python - 张量板中没有标量数据

Question

我一直在阅读有关标量的张量板文档，但在张量板中呈现它时遇到问题。

我在 Windows 10 中安装了 pip install tensorflow

我的代码如下所示:

import tensorflow as tf

a = tf.constant(7, name='test_variable')
tf.summary.scalar('variable', a)

with tf.Session() as sess:
    tf.summary.FileWriter('my_folder', graph=sess.graph)
    X = tf.global_variables_initializer()
    sess.run(X)

我看到my_folder中有一个文件

在命令提示符中:tensorboard --logdir=my_folder --port 6006

输出:

C:\Users\MM>tensorboard --logdir=my_folder --port 6006
Starting TensorBoard b'54' at http://DESKTOP-9S2D9VF:6006
(Press CTRL+C to quit)

当我打开浏览器时，我得到:

No scalar data was found. 
Probable causes: etc. etc.

Answer 1

您需要运行summary_op并将结果传递给FileWriter。例如:

import tensorflow as tf

a = tf.constant(7, name='test_variable')
tf.summary.scalar('variable', a)
summary_op = tf.summary.merge_all()

with tf.Session() as sess:
    summary_writer = tf.summary.FileWriter('/tmp/summary', graph=sess.graph)
    X = tf.global_variables_initializer()
    sess.run(X)
    summary = sess.run(summary_op)
    summary_writer.add_summary(summary)

来自官方documentation :

Then, you can just run the merged summary op, which will generate a serialized Summary protobuf object with all of your summary data at a given step. Finally, to write this summary data to disk, pass the summary protobuf to a tf.summary.FileWriter.