(h_conv1)  
#
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) 

#1fc
#tf.nn.conv2d:二维的卷积  
#conv2d(input, filter, strides, padding, use_cudnn_on_gpu=None,data_format=None, name=None)  
#filter:A 4-D tensor of shape  
#      `[filter_height, filter_width, in_channels, out_channels]`  
#strides:步长，都是1表示所有点都不会被遗漏。1-D 4值，表示每歌dim的移动步长。  
# padding:边界的处理方式，“SAME"、"VALID”可选  

W_fc1 = weight_variable([7*7*64, 1024])  
b_fc1 = bias_variable([1024])  
h_pool2_flat = tf.reshape(h_pool2, [-1, 7*7*64])  #两次pooling之后28/2/2=7,channels=64
h_fc1 = tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)  

keep_prob = tf.placeholder(tf.float32)  
h_fc1_dropout = tf.nn.dropout(h_fc1, keep_prob) 


#2fc
#tf.nn.max_pool:最大值池化函数，即求2*2区域的最大值，保留最显著的特征。  
#max_pool(value, ksize, strides, padding, data_format="NHWC", name=None)  
#ksize:池化窗口的尺寸  
#strides:[1,2,2,1]表示横竖方向步长为2  
W_fc2 = weight_variable([1024, 10])  
b_fc2 = bias_variable([10])  
y_conv = tf.nn.softmax(tf.matmul(h_fc1_dropout, W_fc2) + b_fc2)  

四、训练

#train
cross_entropy = tf.reduce_mean(-tf.reduce_sum(y_*tf.log(y_conv), reduction_indices=[1]))  
train_step=tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)
#accuracy
correct_prediction=tf.equal(tf.argmax(y_conv,1),tf.argmax(y_,1))
accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))

tf.global_variables_initializer().run() 
for i in range(20000):  
    batch = mnist.train.next_batch(50)  
    if i%100 == 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}) 

五、测试

#在测试集上测试
print("test accuracy %g"%accuracy.eval(feed_dict={x:mnist.test.images, y_:mnist.test.labels, keep_prob:1.0}))  

此代码在我电脑上test accuracy 0.9912