中国大学MOOC-人工智能实践：Tensorflow笔记-课程笔记 Chapter6

本篇博客为学习中国大学MOOC-人工智能实践：Tensorflow笔记课程时的个人笔记记录。具体课程情况可以点击链接查看。（这里推一波中国大学MOOC，很好的学习平台，质量高，种类全，想要学习的话很有用的）
本篇是第六章的学习笔记,前面五章的笔记可以翻看我的博客~

Chapter 6 全连接网络实践

关于上节课留下来的断点续训问题

修改mnist_backward.py文件
修改后的文件内容如下,修改的地方进行了标记

#coding:utf-8
import tensorflow as tf 
from tensorflow.examples.tutorials.mnist import input_data
import mnist_forward
import os

BATCH_SIZE = 200
LEARNING_RATE_BASE = 0.1
LEARNING_RATE_DECAY = 0.99
REGULARIZER = 0.0001
STEPS = 50000
MOVING_AVERAGE_DECAY = 0.99
MODEL_SAVE_PATH = 'G:/model/'   #这里是我选择放置训练好的model的路径,根据自己的需要进行修改
MODEL_NAME = 'mnist_model'
DATA_PATH = 'G:/datasets/mnist' #这里是我放置dataset的路径,根据自己的需要进行修改

def backward(mnist):
    x = tf.placeholder(tf.float32, [None, mnist_forward.INPUT_NODE])
    y_ = tf.placeholder(tf.float32, [None, mnist_forward.OUTPUT_NODE])
    y = mnist_forward.forward(x, REGULARIZER)
    global_step = tf.Variable(0, trainable=False)
    ce = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=y,labels=tf.argmax(y_, 1))
    cem = tf.reduce_mean(ce)
    loss = cem + tf.add_n(tf.get_collection('losses'))
    learning_rate = tf.train.exponential_decay(LEARNING_RATE_BASE,global_step,mnist.train.num_examples / BATCH_SIZE,LEARNING_RATE_DECAY,staircase = True)
    train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_step)
    ema = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY, global_step)
    ema_op = ema.apply(tf.trainable_variables())
    with tf.control_dependencies([train_step, ema_op]):
        train_op = tf.no_op(name = 'train')
    saver = tf.train.Saver()
    with tf.Session() as sess:
        init_op = tf.global_variables_initializer()
        sess.run(init_op)
        # 加入断点续训功能 #################################################modified
        ckpt = tf.train.get_checkpoint_state(MODEL_SAVE_PATH)
        if ckpt and ckpt.model_checkpoint_path:
            saver.restore(sess, ckpt.model_checkpoint_path)
        # 读入原本训练的结果,继续训练############end
        for i in range(STEPS):
            xs, ys = mnist.train.next_batch(BATCH_SIZE)
            _, loss_value, step = sess.run([train_op, loss, global_step], feed_dict={x: xs, y_: ys})
            if i % 1000 == 0:
                print("After %d training steps, loss on training batch is %g." % (step, loss_value))
                saver.save(sess, os.path.join(MODEL_SAVE_PATH,MODEL_NAME),global_step=global_step)

def main():
    mnist = input_data.read_data_sets(DATA_PATH, one_hot = True)
    backward(mnist)

if __name__ == '__main__':
    main()

6.1 输入手写数字图片输出识别结果

Q:如何对输入的真实图片,输出预测的结果,
输入层是784个节点,每个节点是[0,1]之间的浮点数,
输出层是10个可能性概率组成的以为数组

def application():
    testNum = input("input the number of test pictures:")
    for i in range (testNum):
        testPic = raw_input("the path of test pictures:")
        testPicArr = pre_pic(testPic)
        preValue = restore_model(testPicArr)
        print("The prediction number is:",preValue)

相比第五章的代码新增代码文件 mnist_app.py
具体的代码文件 mnist_app.py 内容如下：

#coding:utf-8
import tensorflow as tf 
import numpy as np 
from PIL import Image 
import matplotlib.pyplot as plt
import mnist_backward
import mnist_forward

def restore_model(testPicArr):
    with tf.Graph().as_default() as tg:
        x = tf.placeholder(tf.float32, [None, mnist_forward.INPUT_NODE])
        y = mnist_forward.forward(x, None)
        preValue = tf.arg_max(y, 1)
        variable_averages = tf.train.ExponentialMovingAverage(mnist_backward.MOVING_AVERAGE_DECAY)
        variables_to_restore = variable_averages.variables_to_restore()
        saver = tf.train.Saver(variables_to_restore)

        with tf.Session() as sess:
            ckpt = tf.train.get_checkpoint_state(mnist_backward.MODEL_SAVE_PATH)
            if ckpt and ckpt.model_checkpoint_path:
                saver.restore(sess, ckpt.model_checkpoint_path)
                preValue = sess.run(preValue, feed_dict={x:testPicArr})
                return preValue
            else:
                print("No checkpoint file found")
                return -1

def pre_pic(picName):
    img = Image.open(picName)
    #img = img.convert("L")
    reIm = img.resize((28,28), Image.ANTIALIAS)
    im_arr = np.array(reIm.convert('L'))
    threshold = 100
    for i in range(28):
        for j in range(28):
            im_arr[i][j] = 255-im_arr[i][j]
            if (im_arr[i][j]<threshold):
                im_arr[i][j] = 0
            else: im_arr[i][j] = 255
    plt.figure("figure")
    plt.imshow(im_arr)
    plt.show()
    nm_arr = im_arr.reshape([1, 784])
    nm_arr = nm_arr.astype(np.float32)
    img_ready = np.multiply(nm_arr, 1.0/255.0)
    return img_ready

def application():
    '''
    testNum = int(input("input the number of test picture: "))
    for i in range(testNum):
        testPic = input("the path of test picture: ")
        testPicArr = pre_pic(testPic)
        preValue = restore_model(testPicArr)
        print("The prediction number is:", preValue)
        '''
    for i in range(10):
        imName = 'pic/'+str(i)+'.jpg'
        print("ImageName is:", imName)
        testPicArr = pre_pic(imName)
        preValue = restore_model(testPicArr)
        print("The prediction number is:", preValue)

def main():
    application()

if __name__ == '__main__':
    main()

注意:这里我修改了老师上课使用的代码,改成自动循环十张测试图片,每次读入一张图片,然后处理完成后显示该图片,手动关闭图片窗口程序会继续运行输出测试结果.老师原本使用的代码我注释掉了,可以自行修改.
测试使用的十张手写数字图片我没有找到,于是自己制作了一份,从0~9十张手写数字图片,压缩了一下上传到了CSDN,需要的朋友可以自行下载,(可能我写的太丑了,有几个数字就是识别不对,不过个人感觉也很正常,毕竟只是这么简单的全连接层,识别效果不是很理想也很正常,加上我自己制作的数据集和mnist本身的样式可能差的比较远.)
十张测试图片下载链接:https://download.csdn.net/download/tuzixini/10560123
下载后解压放在和代码同一目录下就好.

6.2 制作数据集

Q:如何制作数据集,实现特定应用
tfrecords文件
tfrecords文件是一种二进制文件,可先将图片和标签制作成该格式的文件,使用tensorflow进行数据读取,会提高内存利用率.
使用tf.train.Example的协议存储训练数据,训练数据的特征用键值对的形式表示.
如:’img_raw’:值 ‘label’:值 值是Byteslist/FloatList/int64List
用SerializeToString()把数据序列化成字符串存储.
生成tfrecords文件

writer = tf.python_io.TFRecordWriter(tfRecoderName) #新建一个writer
for 循环遍历每张图片和标签:
    example = tf.train.Example(features=tf.train.Features(feature={'img_raw':tf.train.Feature(bytes_list=tf.train.BytesList(value=[img_raw])), 'label':tf.train.Feature(int64_list=tf.train.List(value=labels))})) # 把每张图片和标签封装到Example中
    writer.write(example.SerializeToString()) #把example进行序列化
writer.close()

解析tfrecords文件

filename_queue = tf.train.string_input_producer([tfRecord_path])
reader = tf.TFRecordReader() # 新建一个reader
_, serialized_example = reader.read(filename_queue)
features = tf.parse_single_example(serialized_example, feature={'img_raw':tf.FixedLenFeature([],tf.string),'label':tf.FixedLenFeature([10],tf.int64)})
img = tf.decode_raw(features['img_raw'],tf.uint8)
img.set_shape([784])
img = tf.cast(img,tf.float32)*(1./255)
label = tf.cast(features['label'], tf.float32) 

相比6.1 新增代码文件 mnist_generateds.py, 同时修改了mnist_backward.py文件和mnist_test.py文件中图片和标签获取的接口.
mnist_generateds.py 代码内容如下:

#coding:utf-8
import tensorflow as tf 
import numpy as np 
from PIL import Image
import os 

image_train_path = './mnist_data_jpg/mnist_train_jpg_60000/'
label_train_path = './mnist_data_jpg/mnist_train_jpg_60000.txt'
tfRecord_train = './data/mnist_train.tfrecords'
image_test_path = './mnist_data_jpg/mnist_test_jpg_10000/'
label_test_path = './mnist_data_jpg/mnist_train_jpg_10000.txt'
tfRecord_test = './data/mnist_test.tfrecords'
data_path = './data'
resize_height = 28
resize_width = 28

def write_tfRecord(tfRecordName, image_path, label_path):
    writer = tf.python_io.TFRecordWriter(tfRecordName)
    num_pic = 0
    f = open(label_path, 'r')
    contents = f.readlines()
    f.close()
    for content in contents:
        value = content.split()
        image_path = image_path + value[0]
        img = Image.open(image_path)
        img_raw = img.tobytes()
        labels = [0]*10
        labels[int(value[1])] = 1
        example = tf.train.Example(features=tf.train.Features(feature={
            'img_raw': tf.train.Feature(bytes_list=tf.train.BytesList(value=[img_raw])),
            'label': tf.train.Feature(int64_list=tf.train.Int64List(value=labels))
        }))
        writer.write(example.SerializeToString())
        num_pic += 1
        print("the number of picture:", num_pic)
    writer.close()
    print("Write tfrecord sussessful")

def generate_tfRecord():
    isExists = os.path.exists(data_path)
    if not isExists:
        os.makedirs(data_path)
        print("The directory was create successfully")
    else:
        print ("directory already exists")
    write_tfRecord(tfRecord_train, image_train_path, label_train_path)
    write_tfRecord(tfRecord_test, image_test_path, label_test_path)

def read_tfRecord(tfRecord_path):
    filename_queue = tf.train.string_input_producer([tfRecord_path])
    reader = tf.TFRecordReader() # 新建一个reader
    _, serialized_example = reader.read(filename_queue)
    features = tf.parse_single_example(serialized_example, feature={'img_raw':tf.FixedLenFeature([],tf.string),'label':tf.FixedLenFeature([10],tf.int64)})
    img = tf.decode_raw(features['img_raw'],tf.uint8)
    img.set_shape([784])
    img = tf.cast(img,tf.float32)*(1./255)
    label = tf.cast(features['label'], tf.float32)
    return img, label

def get_tfRecord(num, isTrain=True):
    if isTrain:
        tfRecord_path = tfRecord_train
    else:
        tfRecord_path= tfRecord_test
    img, label = read_tfRecord(tfRecord_path)
    img_batch, label_batch = tf.train.shuffle_batch([img, label],batch_size=num,num_threads=2,capacity=1000,min_after_dequeue=700)
    return img_batch, label_batch

def main():
    generate_tfRecord()

if __name__ == '__main__':
    main()

使用多线程提升图片标签批获取的效率
把批获取的操作放到线程协调器开启和关闭的中间

# 开启线程协调器
coord = tf.train.Coordinator()
threads = tf.train.start_queue_runners(sess=sess, coord=coord)

pass

# 关闭线程协调器
coord.request_stop()
coord.join(threads)

修改后用于Chapter 6 的mnist_backward.py代码如下

#coding:utf-8
import tensorflow as tf 
from tensorflow.examples.tutorials.mnist import input_data
import mnist_forward
import os
# chapter 6 添加 ## start
import mnist_generateds
## end

BATCH_SIZE = 200
LEARNING_RATE_BASE = 0.1
LEARNING_RATE_DECAY = 0.99
REGULARIZER = 0.0001
STEPS = 50000
MOVING_AVERAGE_DECAY = 0.99
MODEL_SAVE_PATH = 'G:/model/'   #这里是我选择放置训练好的model的路径,根据自己的需要进行修改
MODEL_NAME = 'mnist_model'
DATA_PATH = 'G:/datasets/mnist' #这里是我放置dataset的路径,根据自己的需要进行修改
# chapter 6 添加 ## start
train_num_examples = 60000
## end

def backward(mnist):
    x = tf.placeholder(tf.float32, [None, mnist_forward.INPUT_NODE])
    y_ = tf.placeholder(tf.float32, [None, mnist_forward.OUTPUT_NODE])
    y = mnist_forward.forward(x, REGULARIZER)
    global_step = tf.Variable(0, trainable=False)
    ce = tf.nn.sparse_softmax_cross_entropy_with_logits(logits=y,labels=tf.argmax(y_, 1))
    cem = tf.reduce_mean(ce)
    loss = cem + tf.add_n(tf.get_collection('losses'))

    # chapter 5 使用,在chapter 6被注释
    # learning_rate = tf.train.exponential_decay(LEARNING_RATE_BASE,global_step,mnist.train.num_examples / BATCH_SIZE,LEARNING_RATE_DECAY,staircase = True)
    # 替换为:
    learning_rate = tf.train.exponential_decay(LEARNING_RATE_BASE,global_step,train_num_examples / BATCH_SIZE,LEARNING_RATE_DECAY,staircase = True)
    ## end

    train_step = tf.train.GradientDescentOptimizer(learning_rate).minimize(loss, global_step=global_step)
    ema = tf.train.ExponentialMovingAverage(MOVING_AVERAGE_DECAY, global_step)
    ema_op = ema.apply(tf.trainable_variables())
    with tf.control_dependencies([train_step, ema_op]):
        train_op = tf.no_op(name = 'train')
    saver = tf.train.Saver()

    # chapter 6 添加 ## start
    img_batch, label_batch = mnist_generateds.generate_tfRecord(BATCH_SIZE, isTrain=True)
    ## end

    with tf.Session() as sess:
        init_op = tf.global_variables_initializer()
        sess.run(init_op)
        # 加入断点续训功能 ###########################################################################modified
        ckpt = tf.train.get_checkpoint_state(MODEL_SAVE_PATH)
        if ckpt and ckpt.model_checkpoint_path:
            saver.restore(sess, ckpt.model_checkpoint_path)
        # 读入原本训练的结果,继续训练

        # chapter 6 添加 ## start
        coord = tf.train.Coordinator()
        threads = tf.train.start_queue_runners(sess=sess, coord=coord)
        ## end

        for i in range(STEPS):
            # chapter 5 使用,在chapter 6被注释
            # xs, ys = mnist.train.next_batch(BATCH_SIZE)
            # 替换为:
            xs, ys = sess.run([img_batch, label_batch])
            ## end

            _, loss_value, step = sess.run([train_op, loss, global_step], feed_dict={x: xs, y_: ys})
            if i % 1000 == 0:
                print("After %d training steps, loss on training batch is %g." % (step, loss_value))
                saver.save(sess, os.path.join(MODEL_SAVE_PATH,MODEL_NAME),global_step=global_step)

        # chapter 6 添加 ## start
        coord.request_stop()
        coord.join(threads)
        ## end

def main():
    mnist = input_data.read_data_sets(DATA_PATH, one_hot = True)
    backward(mnist)

if __name__ == '__main__':
    main()

修改后用于Chapter6 的 mnist_test.py 文件代码如下:

#coding:utf-8
import time
import tensorflow as tf 
from tensorflow.examples.tutorials.mnist import input_data
import mnist_backward
import mnist_forward
## Chapter 6 添加
import mnist_generateds
# end
TEST_INTERVAL_SECS = 5
DATA_PATH = 'G:/datasets/mnist' #这里是我放置dataset的路径,根据自己的需要进行修改

## chapter 6 添加
TEST_NUM = 10000
# end

def test(mnist):
    with tf.Graph().as_default() as g:
        x = tf.placeholder(tf.float32, [None, mnist_forward.INPUT_NODE])
        y_ = tf.placeholder(tf.float32, [None, mnist_forward.OUTPUT_NODE])
        y = mnist_forward.forward(x, None)
        ema = tf.train.ExponentialMovingAverage(mnist_backward.MOVING_AVERAGE_DECAY)
        ema_restore = ema.variables_to_restore()
        saver = tf.train.Saver(ema_restore)
        correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
        accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

        ## Chapter 6 添加
        img_batch, label_batch = mnist_generateds.get_tfRecord(TEST_num, isTrain=False)
        # end

        while True:
            with tf.Session() as sess:
                ckpt = tf.train.get_checkpoint_state(mnist_backward.MODEL_SAVE_PATH)
                if ckpt and ckpt.model_checkpoint_path:
                    saver.restore(sess, ckpt.model_checkpoint_path)
                    global_step = ckpt.model_checkpoint_path.split('/')[-1].split('-')[-1]

                    ## chapter 6 添加
                    coord = tf.train.Coordinator()
                    threads = tf.train.start_queue_runners(sess=sess, coord=coord)
                    xs,ys = sess.run([img_batch, label_batch])
                    # end
                    ## chapter 5 使用,在chapter 6注释
                    # accuracy_score = sess.run(accuracy, feed_dict={x: mnist.test.images, y_: mnist.test.labels})
                    # 替换为:
                    accuracy_score = sess.run(accuracy, feed_dict={x: xs, y_: ys})
                    # end
                    print("After %s training steps, test accuracy = %g" % (global_step, accuracy_score))

                    ## chapter 6添加
                    coord.request_stop()
                    coord.join(threads)
                    # end
                else:
                    print("No checkpoint file found!")
                    return
            time.sleep(TEST_INTERVAL_SECS)

def main():
    mnist = input_data.read_data_sets(DATA_PATH, one_hot=True)
    test(mnist)

if __name__ == '__main__':
    main()