TensorFlow--(四)MNIST手写体数字识别问题

最新推荐文章于 2024-10-15 15:12:58 发布

原创最新推荐文章于 2024-10-15 15:12:58 发布 · 279 阅读

0 ·

CC 4.0 BY-SA版权

Tensorflow 专栏收录该内容

7 篇文章

订阅专栏

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

Successfully downloaded train-images-idx3-ubyte.gz 9912422 bytes.
Extracting ../datasets/MNIST_data/train-images-idx3-ubyte.gz
Successfully downloaded train-labels-idx1-ubyte.gz 28881 bytes.
Extracting ../datasets/MNIST_data/train-labels-idx1-ubyte.gz
Successfully downloaded t10k-images-idx3-ubyte.gz 1648877 bytes.
Extracting ../datasets/MNIST_data/t10k-images-idx3-ubyte.gz
Successfully downloaded t10k-labels-idx1-ubyte.gz 4542 bytes.
Extracting ../datasets/MNIST_data/t10k-labels-idx1-ubyte.gz

print "Training data size: ", mnist.train.num_examples
print "Validating data size: ", mnist.validation.num_examples
print "Testing data size: ", mnist.test.num_examples

Training data size:  55000
Validating data size:  5000
Testing data size:  10000

'''
一个辅助函数,给定神经网络的输入和所有参数,计算神经网络的前向传播结果.
在这里定义了一个使用ReLU激活函数的三层全连接网络.
通过加入隐藏层实现了多层网络结构,通过ReLU激活函数实现了去线性化
在这个函数中也支持传入用于计算参数平均值的类,这样方便在测试时使用滑动平均模型
'''
def inference(input_tensor,avg_class,weights1,biases1,weights2,biases2):
    
    #当没有提供滑动平均类时,直接使用参数当前的取值
    if avg_class == None:
        #计算隐藏层的前向传播结果,这里使用了ReLU激活函数
        layer1 = tf.nn.relu(tf.matmul(input_tensor,weights1)+biases1)
        
        #计算输出层的前向传播结果.
        #因为在计算损失函数时一并计算softmax函数,所以这里不需要加入激活函数.而且不加入softmax不会影响预测结果.
        #因为预测时使用的是不同类型对应节点输出值的相对大小,有没有softmax层对最后分类结果的计算没有影响.
        #于是在计算整个神经网络的前向传播时可以不加入最后的softmax层
        return tf.matmul(layer1,weights2)+biases2
    else:
        #首先使用avg_class.average()函数来计算得出变量的滑动平均值,然后再计算相对应的神经网络前向传播结果
        layer1 = tf.nn.relu(tf.matmul(input_tensor,avg_class.average(weights1))+avg_class.average(biases1))
        return tf.matmul(layer1,avg_class.average(weights2))+avg_class.average(biases2)

#训练模型的过程
def train(mnist):
    x = tf.placeholder(tf.float32,[None, INPUT_NODE],name='x-input')
    y_ = tf.placeholder(tf.float32,[None,OUTPUT_NODE],name='y-input')
    
    #生成隐藏层的参数
    weights1 = tf.Variable(tf.truncated_normal([INPUT_NODE,LAYER1_NODE],stddev=0.1))
    biases1 = tf.Variable(tf.constant(0.1,shape=[LAYER1_NODE]))
    
    #生成输出层的参数
    weights2 = tf.Variable(tf.truncated_normal([LAYER1_NODE,OUTPUT_NODE],stddev=0.1))
    biases2 = tf.Variable(tf.constant(0.1,shape=[OUTPUT_NODE]))
    
    #计算在当前参数下神经网络前向传播的结果.这里给出的用于计算滑动平均的类为None,所以函数不会使用参数的滑动平均值
    y = inference(x,None,weights1,biases1,weights2,biases2)
    
    #定义存储训练轮数的变量.这个变量不需要计算滑动平均值,所以这里指定这个变量为不可训练的变量(trainable=False).
    #在TensorFlow训练神经网络时,一般会将训练轮数的变量指定为不可训练的参数
    global_step = tf.Variable(0,trainable=False)  
    
    #给定滑动平均衰减率和训练轮数的变量,初始化滑动平均类.
    #给定训练轮数的变量可以加快训练早期的更新速度
    variable_averages = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY,global_step)
    
    #在所有代表神经网络参数的变量上使用滑动平均.其他辅助变量(比如global_step)就不需要了.
    #tf.trainable_variables()返回的就是图上集合GrapKeys.TRAINABLE_VARIABLES中的元素.这个集合的元素就是所有没有指定trainalbe=False的参数
    variables_averages_op = variable_averages.apply(tf.trainable_variables())
    
    #计算使用了滑动平均之后的前向传播结果.
    #滑动平均不会改变变量本身的取值,而是会维护一个影子变量来记录其滑动平均值.
    #所以当需要使用这个滑动平均值时,需要明确调用average函数
    average_y = inference(x,variable_averages,weights1,biases1,weights2,biases2)
    
    #计算交叉熵作为刻画预测值和真实值之间差距的损失函数
    #这里使用了TensorFlow中提供的 tf.nn.sparse_softmax_cross_entropy_with_logits()函数来计算交叉熵
    #当分类问题只要一个正确答案时,可以使用这个函数来加速交叉熵的计算.
    #MNIST问题的图片中只包含了0~9中的一个数字,所以可以使用这个函数来计算交叉熵损失
    #这个函数的第一个参数:是神经网络不包括softmax层的前向传播结果
    #这个函数的第二个参数:是训练数据的正确答案.因为标准答案是一个长度为10的一维数组,而该函数需要提供的是一个正确答案的数字
    #tf.argmax()函数得到的正确答案对应的类别编号
    cross_entropy = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=y,labels=tf.argmax(y_,1))
    
    #计算在当前batch中所有样例的交叉熵平均值
    cross_entropy_mean = tf.reduce_mean(cross_entropy)
    
    #计算L2正则化损失函数
    regularizer = tf.contrib.layers.l2_regularizer(REGULARIZATION_RATE)
    #计算模型的正则化损失.一般只计算神经网络边上权重的正则化损失,而不使用偏置项
    regularization = regularizer(weights1)+regularizer(weights2)
    
    #总损失等于交叉熵损失和正则化损失的和
    loss = cross_entropy_mean + regularization
    
    #设置指数衰减的学习率
    learning_rate = tf.train.exponential_decay(
                                                LEARNING_RATE_BASE,                     #基础的学习率,随着迭代的进行,更新变量时使用的学习率在这个基础上递减
                                                global_step,                            #当前迭代的轮数
                                                mnist.train.num_examples / BATCH_SIZE,  #过完所有的训练数据需要的迭代次数
                                                LEARNING_RATE_DECAY)                    #学习率衰减速度
    
   #使用tf.train.GradientDescentOptimizer()优化算法来优化损失函数.注意这里损失函数包含了交叉熵损失和L2正则化损失 
    train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss,global_step=global_step)
    
    with tf.control_dependencies([train_step,variables_averages_op]):
        train_op = tf.no_op(name='train')
        
        correct_prediction = tf.equal(tf.argmax(average_y,1),tf.argmax(y_,1))
        accuracy = tf.reduce_mean(tf.cast(correct_prediction,tf.float32))
    
    #初始化会话并开始训练过程
    with tf.Session() as sess:
        tf.global_variables_initializer().run()
        
        #准备验证数据.
        #一般在神经网络的训练过程中会通过验证数据来大致判断停止的条件和评判训练的效果
        validate_feed = {x:mnist.validation.images,y_:mnist.validation.labels}
        
        #准备测试数据.
        #在真实的应用中,这部分数据在训练时是不可见的,这个数据只是作为模型优劣的最后评价标准
        test_feed = {x:mnist.test.images,y_:mnist.test.labels}
        
        #迭代训练神经网络
        for i in range(TRAINING_STEPS):
            #没1000轮输出一次在验证数据集上的测试结果
            if i % 1000 == 0:
                
                #计算滑动平均模型在验证数据上的结果.
                #因为MNIST数据集比较小,所以一次可以处理所有的验证数据.
                #为了计算方便,本样例程序没有将验证数据划分为更小batch.
                #当神经网络模型比较复杂或者验证数据比较大时,太小的batch会导致计算时间过长甚至发生内存溢出的错误
                validate_acc = sess.run(accuracy,feed_dict=validate_feed)
                print("After %d training step(s),validation accuracy using average model is %g " % (i,validate_acc))
        
            #产生这一轮使用的一个batch的训练结果,并运行训练过程
            xs,ys = mnist.train.next_batch(BATCH_SIZE)
            sess.run(train_op,feed_dict={x:xs,y_:ys})
    
        #在训练结束之后,在测试数据上检测神经网络模型的最终正确率
        test_acc = sess.run(accuracy,feed_dict=test_feed)
        print("After %d training step(s),test accuracy using average model is %g " % (TRAINING_STEPS,test_acc))

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

INPUT_NODE = 784     #输入层的节点数.对于MNITST数据集,这个就等于图片的像素, 28x28=784
OUTPUT_NODE = 10     #输出层的节点数.这个等于类别的数据.因为在MNIST数据集中需要区分的是0~9这10个数,所以这里输出层的节点数为10
LAYER1_NODE = 500    #隐藏层的节点数.这里使用只有一个隐藏层的网络结构作为样例.这个隐藏层有500个节点
BATCH_SIZE = 100     #一个训练batch中的训练数据个数.
                     #数据越小时,训练过程越接近随机梯度下降;数据越大时,训练越接近梯度下降

LEARNING_RATE_BASE = 0.8         #基础的学习率
LEARNING_RATE_DECAY = 0.99       #学习率的衰减率
REGULARIZATION_RATE = 0.0001     #描述模型复杂度的正则化项在损失函数中的系数
TRAINING_STEPS = 30000           #训练轮数
MOVING_AVERAGE_DECAY = 0.99      #滑动平均衰减率

#声明处理MINST数据集的类,这个类在初始化时会自动下载数据
mnist = input_data.read_data_sets("../datasets/MNIST_data/",one_hot=True)
train(mnist)

Extracting ../datasets/MNIST_data/train-images-idx3-ubyte.gz
Extracting ../datasets/MNIST_data/train-labels-idx1-ubyte.gz
Extracting ../datasets/MNIST_data/t10k-images-idx3-ubyte.gz
Extracting ../datasets/MNIST_data/t10k-labels-idx1-ubyte.gz
After 0 training step(s),validation accuracy using average model is 0.0862 
After 1000 training step(s),validation accuracy using average model is 0.977 
After 2000 training step(s),validation accuracy using average model is 0.9798 
After 3000 training step(s),validation accuracy using average model is 0.9832 
After 4000 training step(s),validation accuracy using average model is 0.9844 
After 5000 training step(s),validation accuracy using average model is 0.9848 
After 6000 training step(s),validation accuracy using average model is 0.9848 
After 7000 training step(s),validation accuracy using average model is 0.985 
After 8000 training step(s),validation accuracy using average model is 0.9852 
After 9000 training step(s),validation accuracy using average model is 0.9854 
After 10000 training step(s),validation accuracy using average model is 0.986 
After 11000 training step(s),validation accuracy using average model is 0.9856 
After 12000 training step(s),validation accuracy using average model is 0.986 
After 13000 training step(s),validation accuracy using average model is 0.9862 
After 14000 training step(s),validation accuracy using average model is 0.9866 
After 15000 training step(s),validation accuracy using average model is 0.9868 
After 16000 training step(s),validation accuracy using average model is 0.9868 
After 17000 training step(s),validation accuracy using average model is 0.9858 
After 18000 training step(s),validation accuracy using average model is 0.9868 
After 19000 training step(s),validation accuracy using average model is 0.9864 
After 20000 training step(s),validation accuracy using average model is 0.9864 
After 21000 training step(s),validation accuracy using average model is 0.9862 
After 22000 training step(s),validation accuracy using average model is 0.9858 
After 23000 training step(s),validation accuracy using average model is 0.9868 
After 24000 training step(s),validation accuracy using average model is 0.9872 
After 25000 training step(s),validation accuracy using average model is 0.9864 
After 26000 training step(s),validation accuracy using average model is 0.9866 
After 27000 training step(s),validation accuracy using average model is 0.9872 
After 28000 training step(s),validation accuracy using average model is 0.9872 
After 29000 training step(s),validation accuracy using average model is 0.9868 
After 30000 training step(s),test accuracy using average model is 0.9835