| linux安装nvidia | 您所在的位置：网站首页 › cuda102 升级到cuda110 › linux安装nvidia |
linux安装nvidia
|
linux安装nvidia_dali
查看cuda版本 (base) lthpc@lthpc:~$ nvcc -V nvcc: NVIDIA (R) Cuda compiler driver Copyright (c) 2005-2018 NVIDIA Corporation Built on Sat_Aug_25_21:08:01_CDT_2018 Cuda compilation tools, release 10.0, V10.0.130查看python版本 (base) lthpc@lthpc:~$ python3 -V Python 3.7.3如果linux主机可以联网 pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/cuda/10.0 nvidia-dali如果不能联网 进入下面网站https://developer.download.nvidia.com/compute/redist/cuda/10.0 选择对应版本
pip安装 (base) lthpc@lthpc:~$ cd /home/library (base) lthpc@lthpc:/home/library$ pip install nvidia_dali-0.21.0-1239036-cp37-cp37m-manylinux1_x86_64.whl --user (base) lthpc@lthpc:/home/library$ pip install nvidia_dali-0.21.0-1239036-cp37-cp37m-manylinux1_x86_64.whl --user Looking in indexes: https://pypi.tuna.tsinghua.edu.cn/simple Processing ./nvidia_dali-0.21.0-1239036-cp37-cp37m-manylinux1_x86_64.whl Requirement already satisfied: future in /opt/anaconda3-2019.3/lib/python3.7/site-packages (from nvidia-dali==0.21.0) (0.17.1) Installing collected packages: nvidia-dali Successfully installed nvidia-dali-0.21.0使用教程https://blog.csdn.net/minstyrain/article/details/99696557 代码https://gitee.com/lilujunai/Efficient_ImageNet_Classification?_from=gitee_search |
【本文地址】
公司简介
联系我们