linux环境下pytorch（GPU版本）的安装

1.安装anaconda

https://repo.anaconda.com/archive/
bash Anaconda3-2023.09-0-Linux-x86_64.sh

2.配置环境

xbgydx@ubuntu:~$ nano ~/.bashrc
xbgydx@ubuntu:~$ export PATH="/home/xbgydx/anaconda3/bin:$PATH"
按 Ctrl + X，然后按 Y 保存更改，最后按 Enter 退出。

source ~/.bashrc

anaconda -V（注意a要小写，V要大写）；
conda -V ,显示版本信息

conda init bash
conda activate pytorch
 

3.安装pytorch

conda create -n pytorch python=3.9
conda info --envs

nvcc -V
sudo apt install nvidia-cuda-toolkit

可选：

~/.bashrc

export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
source ~/.bashrc

pip install torch==1.8.1+cu101 torchvision==0.9.1+cu101 torchaudio==0.8.1 -f https://download.pytorch.org/whl/torch_stable.html  -i  https://pypi.tuna.tsinghua.edu.cn/simple

4.检测是否成功

python
import torch
print(torch.__version__)  # torch版本
print(torch.cuda.is_available())   # 能否调用cuda Ture/False