tensorflow - tensorflow不使用gpu错误无法加载动态库'libcudnn.so.8'
问题描述
我无法解决这个问题:
我有 GTX-1050。
(base) (venv) wojtek@wojtek-GF63-8RC:~/_mgr/Lenet$ python train_network.py --dataset images --model 90_180_270_plus_originalne
2021-05-04 02:29:38.383821: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:937] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2021-05-04 02:29:38.389729: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudnn.so.8'; dlerror: libcudnn.so.8: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda-11.2/lib64:/usr/local/cuda/lib64::/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64
2021-05-04 02:29:38.389752: W tensorflow/core/common_runtime/gpu/gpu_device.cc:1765] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.
Skipping registering GPU devices...
Num GPUs Available: 0
英伟达 smi 输出:
(base) (venv) wojtek@wojtek-GF63-8RC:~/_mgr/Lenet$ nvidia-smi
Tue May 4 02:32:09 2021
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 465.19.01 Driver Version: 465.19.01 CUDA Version: 11.3 |
|-------------------------------+----------------------+----------------------+
| 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 ... On | 00000000:01:00.0 Off | N/A |
| N/A 74C P0 N/A / N/A | 488MiB / 4042MiB | 5% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=============================================================================|
| 0 N/A N/A 1211 G /usr/lib/xorg/Xorg 47MiB |
| 0 N/A N/A 1371 G /usr/bin/gnome-shell 88MiB |
| 0 N/A N/A 1674 G /usr/lib/xorg/Xorg 309MiB |
| 0 N/A N/A 1852 G /usr/bin/gnome-shell 16MiB |
| 0 N/A N/A 2206 G ...AAAAAAAA== --shared-files 8MiB |
| 0 N/A N/A 2306 G ...gAAAAAAAAA --shared-files 9MiB |
| 0 N/A N/A 5615 G /usr/lib/firefox/firefox 1MiB |
| 0 N/A N/A 6433 G /usr/lib/firefox/firefox 1MiB |
| 0 N/A N/A 6690 G /usr/lib/firefox/firefox 1MiB |
(base) (venv) wojtek@wojtek-GF63-8RC:~/_mgr/Lenet$ nvcc -V
nvcc:NVIDIA (R) Cuda 编译器驱动程序 版权所有 (c) 2005-2020 NVIDIA Corporation 构建于 Mon_Nov_30_19:08:53_PST_2020 Cuda 编译工具,版本 11.2,V11.2.67 构建 cuda_11.2.r11.2/compiler.29373293_0
解决方案
推荐阅读
- symfony - 无法连接到主机,elasticsearch 关闭?Symfony 4 elastica/rulfin 客户端。码头工人
- c# - unity2D - 填充量边缘指示器
- java - 当我从移动 onepluse 3 Android oreo 的下载文件夹中选择文件/图像/.doc 时,Android 应用程序崩溃
- arrays - 为 Alamofire POST 请求 Swift 4 创建主体
- c++ - 如何打印没有空格作为输入的整数?
- subquery - SQL Oracle 子查询
- linux - 检查 Linux 包然后跳过
- python - Keras load_model 和 get_default_graph.finalize()
- javascript - 当孩子集中注意力时,焦点控制下拉菜单会关闭
- unity3d - 用触摸向各个方向移动玩家?,Unity