WSL2中TensorFlow无法识别CUDA驱动问题求助

运行TensorFlow时出现以下错误，无法启用GPU加速：

>>> import tensorflow as tf
2023-04-19 17:13:12.944168: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.
2023-04-19 17:13:13.056290: I tensorflow/tsl/cuda/cudart_stub.cc:28] Could not find cuda drivers on your machine, GPU will not be used.
2023-04-19 17:13:13.057124: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.
To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.
2023-04-19 17:13:14.369949: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT
>>> tf.__version__
'2.12.0'

环境配置信息 #

• CUDA版本：

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2023 NVIDIA Corporation
Built on Tue_Feb__7_19:32:13_PST_2023
Cuda compilation tools, release 12.1, V12.1.66
Build cuda_12.1.r12.1/compiler.32415258_0

• GPU驱动信息：

+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 530.46                 Driver Version: 531.61       CUDA Version: 12.1     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                  Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf            Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  NVIDIA GeForce GTX 1650         On | 00000000:01:00.0 Off |                  N/A |
| N/A   44C    P8                4W /  N/A|      0MiB /  4096MiB |      0%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
|    0   N/A  N/A        24      G   /Xwayland                                 N/A      |
+---------------------------------------------------------------------------------------+

• PATH环境变量：

/usr/local/cuda/bin:/usr/local/cuda-12.1/bin

• LD_LIBRARY_PATH环境变量：

/usr/local/cuda/include:/usr/local/cuda/lib64:/usr/local/cuda-12.1/lib64:/usr/local/cuda/extras/CUPTI/lib64

已安装TensorFlow、CUDA驱动及cuDNN，配置相关环境变量后仍无法识别CUDA驱动，求解决方案。

解决方案 #

1. 匹配版本兼容性 #

TensorFlow 2.12.0官方支持的CUDA版本为11.8，当前使用的CUDA 12.1不在兼容列表内，这是核心问题。降级CUDA到11.8版本，确保与TensorFlow版本匹配。

2. 修正WSL2驱动安装逻辑 #

WSL2不需要在Linux子系统内安装NVIDIA驱动，驱动必须安装在Windows主机上，且需为WSL专用驱动（非普通桌面驱动），版本需适配CUDA 11.8（对应驱动版本≥522.06）。

3. 调整环境变量 #

• LD_LIBRARY_PATH中不应包含头文件路径/usr/local/cuda/include，修正为：

  export LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64:$LD_LIBRARY_PATH
  

• 确保/usr/local/cuda是CUDA 11.8的软链接，通过ls -l /usr/local/cuda查看指向，若指向12.1，重新创建软链接：

  sudo rm /usr/local/cuda
  sudo ln -s /usr/local/cuda-11.8 /usr/local/cuda
  

4. 验证cuDNN配置 #

安装与CUDA 11.8兼容的cuDNN版本（推荐cuDNN 8.6.0），将cuDNN文件复制到CUDA目录：

sudo cp cudnn-linux-x86_64-8.6.0.163_cuda11-archive/lib/libcudnn* /usr/local/cuda/lib64/
sudo cp cudnn-linux-x86_64-8.6.0.163_cuda11-archive/include/cudnn*.h /usr/local/cuda/include/
sudo chmod a+r /usr/local/cuda/lib64/libcudnn* /usr/local/cuda/include/cudnn*.h

5. 重新安装TensorFlow #

卸载现有TensorFlow后，安装对应版本：

pip uninstall tensorflow
pip install tensorflow==2.12.0

6. 验证GPU可用性 #

完成配置后，运行以下代码确认：

import tensorflow as tf
print(tf.config.list_physical_devices('GPU'))

输出包含GPU设备信息则配置成功。

内容的提问来源于stack exchange，提问作者Helios