11手写数字识别-卷积神经网络

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

mnist = input_data.read_data_sets("MNIST_data", one_hot=True)

#每批次的大小
batch_size = 100
#计算一共有多少批次
n_batch = mnist.train.num_examples // batch_size

#定义函数，初始化权值
def weight_variable(shape):#shape表示权值的形状
    initial = tf.truncated_normal(shape, stddev=0.1) #生成截断的正态分布
    return tf.Variable(initial)#返回初始化结果
#初始化偏执值
def bias_variable(shape):
    initial = tf.constant(0.1, shape=shape)
    return tf.Variable(initial)
#定义卷积层
def conv2d(x, W):
    #使用tf.nn.conv2d库，conv2d表示2维卷积操作
    #x input tensor of shape[batch，in_height,in_widthin_channels]，批次大小、长、宽、通道数
    #w filter/kernel tensor of shape [filter_height，filter_width,in_channels,out_channels]，滤波器长、宽、输入通道数、输出通道数
    # strides[O] = strides[3] = 1 . strides[1]代表x方向的步长，strides[2]代表y方向的步长
    #padding: A string` from:""SAME""，"VALID"
    return tf.nn.conv2d(x,W,strides=[1,1,1,1],padding='SAME')
#池化层
def max_pool_2x2(x):
    #ksize [1,x, y,1],步长，x, y方向窗口大小，池化窗口为2乘2；strides=[1,2,2,1]步长
    return tf.nn.max_pool(x,ksize=[1,2,2,1],strides=[1,2,2,1],padding='SAME')

#定义两个placeholder
x=tf.placeholder(tf.float32,[None,784]) #28*28
y=tf.placeholder(tf.float32,[None,10])

#改变x的格式转为4D的向量[batch, in_height, in_width, in_channels]，图像是2维的而784是一维的。
x_image = tf.reshape(x,[-1,28,28,1])

#初始化第一个卷积层的权值和偏置
W_conv1=weight_variable([5,5,1,32]) #5*5的采样窗口，32个卷积核从一个平面抽取特征。输入1个通道，输出32个通道，采样完得到32个特征平面
b_conv1=bias_variable([32]) #每一个卷积核一个偏执值 

#把x_image和权重向量进行卷积，再加上偏执值，然后应用于relu激活函数
h_conv1=tf.nn.relu(conv2d(x_image,W_conv1)+b_conv1)#第一个卷积层计算后所得结果
h_pool1=max_pool_2x2(h_conv1) #进行max-pooling，经过池化计算所得结果 

#初始化第二个卷积层的权重和偏置
W_conv2=weight_variable([5,5,32,64]) #5*5的采样窗口，64个卷积核从32个平面抽取特征
b_conv2=bias_variable([64]) #每一个卷积核一个偏执值

#把h_pool1和权值向量进行卷积，再加上偏执值，然后应用于relu激活函数
h_conv2=tf.nn.relu(conv2d(h_pool1,W_conv2)+b_conv2)
h_pool2=max_pool_2x2(h_conv2)#进行maxpooling

#28*28的图像第一次卷积后还是28*28,第一次池化后变成14*14，最后得到32张14*14的平面
#第二次卷积后为14*14，第二次池化后为7*7，最后得到64张7*7的平面
#经过上面的操作后得到64张7*7的平面

#初始化第一个全连接的权值
W_fc1=weight_variable([7*7*64,100]) #上一层有7*7*64个神经元，全连接层有1024个神经元
b_fc1=bias_variable([100]) #1024个节点

#把池化层2的输出扁平化为1维
h_pool2_flat=tf.reshape(h_pool2,[-1,7*7*64])#-1代表任意值此处batch为100
#求第一个全连接层的输出
h_fc1=tf.nn.relu(tf.matmul(h_pool2_flat, W_fc1) + b_fc1)

#keep_prob用来表示神经元的输出概率
keep_prob=tf.placeholder(tf.float32)
h_fc1_drop=tf.nn.dropout(h_fc1,keep_prob)

#初始化第二个全连接层
W_fc2=weight_variable([100,10])
b_fc2=bias_variable([10])

#计算输出
prediction = tf.nn.softmax(tf.matmul(h_fc1_drop, W_fc2)+b_fc2)

#交叉墒代价函数
cross_entropy=tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y,logits=prediction))
#使用AdamOptimizier进行优化
train_step=tf.train.AdamOptimizer(1e-4).minimize(cross_entropy)

#结果存放在一个布尔列表中
correct_prediction=tf.equal(tf.argmax(prediction,1), tf.argmax(y,1)) #argmax返回一维张量中最大的值所在的位置

#求准确率
accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    for epoch in range(21):
        for batch in range(n_batch):
            batch_xs,batch_ys=mnist.train.next_batch(batch_size)
            sess.run(train_step,feed_dict={x:batch_xs,y:batch_ys,keep_prob:0.7})

        acc=sess.run(accuracy,feed_dict={x:mnist.test.images, y:mnist.test.labels,keep_prob:1.0})
        print ("Iter "+str(epoch) + ", Testing Accuracy= " +str(acc))

WARNING:tensorflow:From <ipython-input-1-82e795e89438>:4: read_data_sets (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.
Instructions for updating:
Please use alternatives such as official/mnist/dataset.py from tensorflow/models.
WARNING:tensorflow:From C:\Users\CMM\Anaconda3\envs\Multi-Channelpytorch1py36\lib\site-packages\tensorflow_core\contrib\learn\python\learn\datasets\mnist.py:260: maybe_download (from tensorflow.contrib.learn.python.learn.datasets.base) is deprecated and will be removed in a future version.
Instructions for updating:
Please write your own downloading logic.
WARNING:tensorflow:From C:\Users\CMM\Anaconda3\envs\Multi-Channelpytorch1py36\lib\site-packages\tensorflow_core\contrib\learn\python\learn\datasets\mnist.py:262: extract_images (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.
Instructions for updating:
Please use tf.data to implement this functionality.
Extracting MNIST_data\train-images-idx3-ubyte.gz
WARNING:tensorflow:From C:\Users\CMM\Anaconda3\envs\Multi-Channelpytorch1py36\lib\site-packages\tensorflow_core\contrib\learn\python\learn\datasets\mnist.py:267: extract_labels (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.
Instructions for updating:
Please use tf.data to implement this functionality.
Extracting MNIST_data\train-labels-idx1-ubyte.gz
WARNING:tensorflow:From C:\Users\CMM\Anaconda3\envs\Multi-Channelpytorch1py36\lib\site-packages\tensorflow_core\contrib\learn\python\learn\datasets\mnist.py:110: dense_to_one_hot (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.
Instructions for updating:
Please use tf.one_hot on tensors.
Extracting MNIST_data\t10k-images-idx3-ubyte.gz
Extracting MNIST_data\t10k-labels-idx1-ubyte.gz
WARNING:tensorflow:From C:\Users\CMM\Anaconda3\envs\Multi-Channelpytorch1py36\lib\site-packages\tensorflow_core\contrib\learn\python\learn\datasets\mnist.py:290: DataSet.__init__ (from tensorflow.contrib.learn.python.learn.datasets.mnist) is deprecated and will be removed in a future version.
Instructions for updating:
Please use alternatives such as official/mnist/dataset.py from tensorflow/models.
WARNING:tensorflow:From <ipython-input-1-82e795e89438>:70: calling dropout (from tensorflow.python.ops.nn_ops) with keep_prob is deprecated and will be removed in a future version.
Instructions for updating:
Please use `rate` instead of `keep_prob`. Rate should be set to `rate = 1 - keep_prob`.
WARNING:tensorflow:From <ipython-input-1-82e795e89438>:80: softmax_cross_entropy_with_logits (from tensorflow.python.ops.nn_ops) is deprecated and will be removed in a future version.
Instructions for updating:

Future major versions of TensorFlow will allow gradients to flow
into the labels input on backprop by default.

See `tf.nn.softmax_cross_entropy_with_logits_v2`.

Iter 0, Testing Accuracy= 0.9204
Iter 1, Testing Accuracy= 0.949
Iter 2, Testing Accuracy= 0.9614
Iter 3, Testing Accuracy= 0.9664
Iter 4, Testing Accuracy= 0.9719
Iter 5, Testing Accuracy= 0.9755
Iter 6, Testing Accuracy= 0.9781
Iter 7, Testing Accuracy= 0.9796
Iter 8, Testing Accuracy= 0.9814
Iter 9, Testing Accuracy= 0.982
Iter 10, Testing Accuracy= 0.9833
Iter 11, Testing Accuracy= 0.983
Iter 12, Testing Accuracy= 0.9854
Iter 13, Testing Accuracy= 0.9855
Iter 14, Testing Accuracy= 0.9849
Iter 15, Testing Accuracy= 0.9867
Iter 16, Testing Accuracy= 0.9855
Iter 17, Testing Accuracy= 0.9879
Iter 18, Testing Accuracy= 0.9878
Iter 19, Testing Accuracy= 0.9877
Iter 20, Testing Accuracy= 0.9889