ai-benchmark运行时报错libcudart.so.11.0解决方法

系统ubuntu server 20.4

tensorflow版本:2.6

pytho版本： 3.6

报错：

[root@localhost ~]# ai-benchmark
2021-12-21 16:41:15.203709: W tensorflow/stream_executor/platform/default/dso_lo                                     ader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcuda                                     rt.so.11.0: cannot open shared object file: No such file or directory
2021-12-21 16:41:15.203814: I tensorflow/stream_executor/cuda/cudart_stub.cc:29]                                      Ignore above cudart dlerror if you do not have a GPU set up on your machine.

>>   AI-Benchmark-v.0.1.2
>>   Let the AI Games begin..

2021-12-21 16:41:26.206482: I tensorflow/core/platform/cpu_feature_guard.cc:142]                                      This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (on                                     eDNN) to use the following CPU instructions in performance-critical operations:                                       AVX2 FMA
To enable them in other operations, rebuild TensorFlow with the appropriate comp                                     iler flags.
2021-12-21 16:41:26.516635: E tensorflow/stream_executor/cuda/cuda_driver.cc:271                                     ] failed call to cuInit: CUDA_ERROR_NO_DEVICE: no CUDA-capable device is detecte                                     d
2021-12-21 16:41:26.516671: I tensorflow/stream_executor/cuda/cuda_diagnostics.c                                     c:156] kernel driver does not appear to be running on this host (localhost.local                                     domain): /proc/driver/nvidia/version does not exist
*  TF Version: 2.6.0
*  Platform: Linux-3.10.0-1160.49.1.el7.x86_64-x86_64-with-centos-7.9.2009-Core
*  CPU: N/A
*  CPU RAM: 16 GB
解决方法：安装cuda工具，详细参见解决Could not load dynamic library ‘libcudart.so.11.0‘； dlerror: libcudart.so.11.0妙招_xiaopang_love_study的博客-CSDN博客_libcudart.so.11.0