Yolo-V4 Installation On Ubuntu 18.04

Hello friends, Installation of YOLOv4 on Ubuntu 18.04.

Pre-Requirements:

1. Python 3.6
2. Git
3. CMake >=3.12
4. CUDA >= 10.0
5. CuDNN 7.6.5
6. OpenCV 4.0.0
7. OpenCV Contrib (latest)
8. YOLOv4
9. Last but not least a good internet connection

Steps are as follows:

Step 1: Install Git

• sudo apt update
• sudo apt-get install git

Step 2: Install Nvidia Driver using following commands:-

• ubuntu-drivers devices
• then select the recommended drivers
• Example command => “sudo apt-get install nvidia-driver-460”

Step 3: Install CUDA 10.0

• link for cuda file(cuda-10.0)
• After Successful downloading all the files open terminal in that location where your file downloaded and type “sudo sh ‘CUDA filename’”
• First select “accept” for Eula options
• then select ’n’ / no for graphic installation
• then select “y” / yes for all others options
• press enter for default locations.
• copy all the lines and paste in bashrc file and save the file if this are not already available in your bashrc file.
• # NVIDIA CUDA Path
• export PATH=/usr/local/cuda-10.0/bin:$PATH
• export LD_LIBRARY_PATH=/usr/local/cuda-10.0/lib64

Step 4: Install CuDNN

• To install CuDNN you have to download 3 files are as follows:

1. cuDNN Runtime Library for Ubuntu18.04 (Deb)
2. cuDNN Developer Library for Ubuntu18.04 (Deb)
3. cuDNN Code Samples and User Guide for Ubuntu18.04 (Deb) (Not Mandatory)

• After Successful downloading all the files open terminal in that location where your file downloaded and type “sudo dpkg -i ‘cuDNN Runtime file name’ “ and press enter.
• then type same command with the 2nd file name is developer file (dev file).
• and also same command for the 3rd file name I.e document file(Sample file).
• After all, type “nvidia-smi” and press enter to be sure you have installed nvidia graphic driver.
• Now, type “nvcc — version” that will show you the CUDA version that installed in your system.

Step 5: Now time to install OpenCV by build, follow these commands

• sudo apt-get update
• sudo apt-get upgrade
• sudo apt-get install build-essential cmake unzip pkg-config
• sudo apt-get install libjpeg-dev libpng-dev libtiff-dev
• sudo apt-get install libjasper-dev
• sudo add-apt-repository “deb http://security.ubuntu.com/ubuntu xenial-security main”
• sudo apt update
• sudo apt install libjasper1 libjasper-dev
• sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev libv4l-dev
• sudo apt-get install libxvidcore-dev libx264-dev
• sudo apt-get install libgtk-3-dev
• sudo apt-get install libatlas-base-dev gfortran
• sudo apt-get install python3.6-dev
• wget -O opencv.zip https://github.com/opencv/opencv/archive/4.0.0.zip
• wget -O opencv_contrib.zip https://github.com/opencv/opencv_contrib/archive/4.0.0.zip
• unzip opencv.zip
• unzip opencv_contrib.zip
• mv opencv-4.0.0 opencv
• mv opencv_contrib-4.0.0 opencv_contrib
• wget https://bootstrap.pypa.io/get-pip.py
• sudo python3 get-pip.py
• sudo pip install virtualenv virtualenvwrapper
• sudo rm -rf ~/get-pip.py ~/.cache/pip
• echo -e “\n# virtualenv and virtualenvwrapper” >> ~/.bashrc
• echo “export WORKON_HOME=$HOME/.virtualenvs” >> ~/.bashrc
• echo “export VIRTUALENVWRAPPER_PYTHON=/usr/bin/python3” >> ~/.bashrc
• echo “source /usr/local/bin/virtualenvwrapper.sh” >> ~/.bashrc
• source ~/.bashrc
• mkvirtualenv cv -p python3
• workon cv
• pip install numpy
• cd opencv
• mkdir build
• cd build
• cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D INSTALL_PYTHON_EXAMPLES=ON -D BUILD_opencv_python2=OFF -D INSTALL_C_EXAMPLES=OFF -D OPENCV_ENABLE_NONFREE=ON -D OPENCV_EXTRA_MODULES_PATH=/home/antpc/tf/opencv/opencv_contrib/modules -D PYTHON_DEFAULT_EXECUTABLE=~/.virtualenvs/cv/bin/python -D BUILD_EXAMPLES=ON ..
• Once CMake finishes, it’s important that you inspect the output. Your output should look similar to mine below:

• (After Successful configuration and generating be sure it build with “Python3” and second thing is non-free algorithm is “yes”)
• make -j4 (Here 4 is the no. of cores),(To check your system cores type “nproc” in terminal and press enter)

• sudo make install
• sudo ldconfig
• ls /usr/local/python/cv2/python-3.6(to confirm this file “cv2.cpython-36m-x86_64-linux-gnu.so” )
• go to that location by using command “cd /usr/local/python/cv2/python-3.6”
• sudo mv cv2.cpython-36m-x86_64-linux-gnu.so cv2.so
• cd ~/.virtualenvs/cv/lib/python3.6/site-packages/
• ln -s /usr/local/python/cv2/python-3.6/cv2.so cv2.so
• Now Test your OpenCV installation
• workon cv
• python
• import cv2
• cv2.__version__
• quit()

Step 6: Darknet YOLOv4 installation Steps:

• Goto this link and download the repo in zip “https://github.com/AlexeyAB/darknet/tree/darknet_yolo_v4_pre"
• unzip downloaded file “darknet_yolo_v4_pre”
• open terminal on that folder and activate your virtualenvs by this command==>
• workon cv
• sudo apt install libopencv-dev python3-opencv
• sudo apt install make git g++
• on the folder open make file and change these ==>
• GPU=1, CUDNN=1, CUDNN_HALF=1, OPENCV=1, LIBSO=1.
• make
• mkdir build_release
• cd build_release
• sudo apt remove cmake
• pip install cmake — upgrade
• type “which cmake” and add that path in the .bashrc file that are in home location.
• cmake ..
• make
• copy the darknet and libdark.so from the build_release to the darknet folder.
• You also have to rename ( libdark.so -> libdarknet.so )
• Confirm your installation by going to the darknet folder with same “cv environment” and type “python” and then “import darknet”.

Thank You✌