in , , ,

How To Install PyTorch YOLOv5 With CUDA On Ubuntu 20.04


Today we will try to build our environment to host PyTorch YOLOv5 You Only Look Once The most famous real-time object detection algorithm library with the Nvidia CUDA Driver support

We will need to do the following list

  • Update Our Ubuntu 20.04
  • Install Python And PIP
  • Nvidia CUDA Driver with CUDA Toolkit version 11.3 for GPU Computing
  • Nvidia cuDNN The Deep Nural Network Library version libcudnn8_8.2.0.53-1+cuda11.3
  • Install Pytorch version 1.11.0 stable
  • Install YOLOv5

Step1: Update Ubuntu 20.04

# apt update
# apt upgrade

Step2: Install Python PIP AND Requirements

# apt install python3 python-is-python3 python3-dev pip virtualenv
# apt install python3-pil
# apt install ffmpeg libsm6 libxext6

Step3: Install And Check Prerequests

Pre-Requests Packages

Now we need to make sure that all Linux required headers are installed.

# apt install linux-headers-$(uname -r) -y

Check If Build-Essential development libraries and tools are installed

# apt install build-essential

Detect GPU CUDA Support

Check your GPU Graphic Card for the CUDA Enabled feature, the compute capability listed here for Nvidia CUDA GPUs Graphic Cards.

# lspci | grep -i nvidia

If (you found your GPU Card is CUDA Supported) { continue applying the following steps } else { Skip CUDA and cuDNN Installation Steps }.

Install Nvidia CUDA Driver Official Repository

According to the requirements version we will use the Nvidia CUDA Driver repository for Ubuntu 18.04 LTS, It's OK and compatible with Ubuntu 20.04 LTS

# wget -O /etc/apt/preferences.d/cuda-repository-pin-600
# apt-key adv --fetch-keys
# add-apt-repository "deb /"
# apt-get update

Check PyTorch Requirements

When looking for the stable version of PyTorch GPU CUDA Support requirements, it's the CUDA 11.33, while YOLOv5 we need to install Python>=3.8 environments with PyTorch>=1.7, plus a list of the following requirement packages

Install PyTorch And YOLOv5

Step4: Install CUDA 11.3 And cuDNN 8.2

You can detect and select the correct version from the Nvidia official repository URL of the packages list.

# apt install --no-install-recommends cuda=11.3.0-1
# apt install --no-install-recommends libcudnn8= libcudnn8-dev=
# echo 'export PATH=/usr/local/cuda/bin${PATH:+:${PATH}}' >> ~/.bashrc

Step5: Install PyTorch

Install PyTorch With GPU CUDA Support

# pip install --no-cache-dir torch torchvision torchaudio --extra-index-url

Install PyTorch CPU Only

# pip install --no-cache-dir torch torchvision torchaudio

Check Your Installation

# python
Python 3.8.10 (default, Mar 15 2022, 12:22:08)
[GCC 9.4.0] on linux
Type "help", "copyright", "credits" or "license" for more information.
>>> import torch
>>> torch.__version__

Step6: Install YOLOv5

# git clone
# cd yolov5
# pip install --no-cache-dir -r requirements.txt

And you can start with YOLOv5 your hello world example at and

What do you think?

Leave a Reply

Your email address will not be published. Required fields are marked *

      How Do I Buy Ethereum

      How Do I Buy Ethereum?

      Nvidia GPU CUDA Share With Docker Container

      How To Share Your Host GPU With Docker Container