Ubuntu Setup

0. Delete Ubuntu

双系统删除ubuntu

0.1 Open Diskpart

• win+R: diskpart

0.2 Commands in Diskpart

• Check available disks:

  list disk
  

• Select the disk where Ubuntu is installed (e.g., Disk 1):

  select disk <disk_number>
  

• Check partitions on the selected disk:

  list partition
  

• Select the partitio(System)n containing Ubuntu system files (e.g., Partition 1):

  select partition 1
  

• Assign a drive letter (P:):

  assign letter=P
  

0.3 Delete Ubuntu System Files

• Run Notepad as Administrator.
• Navigate to P: in File Explorer and delete the Ubuntu folder in EFI folder.

0.4. Remove Assigned Drive Letter

• Return to Diskpart and run the following command:

  remove letter=P
  

0.5 Delete Ubuntu Partition Using Disk Management

• Open Disk Management (Win + X → Disk Management).
• Locate the partition containing Ubuntu.
• Right-click and select “Delete Volume”.
• (Optional) Extend another partition into the unallocated space.

1. Install Ubuntu

• Rufus
• Ubuntu Download

2. Install Anaconda3

2.1 Download

Anaconda3

cd Downloads

bash Anaconda3-xxxx.xx-Linux-x86_64.sh

2.2 bashrc setting

nano ~/.bashrc

• export

export PATH="/home/username/anaconda3/bin:$PATH"

source ~/.bashrc

---

2.3 Command

conda create -n your_env_name python=X.X（3.6,3.7,etc.）

source activate your_env_name(env_name)

source deactivate your_env_name(env_name)

conda remove -n your_env_name(env_name) --all

conda list

conda install package_name(package_name)
conda install scrapy==1.3 
conda install -n env_name package_name

#Check env
conda env list 
#or
conda info -e
#or
conda info --envs

conda update conda

conda update anaconda

conda update --all

conda update python

2.4 put images source

3.Install Pycharm

Pycharm

4.ROS

• ROS2 - humble - ubuntu 22.04
• ROS1 - ubuntu 20.04

5.Install Nvidia driver

ubuntu-drivers devices

sudo apt-get install nvidia-driver-535

• hardware key

Enroll MOK 

• reboot and check

nvidia-smi

---

6.CUDA

CUDA

• nano ~/.bashrc

  export PATH=/usr/local/cuda-12.3/bin${PATH:+:${PATH}}
  export LD_LIBRARY_PATH=/usr/local/cuda-12.3/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
  

---

7.cuDNN

cudnn

• Extract

cd ~/Downloads/cudnn-xxxxxxxxxxxxx`

• move

sudo cp include/cudnn*.h /usr/local/cuda/include 
sudo cp -P lib/libcudnn* /usr/local/cuda/lib64 
sudo chmod a+r /usr/local/cuda/include/cudnn*.h /usr/local/cuda/lib64/libcudnn*

• Check

cat /usr/local/cuda-12.3/include/cudnn_version.h | grep CUDNN_MAJOR -A 2

---

8.Pytorch

https://pytorch.org/get-started/previous-versions/

• Check

>>import torch

>>print(torch.__version__)

>>print(torch.version.cuda)

>>print(torch.backends.cudnn.version())

>>torch.cuda.is_available()

>>torch.cuda.device_count()

>>torch.cuda.get_device_name(0)

>>torch.cuda.current_device()

---

9.Docker

Do not install docker-desktop!

• Install using the apt repository

---