【tensorflow】识别图中模糊手写数字

1、导入NMIST数据集。

手动下载：http://yann.lecun.com/exdb/mnist/

自动下载：

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

 

2、分析MNIST样本特点定义变量。

由于输入图片是个550000*784的矩阵，所以先创建一个[None,784]的占位符x和[None,10]的占位符y，然后使用feed机制将图片和标签输入进去。

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

tf.reset_default_graph()
#定义占位符
x=tf.placeholder(tf.float32,[None,784])# mnist data维度 28*28=784
y=tf.placeholder(tf.flloat32,[None,10])## 0-9 数字=> 10 classes

3、构建模型。

①定义学习参数：使用Variable定义学习参数。

w=tf.Variable(tf.random_normal([784,10]))
b=tf.Variable(tf.zeros([10]))

 

②定义输出节点

pred=tf.nn.softmax(tf.matul(x,w)+b)

这里的x是一个二维张量，拥有多个输入。然后在加上b，把它们的和输入到tf.nn.softmax函数里。

softmax详解参考：https://blog.csdn.net/bitcarmanlee/article/details/82320853?depth_1-utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromBaidu-1&utm_source=distribute.pc_relevant.none-task-blog-BlogCommendFromBaidu-1

③定义反向传播的结构

#损失函数
cost=tf.reduce_mean(-tf.reduce_sum(y*tf.log(pred), reduction_indices=1))
#定义参数
learning_rate=0.01
#使用梯度下降优化器
optimizer=tf.train.GradientDescentOptimzer(learning_rate).minimize(cost)

整个过程就是不断让损失值cost变小，因为损失值越小，才能表明输出的结果跟标签的数据越相近。当cost小到我们的需求时，这时的b和w就是训练出来的合适值。

 

4、训练模型并输出中间状态参数。

training_epochs=25 #整个训练迭代25次
batch_size=100 #训练过程中一次取100条数据进行训练
display_step=1#每训练一次就把具体的中间状态显示出来

#启动session
with tf.Session() as sess:
    see.run(tf.global_variables_initializer())

    #启动循环开始训练
    for epoch in rande(traning_epochs):
        avg_cost=0
        total_batch = int(mnist.train.num_examples/batch_size)
        # 遍历全部数据集
        for i in range(total_batch):
            batch_xs, batch_ys = mnist.train.next_batch(batch_size)
            # 运行优化器
            c = sess.run([optimizer, cost], feed_dict={x: batch_xs,y: batch_ys})
            # 计算loss值
            avg_cost += c / total_batch
        # 显示训练中的详细信息
        if (epoch+1) % display_step == 0:
            print ("Epoch:", '%04d' % (epoch+1), "cost=", "{:.9f}".format(avg_cost))

    print( " Finished!")

5、测试模型

测试错误率的算法：直接判断预测结果与真实标签是否相同，如果相同，就表明是正确的，如果不相同，就表示是错误的。然后正确的个数除以总个数，得到的即为正确率。

correct_prediction=tf.equal(tf.argmax(pred,1),tf.argmax(y, 1)))
    #计算准确率
    accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32)))
    print("Accuracy:"，accuracy.eval(({x: mnist.test.images, y: mnist.test.labels}))))

6、保存模型

#保存模型
save_path=saver.save(sess,model_path)
print("Model saved in file:%s"%save_path)

7、读取模型

with tf.Session() as sess:
    #初始化变量
    sess.run(tf.global_variables_initializer())
    #恢复模型变量
    saver.restre(sess,model_path)

    # 测试 model
    correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
    # 计算准确率
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
    print ("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))
    
    output=tf.argmax(pred,1)
    batch_xs,batch_ys=mnist.train.next_batch(2)
    outputval,predv=see.run([output,pred],feed_dict={x: batch_xs})
    print(outputval,predv,batch_ys)

    im=batch_xs[0]
    im=im.reshape(-1,28)
    pylab.imshow(im)
    pylab.show()

完整代码

import tensorflow as tf #导入tensorflow库
from tensorflow.examples.tutorials.mnist import input_data
mnist = input_data.read_data_sets("MNIST_data/", one_hot=True)
import pylab 

tf.reset_default_graph()
# tf Graph Input
x = tf.placeholder(tf.float32, [None, 784]) # mnist data维度 28*28=784
y = tf.placeholder(tf.float32, [None, 10]) # 0-9 数字=> 10 classes

# Set model weights
W = tf.Variable(tf.random_normal([784, 10]))
b = tf.Variable(tf.zeros([10]))

# 构建模型
pred = tf.nn.softmax(tf.matmul(x, W) + b) # Softmax分类

# Minimize error using cross entropy
cost = tf.reduce_mean(-tf.reduce_sum(y*tf.log(pred), reduction_indices=1))

#参数设置
learning_rate = 0.01
# 使用梯度下降优化器
optimizer = tf.train.GradientDescentOptimizer(learning_rate).minimize(cost)

training_epochs = 25
batch_size = 100
display_step = 1
saver = tf.train.Saver()
model_path = "log/521model.ckpt"

# 启动session
with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())# Initializing OP

    # 启动循环开始训练
    for epoch in range(training_epochs):
        avg_cost = 0.
        total_batch = int(mnist.train.num_examples/batch_size)
        # 遍历全部数据集
        for i in range(total_batch):
            batch_xs, batch_ys = mnist.train.next_batch(batch_size)
            # Run optimization op (backprop) and cost op (to get loss value)
            _, c = sess.run([optimizer, cost], feed_dict={x: batch_xs,
                                                          y: batch_ys})
            # Compute average loss
            avg_cost += c / total_batch
        # 显示训练中的详细信息
        if (epoch+1) % display_step == 0:
            print ("Epoch:", '%04d' % (epoch+1), "cost=", "{:.9f}".format(avg_cost))

    print( " Finished!")

    # 测试 model
    correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
    # 计算准确率
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
    print ("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))

    # Save model weights to disk
    save_path = saver.save(sess, model_path)
    print("Model saved in file: %s" % save_path)



#读取模型
print("Starting 2nd session...")
with tf.Session() as sess:
    # Initialize variables
    sess.run(tf.global_variables_initializer())
    # Restore model weights from previously saved model
    saver.restore(sess, model_path)
    
     # 测试 model
    correct_prediction = tf.equal(tf.argmax(pred, 1), tf.argmax(y, 1))
    # 计算准确率
    accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))
    print ("Accuracy:", accuracy.eval({x: mnist.test.images, y: mnist.test.labels}))
    
    output = tf.argmax(pred, 1)
    batch_xs, batch_ys = mnist.train.next_batch(2)
    outputval,predv = sess.run([output,pred], feed_dict={x: batch_xs})
    print(outputval,predv,batch_ys)

    im = batch_xs[0]
    im = im.reshape(-1,28)
    pylab.imshow(im)
    pylab.show()
    
    im = batch_xs[1]
    im = im.reshape(-1,28)
    pylab.imshow(im)
    pylab.show()