This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.

tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
tensorflow.python.framework.errors_impl.UnknownError: Failed to get convolution algorithm. This is probably because cuDNN failed to initialize, so try looking to see if a warning log message was printed above.
[[node sequential/conv2d/Conv2D (defined at /code/tf_google_certify/chp1/conv.py:27) ]] [Op:__inference_train_function_675]

Function call stack:
train_function

解决方法

添加如下代码

from tensorflow.compat.v1 import ConfigProto
from tensorflow.compat.v1 import InteractiveSession

config = ConfigProto()
config.gpu_options.allow_growth = True
session = InteractiveSession(config=config)

原文链接

https://github.com/tensorflow/tensorflow/issues/25160#issuecomment-782764092

本栏目推荐文章
• org.springframework.kafka.listener.ListenerExecutionFailedException: Listener method could not be invoked with the incoming message
• 通过ssh在服务器上使用mujoco，报错Failed to initialize GLFW；ERROR: could not initialize GLFW
• message报错 crond postdrop maildrop permission denied
• Cannot re-initialize CUDA in forked subprocess.
• kafka：ERROR Shutdown broker because all log dirs（Windows）
• mysql安装：mysqld --initialize -console无反应解决方法（其一）
• vue3 v3-easyui messager消息框封装
• 编译内核 ，不能进入OS，卡在 load initial ramdisk
• 神经网络优化篇：详解神经网络的权重初始化（Weight Initialization for Deep NetworksVanishing / Exploding gradients）
• Element implicitly has an 'any' type because expression of type 'string' can't be used to index type

initialize probably because looking messageinitialize probably because looking look looking look-up look-up table look up generics custom first look for packages looking details probably probability ownership closer rust look