Google Colab 加载运行时 CuDNN 库：8.0.5 但源代码编译时使用：8.1.0
Google Colab Loaded runtime CuDNN library: 8.0.5 but source was compiled with: 8.1.0
在 Google Collab 中使用 Tensorflow 2 对象检测 API 进行训练时，出现以下错误：
2022-03-07 15:27:51.475298: E tensorflow/stream_executor/cuda/cuda_dnn.cc:359] Loaded runtime CuDNN library: 8.0.5 but source was compiled with: 8.1.0. CuDNN library needs to have matching major version and equal or higher minor version. If using a binary install, upgrade your CuDNN library. If building from sources, make sure the library loaded at runtime is compatible with the version specified during compile configuration.
2022-03-07 15:27:51.477142: W tensorflow/core/framework/op_kernel.cc:1745] OP_REQUIRES failed at conv_ops.cc:1120 : UNKNOWN: 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.
我尝试在 Google collab 中更新我的 CuDNN 库，使用这个 Whosebug 答案：
Loaded runtime CuDNN library: 8.0.5 but source was compiled with: 8.1.0. when using google colab。
但是，这并不能解决问题。
我正在使用 Tensorflow 2.7.0。
以下是问题的解决方法，如官方发布的Github
这可能并不总是有效，也可能不安全，但作为暂时的解决方法，您可以尝试 运行 !apt install --allow-change-held-packages libcudnn8=8.1.0.77-1+cuda11.2
在 Google Collab 中使用 Tensorflow 2 对象检测 API 进行训练时，出现以下错误：
2022-03-07 15:27:51.475298: E tensorflow/stream_executor/cuda/cuda_dnn.cc:359] Loaded runtime CuDNN library: 8.0.5 but source was compiled with: 8.1.0. CuDNN library needs to have matching major version and equal or higher minor version. If using a binary install, upgrade your CuDNN library. If building from sources, make sure the library loaded at runtime is compatible with the version specified during compile configuration.
2022-03-07 15:27:51.477142: W tensorflow/core/framework/op_kernel.cc:1745] OP_REQUIRES failed at conv_ops.cc:1120 : UNKNOWN: 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.
我尝试在 Google collab 中更新我的 CuDNN 库，使用这个 Whosebug 答案： Loaded runtime CuDNN library: 8.0.5 but source was compiled with: 8.1.0. when using google colab。 但是，这并不能解决问题。
我正在使用 Tensorflow 2.7.0。
以下是问题的解决方法，如官方发布的Github
这可能并不总是有效，也可能不安全，但作为暂时的解决方法，您可以尝试 运行 !apt install --allow-change-held-packages libcudnn8=8.1.0.77-1+cuda11.2