目录
0.目标环境
windows+ubuntu共存
nvidia driver
cuda
nvidia-docker
pycharm
chrome
python3.8
pip1.安装双系统
由于本机上是有自带的windows系统的，所以，这里直接使用u盘安装ubuntu系统。
1）制作ubuntu20.04的u盘
UltralSo+官网下载系统+u盘
2）安装
https://blog.csdn.net/zr459927180/article/details/51627910
安装时分辨率无法调整，看不到分区选项时：https://www.zhihu.com/question/299230727
2.安装chrome,输入法
chrome:https://blog.csdn.net/snowdream86/article/details/106160498
输入法:
https://pinyin.sogou.com/linux/?r=pinyin
https://www.cnblogs.com/Areas/p/13438171.html
https://zhuanlan.zhihu.com/p/142206571
3.安装gcc
https://blog.csdn.net/m0_37412775/article/details/109355044
sudo apt-get update
sudo apt-get upgrade
sudo apt-get install build-essential
sudo apt-get -y install gcc-7 g++-7change gcc version:https://linuxconfig.org/how-to-switch-between-multiple-gcc-and-g-compiler-versions-on-ubuntu-20-04-lts-focal-fossa
4.安装显卡驱动与高版本cuda
cp cuda/lib64/* /usr/local/cuda-11.1/lib64/
cp cuda/include/* /usr/local/cuda-11.1/include/
1）下载安装显卡驱动
https://blog.csdn.net/Thanlon/article/details/106125738
2）安装cuda-11.1.0
https://blog.csdn.net/qq_36999834/article/details/107589779
3）安装cudnn
wget https://developer.download.nvidia.com/compute/cuda/11.1.0/local_installers/cuda_11.1.0_455.23.05_linux.run
sudo sh cuda_11.1.0_455.23.05_linux.run参考：
https://blog.csdn.net/qq_33200967/article/details/80689543
5.安装pycharm
https://www.jetbrains.com/pycharm/download/other.html
https://blog.csdn.net/feimeng116/article/details/105837483
cp pycharm-professional-2019.3.5.tar.gz /home/xxx/software
tar -zxvf pycharm-professional-2019.3.5.tar.gz -C ./software/pycharm-2019.3/6.安装docker与nvidia-docker
https://docs.docker.com/engine/install/ubuntu/
https://docs.nvidia.com/datacenter/cloud-native/container-toolkit/install-guide.html#nvidia-drivers
sudo apt-get update
sudo apt-get install -y apt-transport-https ca-certificates curl gnupg-agent software-properties-common
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add -
sudo apt-key fingerprint 0EBFCD88
sudo add-apt-repository "deb [arch=amd64] https://download.docker.com/linux/ubuntu \
bionic \
stable"
sudo apt-get update
sudo apt-get install -y docker-ce docker-ce-cli containerd.io
sudo docker run hello-world
sudo usermod -aG docker gmt
sudo systemctl restart docker
sudo systemctl status docker
docker --version
curl https://get.docker.com | sh && sudo systemctl --now enable docker
distribution=$(. /etc/os-release;echo $ID$VERSION_ID) && curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - && curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
sudo apt-get update
sudo apt-get install -y nvidia-docker2
sudo systemctl restart docker导入自己的之前的nvidia-docker 镜像，并创建一个容器：
sudo docker load -i tensorflow.tar
sudo docker images # 查看导入的images
sudo sh ./tf_dep/debug.sh # 创建新容器
# debug.sh
#!/bin/bash
nvidia-docker run -m 8GB -it --net=host \
-v /home/***/tensorflow/:/root \
--name tensorflow tensorflow:v2 /bin/bash
7.联想新机安装显卡驱动
尝试过以下，nvidia-smi都无法找到显卡驱动：
1）ubuntu16.04、ubuntu18.04、ubuntu20.04
2）不同内核版本5.4与5.8
3）使用dmesg查看过kernel的log
4）显卡驱动安装方式（图像界面安装、终端安装），高低版本
5）安装完不同版本的cuda。
最后成功步骤：
1）将Ctrl+Alt+F1，进入bios，修改secure boot为disabled；
2）启动项中ubuntu选项，以5.4.0 kernel进入系统（此处是需要自己去安装内核的）；
3）安装gcc 7，g++ 7；
4）卸载之前安装的驱动，autoinstall
命令如下：
sudo apt-get update
sudo apt-get upgrade
sudo apt-get install build-essential
sudo apt-get -y install gcc-7 g++-7
sudo apt-get remove --purge nvidia*
sudo ubuntu-drivers autoinstall
sudo reboot
注：这里可能是我安装的cuda版本太多了，与前面不太一致。将就看，差别不大。
8.安装pip
ubuntu系统自带的python3.8好像没有pip，这里应该需要自己安装了。对于使用python的，这个必须要有。
sudo apt install -y python3-pip
参考：