如何抑制来自Tensorflow的所有签名警告?

Question

我得到一个警告，我找不到解决方案。显然这:

import os
os.environ['TF_CPP_MIN_LOG_LEVEL'] = '2'  # or any {'0', '1', '2'}
import tensorflow as tf
os.environ['AUTOGRAPH_VERBOSITY'] = '1'

不足以阻止这个烦人的消息:

WARNING:tensorflow:AutoGraph could not transform <function Model.make_train_function.<locals>.train_function at 0x00000281A55264C8> and will run it as-is.
Please report this to the TensorFlow team. When filing the bug, set the verbosity to 10 (on Linux, `export AUTOGRAPH_VERBOSITY=10`) and attach the full output.
Cause: Bad argument number for Name: 4, expecting 3
To silence this warning, decorate the function with @tf.autograph.experimental.do_not_convert

在每个纪元之后弹出。

是的，我已经用@tf.autograph.experimental.do_not_convert装饰了我的每个函数，消息仍然弹出。我目前正在使用Tensorflow 2.2.0

如果你知道为什么会这样，那就太好了。我不知道什么名字。

Answer 1

试试这个:

tf.autograph.set_verbosity(0)

如果不工作，也许logging模块可以在这里提供帮助:

import logging
logging.getLogger("tensorflow").setLevel(logging.ERROR) # this goes *before* tf import
import tensorflow as tf

感谢@Shahrokh Bah的评论。