zoukankan      html  css  js  c++  java
  • 手工设计神经MNIST使分类精度达到98%以上

    设计了两个隐藏层,激活函数是tanh,使用Adam优化算法,学习率随着epoch的增大而调低

    import tensorflow as tf
    from tensorflow.examples.tutorials.mnist import input_data
    
    #载入数据集
    mnist = input_data.read_data_sets("MNIST_data",one_hot=True)
    
    #每个批次的大小
    batch_size = 32
    #计算一共有多少个批次
    n_batch = mnist.train.num_examples // batch_size
    
    #定义两个placeholder
    x = tf.placeholder(tf.float32,[None,784])
    y = tf.placeholder(tf.float32,[None,10])
    keep_prob=tf.placeholder(tf.float32)
    lr = tf.Variable(0.001, dtype=tf.float32)
    
    #创建一个简单的神经网络
    W1 = tf.Variable(tf.truncated_normal([784,500],stddev=0.1))
    b1 = tf.Variable(tf.zeros([500])+0.1)
    L1 = tf.nn.tanh(tf.matmul(x,W1)+b1)
    L1_drop = tf.nn.dropout(L1,keep_prob) 
    
    W2 = tf.Variable(tf.truncated_normal([500,300],stddev=0.1))
    b2 = tf.Variable(tf.zeros([300])+0.1)
    L2 = tf.nn.tanh(tf.matmul(L1_drop,W2)+b2)
    L2_drop = tf.nn.dropout(L2,keep_prob) 
    
    W3 = tf.Variable(tf.truncated_normal([300,10],stddev=0.1))
    b3 = tf.Variable(tf.zeros([10])+0.1)
    prediction = tf.nn.softmax(tf.matmul(L2_drop,W3)+b3)
    
    #交叉熵代价函数
    loss = tf.reduce_mean(tf.nn.softmax_cross_entropy_with_logits(labels=y,logits=prediction))
    #训练
    train_step = tf.train.AdamOptimizer(lr).minimize(loss)
    
    #初始化变量
    init = tf.global_variables_initializer()
    
    #结果存放在一个布尔型列表中
    correct_prediction = tf.equal(tf.argmax(y,1),tf.argmax(prediction,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(init)
        for epoch in range(51):
            sess.run(tf.assign(lr, 0.001 * (0.95 ** epoch)))
            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:1.0})
            
            learning_rate = sess.run(lr)
            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) + ", Learning Rate= " + str(learning_rate))


    #
    Iter 0, Testing Accuracy= 0.954, Learning Rate= 0.001
    Iter 1, Testing Accuracy= 0.9624, Learning Rate= 0.00095
    Iter 2, Testing Accuracy= 0.9668, Learning Rate= 0.0009025
    Iter 3, Testing Accuracy= 0.9665, Learning Rate= 0.000857375
    Iter 4, Testing Accuracy= 0.9725, Learning Rate= 0.00081450626
    Iter 5, Testing Accuracy= 0.9738, Learning Rate= 0.0007737809
    Iter 6, Testing Accuracy= 0.9769, Learning Rate= 0.0007350919
    Iter 7, Testing Accuracy= 0.9771, Learning Rate= 0.0006983373
    Iter 8, Testing Accuracy= 0.9777, Learning Rate= 0.0006634204
    Iter 9, Testing Accuracy= 0.9764, Learning Rate= 0.0006302494
    Iter 10, Testing Accuracy= 0.9753, Learning Rate= 0.0005987369
    Iter 11, Testing Accuracy= 0.9779, Learning Rate= 0.0005688001
    Iter 12, Testing Accuracy= 0.9777, Learning Rate= 0.0005403601
    Iter 13, Testing Accuracy= 0.9774, Learning Rate= 0.0005133421
    Iter 14, Testing Accuracy= 0.9772, Learning Rate= 0.000487675
    Iter 15, Testing Accuracy= 0.9803, Learning Rate= 0.00046329122
    Iter 16, Testing Accuracy= 0.9802, Learning Rate= 0.00044012666
    Iter 17, Testing Accuracy= 0.9791, Learning Rate= 0.00041812033
    Iter 18, Testing Accuracy= 0.9806, Learning Rate= 0.00039721432
    Iter 19, Testing Accuracy= 0.9803, Learning Rate= 0.0003773536
    Iter 20, Testing Accuracy= 0.9796, Learning Rate= 0.00035848594
    Iter 21, Testing Accuracy= 0.9803, Learning Rate= 0.00034056162
    Iter 22, Testing Accuracy= 0.9788, Learning Rate= 0.00032353355
    Iter 23, Testing Accuracy= 0.9819, Learning Rate= 0.00030735688
    Iter 24, Testing Accuracy= 0.975, Learning Rate= 0.000291989
    Iter 25, Testing Accuracy= 0.9808, Learning Rate= 0.00027738957
    Iter 26, Testing Accuracy= 0.9814, Learning Rate= 0.0002635201
    Iter 27, Testing Accuracy= 0.9802, Learning Rate= 0.00025034408
    Iter 28, Testing Accuracy= 0.9809, Learning Rate= 0.00023782688
    Iter 29, Testing Accuracy= 0.9811, Learning Rate= 0.00022593554
    Iter 30, Testing Accuracy= 0.9816, Learning Rate= 0.00021463877
    Iter 31, Testing Accuracy= 0.9812, Learning Rate= 0.00020390682
    Iter 32, Testing Accuracy= 0.9815, Learning Rate= 0.00019371149
    Iter 33, Testing Accuracy= 0.9815, Learning Rate= 0.0001840259
    Iter 34, Testing Accuracy= 0.9813, Learning Rate= 0.00017482461
    Iter 35, Testing Accuracy= 0.981, Learning Rate= 0.00016608338
    Iter 36, Testing Accuracy= 0.9806, Learning Rate= 0.00015777921
    Iter 37, Testing Accuracy= 0.9818, Learning Rate= 0.00014989026
    Iter 38, Testing Accuracy= 0.982, Learning Rate= 0.00014239574
    Iter 39, Testing Accuracy= 0.9813, Learning Rate= 0.00013527596
    Iter 40, Testing Accuracy= 0.9818, Learning Rate= 0.00012851215
    Iter 41, Testing Accuracy= 0.9827, Learning Rate= 0.00012208655
    Iter 42, Testing Accuracy= 0.9826, Learning Rate= 0.00011598222
    Iter 43, Testing Accuracy= 0.9814, Learning Rate= 0.00011018311
    Iter 44, Testing Accuracy= 0.9823, Learning Rate= 0.000104673956
    Iter 45, Testing Accuracy= 0.9828, Learning Rate= 9.944026e-05
    Iter 46, Testing Accuracy= 0.9824, Learning Rate= 9.446825e-05
    Iter 47, Testing Accuracy= 0.9824, Learning Rate= 8.974483e-05
    Iter 48, Testing Accuracy= 0.983, Learning Rate= 8.525759e-05
    Iter 49, Testing Accuracy= 0.9827, Learning Rate= 8.099471e-05
    Iter 50, Testing Accuracy= 0.9828, Learning Rate= 7.6944976e-05

    最终达到了0.9828的准确率

    人生苦短,何不用python
  • 相关阅读:
    五种开源协议(GPL,LGPL,BSD,MIT,Apache)
    我想要做的编码机器人
    2017年前端框架、类库、工具大比拼
    Python实现好友全头像的拼接
    Python 中的闭包与装饰器
    Vuex2.0边学边记+两个小例子
    Vue.js写一个SPA登录页面的过程
    react与vue
    20行js代码制作网页刮刮乐
    css3动画详解
  • 原文地址:https://www.cnblogs.com/yqpy/p/11163147.html
Copyright © 2011-2022 走看看