pytorch常见错误
- [1 RuntimeError: CUDA error: no kernel image is available for execution on the device。](#1 RuntimeError: CUDA error: no kernel image is available for execution on the device。)
1 RuntimeError: CUDA error: no kernel image is available for execution on the device。
报错信息:安装带gpu的pytorch的时候,
(1)先警告
NVIDIA GeForce RTX 3090 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 sm_75.
If you want to use the NVIDIA GeForce RTX 3090 GPU with PyTorch, please check the instructions at https://pytorch.org/get-started/locally/
(2)紧接着报错执行深度学习代码报错:
RuntimeError: CUDA error: no kernel image is available for execution on the device。
原因: 是因为pytorch1.6的版本不支持gpu硬件RTX 3090,需要下载更高版本的pytorch才行。
解决: 下载1.8以上版本的pytorch,算力要支持
cpp
// 查看包支持的算力
import torch
torch.cuda.is_available()
// 上面返回true
torch.cuda.get_arch_list()
幸福的烦恼:显卡算力太高而pytorch版本太低不支持_cuda版本太高 pytorch不支持_是Yu欸的博客-CSDN博客