博客
关于我
Tensorboard使用入门,用于可视化loss,accuracy等数据
阅读量:237 次
发布时间:2019-03-01

本文共 2530 字,大约阅读时间需要 8 分钟。

import tensorflow as tffrom tensorflow.examples.tutorials.mnist import input_data# 载入数据集mnist = input_data.read_data_sets("MNIST_data", one_hot=True)# 批次的大小batch_size = 50# 计算一共有多少批次n_batch = mnist.train.num_examples // batch_size# 参数概要def variable_summary(var):    with tf.name_scope("summaries"):        mean = tf.reduce_mean(var)        tf.summary.scalar("mean", mean)   # 平均值    with tf.name_scope("stddev"):        stddev = tf.sqrt(tf.reduce_mean(var-mean))    tf.summary.scalar("stddev", stddev)   # 标准差    tf.summary.scalar("max", tf.reduce_max(var))  # 最大值    tf.summary.scalar("min", tf.reduce_min(var))  # 最小值    tf.summary.histogram("histogram", var)   # 直方图# 输入命名空间with tf.name_scope("input"):    # 定义两个placeholder    x = tf.placeholder(tf.float32, [None, 28 * 28])    y = tf.placeholder(tf.float32, [None, 10])with tf.name_scope("layer"):    # 创建一个简单的神经网络    with tf.name_scope("weight"):        W = tf.Variable(tf.zeros([784, 10]))        variable_summary(W)    with tf.name_scope("biases"):        b = tf.Variable(tf.zeros([10]))        variable_summary(b)    with tf.name_scope("wx_plus_b"):        wx_plus_b = tf.matmul(x, W) + b    with tf.name_scope("softmax"):        prediction = tf.nn.softmax(wx_plus_b)with tf.name_scope("loss"):    # 二次代价函数    loss = tf.reduce_mean(tf.square(y - prediction))    tf.summary.scalar("loss",loss)with tf.name_scope("train"):    # 使用梯度下降法    train_step = tf.train.GradientDescentOptimizer(0.2).minimize(loss)# 初始化变量init = tf.global_variables_initializer()with tf.name_scope("accuracy"):    with tf.name_scope("correct_prediction"):        # 计算正确率,下面是一个布尔型列表        correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(prediction, 1))    with tf.name_scope("accuracy"):        # 求准确率,首先把布尔类型转化为浮点类型        accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))        tf.summary.scalar("accuracy", accuracy)# 合并所有的summarymerged = tf.summary.merge_all()with tf.Session() as sess:    sess.run(init)    # 保存Tensorboard文件    writer = tf.summary.FileWriter("logs/", sess.graph)    for epoch in range(51):        for batch in range(n_batch):            # 使用函数获取一个批次图片            batch_xs, batch_ys = mnist.train.next_batch(batch_size)            summary,_ = sess.run([merged,train_step], feed_dict={x: batch_xs, y: batch_ys})        writer.add_summary(summary, epoch)        acc = sess.run(accuracy, feed_dict={x: mnist.test.images, y: mnist.test.labels})        print("Iter " + str(epoch) + ",Testing Accuracy " + str(acc))
通过在图中定义scalar的方法,保存标量数据,用通过tensorboard可视化出来

转载地址:http://pihv.baihongyu.com/

你可能感兴趣的文章
mysql8.0新特性-自增变量的持久化
查看>>
Mysql8.0注意url变更写法
查看>>
Mysql8.0的特性
查看>>
MySQL8修改密码报错ERROR 1819 (HY000): Your password does not satisfy the current policy requirements
查看>>
MySQL8修改密码的方法
查看>>
Mysql8在Centos上安装后忘记root密码如何重新设置
查看>>
Mysql8在Windows上离线安装时忘记root密码
查看>>
MySQL8找不到my.ini配置文件以及报sql_mode=only_full_group_by解决方案
查看>>
mysql8的安装与卸载
查看>>
MySQL8,体验不一样的安装方式!
查看>>
MySQL: Host '127.0.0.1' is not allowed to connect to this MySQL server
查看>>
Mysql: 对换(替换)两条记录的同一个字段值
查看>>
mysql:Can‘t connect to local MySQL server through socket ‘/var/run/mysqld/mysqld.sock‘解决方法
查看>>
MYSQL:基础——3N范式的表结构设计
查看>>
MYSQL:基础——触发器
查看>>
Mysql:连接报错“closing inbound before receiving peer‘s close_notify”
查看>>
mysqlbinlog报错unknown variable ‘default-character-set=utf8mb4‘
查看>>
mysqldump 参数--lock-tables浅析
查看>>
mysqldump 导出中文乱码
查看>>
mysqldump 导出数据库中每张表的前n条
查看>>