tensorboard展示网络结构

代码

import tensorflow as tffrom tensorflow.examples.tutorials.mnist import input_datatf.compat.v1.disable_eager_execution()#载入数据集mnist=input_data.read_data_sets(\"MNIST_data\",one_hot=True)#每个批次大小batch_size=100#计算一共有多少个批次n_bath=mnist.train.num_examples // batch_sizeprint(n_bath)with tf.name_scope(\'input\'):#定义两个placeholderx=tf.compat.v1.placeholder(tf.float32,[None,784],name=\'x-input\')y=tf.compat.v1.placeholder(tf.float32,[None,10],name=\'y-input\')with tf.name_scope(\'layer\'):#创建一个简单的神经网络with tf.name_scope(\'wights\'):W=tf.Variable(tf.zeros([784,10]),name=\'W\')with tf.name_scope(\'biases\'):b=tf.Variable(tf.zeros([10]),name=\'b\')with tf.name_scope(\'wx_plus_b\'):wx_plus_b=tf.matmul(x,W)+bwith 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))with tf.name_scope(\'train\'):#梯度下降train_step=tf.compat.v1.train.GradientDescentOptimizer(0.2).minimize(loss)#初始化变量init=tf.compat.v1.global_variables_initializer()with tf.name_scope(\'accuracy\'):with tf.name_scope(\'correct_prediction\'):#结果存放在一个布尔型列表中#返回的是一系列的True或False argmax返回一维张量中最大的值所在的位置，对比两个最大位置是否一致correct_prediction=tf.equal(tf.argmax(y,1),tf.argmax(prediction,1))with tf.name_scope(\'accuracy\'):#求准确率#cast：将布尔类型转换为float，将True为1.0，False为0，然后求平均值accuracy=tf.reduce_mean(tf.cast(correct_prediction,tf.float32))with tf.compat.v1.Session() as sess:sess.run(init)writer=tf.compat.v1.summary.FileWriter(\'logs/\',sess.graph)for epoch in range(1):for batch in range(n_bath):#获得一批次的数据，batch_xs为图片，batch_ys为图片标签batch_xs,batch_ys=mnist.train.next_batch(batch_size)#进行训练sess.run(train_step,feed_dict={x:batch_xs,y:batch_ys})#训练完一遍后，测试下准确率的变化acc=sess.run(accuracy,feed_dict={x:mnist.test.images,y:mnist.test.labels})print(\"Iter \"+str(epoch)+\",Testing Accuracy \"+str(acc))

会生成logs/目录，并且目录下的文件我们需要这样子打开