python3下tensorflow练习（五）之CNN下的MNIST识别

1. 数据准备

import tensorflow as tf
import tensorflow.examples.tutorials.mnist.input_data as input_data
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)

2.建立共享函数

def weight(shape):
    return tf.Variable(tf.truncated_normal(shape, stddev=0.1),
                       name ='W')

def bias(shape):
    return tf.Variable(tf.constant(0.1, shape=shape)
                       , name = 'b')

def conv2d(x, W):
    return tf.nn.conv2d(x, W, strides=[1,1,1,1], 
                        padding='SAME')

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

tf.truncated_normal(shape, mean=0.0, stddev=1.0, dtype=tf.float32, seed=None, name=None)
引用：https://blog.csdn.net/u013713117/article/details/65446361
从截断的正态分布中输出随机值。 
生成的值服从具有指定平均值和标准偏差的正态分布，如果生成的值大于平均值2个标准偏差的值则丢弃重新选择。

在正态分布的曲线中，横轴区间（μ-σ，μ+σ）内的面积为68.268949%。 
横轴区间（μ-2σ，μ+2σ）内的面积为95.449974%。 
横轴区间（μ-3σ，μ+3σ）内的面积为99.730020%。 
X落在（μ-3σ，μ+3σ）以外的概率小于千分之三，在实际问题中常认为相应的事件是不会发生的，基本上可以把区间（μ-3σ，μ+3σ）看作是随机变量X实际可能的取值区间，这称之为正态分布的“3σ”原则。 
在tf.truncated_normal中如果x的取值在区间（μ-2σ，μ+2σ）之外则重新进行选择。这样保证了生成的值都在均值附近。

参数:
shape: 一维的张量，也是输出的张量。
mean: 正态分布的均值。
stddev: 正态分布的标准差。
dtype: 输出的类型。
seed: 一个整数，当设置之后，每次生成的随机数都一样。
name: 操作的名字。

3.输入层 Input Layer

with tf.name_scope('Input_Layer'):
    x = tf.placeholder("float",shape=[None, 784]
                       ,name="x")    
    x_image = tf.reshape(x, [-1, 28, 28, 1])

4.卷积层1

with tf.name_scope('C1_Conv'):
    W1 = weight([5,5,1,16])
    b1 = bias([16])
    Conv1=conv2d(x_image, W1)+ b1
    C1_Conv = tf.nn.relu(Conv1 )

with tf.name_scope('C1_Pool'):
    C1_Pool = max_pool_2x2(C1_Conv)

5.卷积层2

with tf.name_scope('C2_Conv'):
    W2 = weight([5,5,16,36])
    b2 = bias([36])
    Conv2=conv2d(C1_Pool, W2)+ b2
    C2_Conv = tf.nn.relu(Conv2)

with tf.name_scope('C2_Pool'):
    C2_Pool = max_pool_2x2(C2_Conv) 

6.全连接层

with tf.name_scope('D_Flat'):
    D_Flat = tf.reshape(C2_Pool, [-1, 1764])#7*7*56=1766

with tf.name_scope('D_Hidden_Layer'):
    W3= weight([1764, 128])
    b3= bias([128])
    D_Hidden = tf.nn.relu(
                  tf.matmul(D_Flat, W3)+b3)
    D_Hidden_Dropout= tf.nn.dropout(D_Hidden, ####若在结构上不定义keep_prob,则需要在结构中定义keep_prob = tf.placeholder(tf.float32)，###在session()中用feed_dict()喂参数进去
                                keep_prob=0.8)####在结构上直接定义参数大小，直接赋值

7.输出层Output

with tf.name_scope('Output_Layer'):
    W4 = weight([128,10])
    b4 = bias([10])
    y_predict= tf.nn.softmax(
                 tf.matmul(D_Hidden_Dropout,
                           W4)+b4)

8.设置训练模型最优化步骤（反向传播更新参数）

with tf.name_scope("optimizer"):
    
    y_label = tf.placeholder("float", shape=[None, 10], 
                              name="y_label")
    
    loss_function = tf.reduce_mean(
                      tf.nn.softmax_cross_entropy_with_logits
                         (logits=y_predict , 
                          labels=y_label))
    
    optimizer = tf.train.AdamOptimizer(learning_rate=0.0001) \
                    .minimize(loss_function)
    saver_path=saver.save(sess, "saveModel/CNN_model1)##保存模型  

9.评估模型准确率

with tf.name_scope("evaluate_model"):
    correct_prediction = tf.equal(tf.argmax(y_predict, 1),
                                  tf.argmax(y_label, 1))
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, "float"))

10.训练模型

trainEpochs = 30##30个周期，每个周期对训练的样本55000样本进行训练
batchSize = 100#批尺寸100
totalBatchs = int(mnist.train.num_examples/batchSize)##55000/100=550
epoch_list=[];accuracy_list=[];loss_list=[];
from time import time
startTime=time()
sess = tf.Session()
sess.run(tf.global_variables_initializer())

for epoch in range(trainEpochs):

    
    for i in range(totalBatchs):
        batch_x, batch_y = mnist.train.next_batch(batchSize)
        sess.run(optimizer,feed_dict={x: batch_x,
                                      y_label: batch_y})
        
    
    loss,acc = sess.run([loss_function,accuracy],
                        feed_dict={x: mnist.validation.images, 
                                   y_label: mnist.validation.labels})

    epoch_list.append(epoch)
    loss_list.append(loss);accuracy_list.append(acc)    
    
    print("Train Epoch:", '%02d' % (epoch+1), \
          "Loss=","{:.9f}".format(loss)," Accuracy=",acc)
    
duration =time()-startTime
print("Train Finished takes:",duration)         

训练结果：

11.可视化cost和acc:

%matplotlib inline
import matplotlib.pyplot as plt
plt.plot(epoch_list, accuracy_list,label="accuracy" )
fig = plt.gcf()
fig.set_size_inches(4,2)
plt.ylim(0.8,1)
plt.ylabel('accuracy')
plt.xlabel('epoch')
plt.legend()
plt.show()

fig = plt.gcf()
fig.set_size_inches(4,2)
plt.plot(epoch_list, loss_list, label = 'loss')
plt.ylabel('loss')
plt.xlabel('epoch')
plt.legend(['loss'], loc='upper left')

12.保存模型

saver = tf.train.Saver()##session()上面
#save_path = saver.save(sess, "saveModel/CNN_model1")这句话写在session()里面
merged = tf.summary.merge_all()
train_writer = tf.summary.FileWriter('log/CNN',sess.graph)

13.加载模型和做出预测

saver = tf.train.Saver()
  with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    saver.restore(sess,"saveModel/CNN_model1")
    y_predict=sess.run(y_predict, feed_dict={x: xxxxx})###做出预测传入要预测的图片xxxxx
    print('test accuracy %g' % accuracy.eval(feed_dict={###打印训练好的模型和测试集相比的准确率
        x: mnist.test.images, y_label: mnist.test.labels, keep_prob: 1.0}))