关于如何使用tensorboard

一、在终端中输入

tensorboard --logdir runs\train(file directory)，当前目录的文件相对地址，比如runs/trains，注意这里只能是在tensorboard的存储event文件的父目录

它会跳出

 点击网址即可使用

二、

（1） 在tensorflow中的用法：

# Create a callback ，初始化tensorboard
tf_callback = tf.keras.callbacks.TensorBoard(log_dir="./logs") 
 
# Pass in the callback when fitting the model ，显示调用tensorboard
model.fit(X_train, y_train, epochs=5, callbacks=[tf_callback])

这里附上完整使用代码：

import tensorflow as tf

fashion_mnist = tf.keras.datasets.fashion_mnist
(train_images, train_labels), (test_images, test_labels) = fashion_mnist.load_data()

train_images = tf.expand_dims(train_images, axis=3)
test_images = tf.expand_dims(test_images, axis=3)

train_images = train_images[:1000]
train_labels = train_labels[:1000]
test_images = test_images[:100]
test_labels = test_labels[:100]


class FasionClassic:
    def __init__(self):
        self.cnn_1 = tf.keras.layers.Conv2D(32, 3, activation=tf.nn.relu)
        self.batch_norm_1 = tf.keras.layers.BatchNormalization()

        self.cnn_2 = tf.keras.layers.Conv2D(64, 3, activation=tf.nn.relu)
        self.batch_norm_2 = tf.keras.layers.BatchNormalizati