基于TensorFlow进行TensorBoard可视化

 # -*- coding: utf-8 -*-
 """
 Created on Thu Nov  1 17:51:28 2018

 @author: zhen
 """

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

 max_steps = 1000
 learning_rate = 0.001
 dropout = 0.9
 data_dir = 'C:/Users/zhen/MNIST_data_bak/'
 log_dir = 'C:/Users/zhen/MNIST_log_bak/'

 mnist = input_data.read_data_sets(data_dir, one_hot=True) # 加载数据，把数据转换成one_hot编码
 sess = tf.InteractiveSession() # 创建内置sess

 with tf.name_scope('input'):
     x = tf.placeholder(tf.float32, [None, 784], name='x-inpupt')
     y_ = tf.placeholder(tf.float32, [None, 10], name='y-input')

 with tf.name_scope("input_reshape"):
     image_shaped_input = tf.reshape(x, [-1, 28, 28, 1])
     tf.summary.image('input', image_shaped_input, 10) # 输出包含图像的summary，该图像有四维张量构建，用于可视化

 # 定义神经网络的初始化方法
 def weight_variable(shape):
     initial = tf.truncated_normal(shape, stddev=0.1) #从截断的正态分布中输出随机值，类似tf.random_normal从正态分布中输出随机值
     return tf.Variable(initial)

 def bias_variable(shape):
     initial = tf.constant(0.1, shape=shape)
     return tf.Variable(initial)

 # 定义Variable变量的数据汇总函数
 def variable_summaries(var):
     with tf.name_scope('summaries'):
         mean = tf.reduce_mean(var) # 求平均值
         tf.summary.scalar('mean', mean) # 输出一个含有标量值的summary protocal buffer，用于可视化
         with tf.name_scope('stddev'):
             stddev = tf.sqrt(tf.reduce_mean(tf.square(var - mean)))
         tf.summary.scalar('stddev', stddev)
         tf.summary.scalar('max', tf.reduce_max(var))
         tf.summary.scalar('min', tf.reduce_min(var))
         tf.summary.histogram('histogram', var) # 用于显示直方图信息

 # 创建MLP多层神经网络
 def nn_layer(input_tensor, input_dim, output_dim, layer_name, act=tf.nn.relu):
     with tf.name_scope(layer_name):
         with tf.name_scope('weights'):
             weights = weight_variable([input_dim, output_dim])
             variable_summaries(weights)
         with tf.name_scope('biases'):
             biases = bias_variable([output_dim])
             variable_summaries(biases)
         with tf.name_scope('Wx_plus_b'):
             preactivate = tf.matmul(input_tensor, weights) + biases # 矩阵乘
             tf.summary.histogram('pre_activations', preactivate)
         activations = act(preactivate, name='activation') # 激活函数relu
         tf.summary.histogram('activations', activations)
         return activations

 hidden1 = nn_layer(x, 784, 500, 'layer1')

 with tf.name_scope('dropout'):
     keep_prob = tf.placeholder(tf.float32)
     tf.summary.scalar('dropout_keep_probability', keep_prob)
     dropped  = tf.nn.dropout(hidden1, keep_prob) # 训练过程中随机舍弃部分神经元，为了防止或减轻过拟合

 y = nn_layer(dropped, 500, 10, 'layer2', act=tf.identity)

 with tf.name_scope('scross_entropy'):
     diff = tf.nn.softmax_cross_entropy_with_logits(logits=y, labels=y_) # 求交叉熵
     with tf.name_scope('total'):
         cross_entropy  = tf.reduce_mean(diff)
 tf.summary.scalar('cross_entropy', cross_entropy)

 with tf.name_scope('train'):
     train_step = tf.train.AdamOptimizer(learning_rate).minimize(cross_entropy) #Adam优化算法,全局最优,引入了二次方梯度矫正
 with tf.name_scope('accuracy'):
     with tf.name_scope('accuracy'):
         correct_prediction = tf.equal(tf.argmax(y, 1), tf.argmax(y_, 1))
     with tf.name_scope('accuracy'):
         accuracy = tf.reduce_mean(tf.cast(correct_prediction, tf.float32))

 tf.summary.scalar('accuracy', accuracy)

 merged = tf.summary.merge_all() # 整合之前定义的所有summary,在此处才开始执行，summary为延迟加载
 train_writer = tf.summary.FileWriter(log_dir + '/train', sess.graph)
 test_writer = tf.summary.FileWriter(log_dir + '/test') # 可视化数据存储在日志文件中
 tf.global_variables_initializer().run()

 def feed_dict(train):
     if train:
         xs, ys = mnist.train.next_batch(100)
         k = dropout
     else:
         xs, ys = mnist.test.images, mnist.test.labels
         k = 1.0
     return {x:xs, y_:ys, keep_prob:k}

 saver = tf.train.Saver()
 for i in range(max_steps):
     if i % 100 == 0:
         summary, acc = sess.run([merged, accuracy], feed_dict(False))
         test_writer.add_summary(summary, i)
         print('Accuray at step %s:%s' % (i, acc))
     else:
         if i % 100 == 99:
             run_options = tf.RunOptions(trace_level=tf.RunOptions.FULL_TRACE) # 定义TensorFlow运行选项
             run_metadata = tf.RunMetadata() # 定义TensorFlow运行元信息，记录训练运行时间及内存占用等信息
             summary, _ = sess.run([merged, train_step], feed_dict=feed_dict(True))
             train_writer.add_run_metadata(run_metadata, 'stp%03d' % i)
             train_writer.add_summary(summary, 1)
             saver.save(sess, log_dir + 'model.ckpt', i)
         else:
             summary, _ = sess.run([merged,train_step], feed_dict=feed_dict(True))
             train_writer.add_summary(summary, i)

 train_writer.close()
 test_writer.close()

　　程序执行完成后，在dos命令窗口或linux窗口运行命令，参数logdir是你程序保存log日志设置的地址。

效果如下：

结果：

　　

　　

　　一层神经网络：

　　

　　二层神经网络：

　　

　　神经网络计算图：

　　

　　神经元输出的分布：

　　　　

　　数据分布直方图：
　　　　　　

　　数据可视化：