NVIDIA GeForce RTX 3050 Laptop GPU with CUDA capability sm_86 is not compatible with the current PyTorch installation.
The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70
root@7cd0ba7d49bc:/app
Python 3.8.10 (default, Jun 22 2022, 20:18:18)
[GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> torch.__version__
'1.12.1+cu102'
>>> print (torch.cuda.is_available());
True
leon@leon:~$ nvidia-smi
Wed Sep 14 15:55:07 2022
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.141.03 Driver Version: 470.141.03 CUDA Version: 11.4 |
|-------------------------------+----------------------+----------------------+
| 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 ... Off | 00000000:01:00.0 On | N/A |
| N/A 39C P8 6W / N/A | 726MiB / 3902MiB | 3% Default |
| | | N/A |
+-------------------------------+----------------------+----------------------+
解决方法:
https://pytorch.org/get-started/locally/#supported-linux-distributions
选择合适的版本,然后进入镜像,卸载原来的torch, 然后用pip重新安装
可以把修复的容器保存为镜像。
接下来我们看一下pytorch对应版本是否有问题:
https://pytorch.org/get-started/previous-versions/
我们可以看到cuda版本如果是11.1,那么所对应torch的版本应该是v1.8.0,但是我们刚才查看版本是1.6.0。接下来就简单啦。升级我们的pytorch。
复制上图对应11.1的命令到cmd又出现如下报错:
torchvision 0.7.0+cu101 has requirement torch1.6.0, but you’ll have torch 1.8.0 which is incompatible.
那我们就单独再download一遍:
pip install –upgrade torchvision0.9.0
报错如下:
Could not install packages due to an EnvironmentError: [WinError 5] 拒绝访问。: ‘d:\anaconda3.7\anaconda\lib\site-packages\torch_c.cp37-win_amd64.pyd’
Consider using the –user option or check the permissions.
参考:
https://blog.csdn.net/Xunuo1995/article/details/115261069
Original: https://blog.csdn.net/shelutai/article/details/126853699
Author: shelutai
Title: NVIDIA GeForce RTX 3050 Laptop GPU with CUDA capability sm_86 is not compatible with current PyTorch
原创文章受到原创版权保护。转载请注明出处:https://www.johngo689.com/708741/
转载文章受原作者版权保护。转载请注明原作者出处!