This README includes commands to install linux packages
Useful links: How to Install
If you execute sudo apt-get autoremove --purge
accidently, the computer may crash.(No desktop to show)
Ref to Ubuntu 14.04 差点重装,崩溃!, one can execute the following to bring the desktop back!
apt-get install ubuntu-desktop
apt-get install unity
apt-get install unity-common
apt-get install unity-lens*
apt-get install unity-services
apt-get install unity-asset-pool
reboot
apt-get install -y software-properties-common python-software-properties
Ref to remove packages with error on apt-get update [duplicate]
The error message when executing apt-get update
:
Get:1 file:/var/cuda-repo-10-0-local-10.0.130-410.48 InRelease
Ign:1 file:/var/cuda-repo-10-0-local-10.0.130-410.48 InRelease
Get:2 file:/var/cuda-repo-10-1-local-10.1.243-418.87.00 InRelease
Ign:2 file:/var/cuda-repo-10-1-local-10.1.243-418.87.00 InRelease
Get:3 file:/var/nv-tensorrt-repo-cuda10.1-trt6.0.1.5-ga-20190913 InRelease
Ign:3 file:/var/nv-tensorrt-repo-cuda10.1-trt6.0.1.5-ga-20190913 InRelease
Get:4 file:/var/cuda-repo-10-0-local-10.0.130-410.48 Release
Err:4 file:/var/cuda-repo-10-0-local-10.0.130-410.48 Release
File not found - /var/cuda-repo-10-0-local-10.0.130-410.48/Release (2: No such file or directory)
Get:5 file:/var/cuda-repo-10-1-local-10.1.243-418.87.00 Release [574 B]
Get:6 file:/var/nv-tensorrt-repo-cuda10.1-trt6.0.1.5-ga-20190913 Release [574 B]
Get:5 file:/var/cuda-repo-10-1-local-10.1.243-418.87.00 Release [574 B]
Get:6 file:/var/nv-tensorrt-repo-cuda10.1-trt6.0.1.5-ga-20190913 Release [574 B]
Ign:9 http://dl.google.com/linux/chrome/deb stable InRelease
Hit:10 http://archive.ubuntukylin.com:10006/ubuntukylin xenial InRelease
Hit:11 http://dl.google.com/linux/chrome/deb stable Release
Hit:13 http://packages.microsoft.com/repos/vscode stable InRelease
Hit:15 http://security.ubuntu.com/ubuntu xenial-security InRelease
Hit:16 http://cn.archive.ubuntu.com/ubuntu xenial InRelease
Hit:17 http://cn.archive.ubuntu.com/ubuntu xenial-updates InRelease
Hit:18 http://cn.archive.ubuntu.com/ubuntu xenial-backports InRelease
Hit:19 http://ppa.launchpad.net/apt-fast/stable/ubuntu xenial InRelease
Hit:20 http://ppa.launchpad.net/graphics-drivers/ppa/ubuntu xenial InRelease
Hit:21 http://ppa.launchpad.net/webupd8team/java/ubuntu xenial InRelease
Hit:22 http://ppa.launchpad.net/yannubuntu/boot-repair/ubuntu xenial InRelease
Ign:12 https://developer.download.nvidia.cn/compute/cuda/repos/ubuntu1604/x86_64 InRelease
Hit:23 https://developer.download.nvidia.cn/compute/cuda/repos/ubuntu1604/x86_64 Release
Reading package lists... Done
E: The repository 'file:/var/cuda-repo-10-0-local-10.0.130-410.48 Release' does not have a Release file.
N: Updating from such a repository can't be done securely, and is therefore disabled by default.
N: See apt-secure(8) manpage for repository creation and user configuration details.
We need to find where does the info of cuda-repo-10-0
be stored:
cd /etc/apt
grep -ir cuda-repo-10-0
It outputs:
sources.list.d/cuda-10-0-local-10.0.130-410.48.list:deb file:///var/cuda-repo-10-0-local-10.0.130-410.48 /
sources.list.d/cuda-10-0-local-10.0.130-410.48.list.save:deb file:///var/cuda-repo-10-0-local-10.0.130-410.48 /
Edit the two files sources.list.d/cuda-10-0-local-10.0.130-410.48.list
and sources.list.d/cuda-10-0-local-10.0.130-410.48.list.save
, and comment their contents.
After that, apt-get update
can be executed successfully.
add-apt-repository ppa:apt-fast/stable
apt-get update
apt-get -y install apt-fast
There will be 3 questions, answer as follow:
Configuring apt-fast
--------------------
1. apt-get 2. apt 3. aptitude
Package manager to install and remove software: 1
Stored in ${_MAXNUM} variable.
Maximum number of connections: 8
This does not affect package manager dialog but download installable packages before package
manager confirmation.
Suppress apt-fast confirmation dialog? [yes/no] yes
apt search <package-name>
How do I see what packages are installed on Ubuntu Linux?
apt list | grep <package-name>
How can I check the available version of a package in the repositories?
apt-cache policy <package-name>
dpkg -L <package-name>
apt-cache madison <package-name>
apt-get install -y sudo
apt install sshpass
apt-get install -y vim
apt-get install -y curl
apt-get install -y less
apt-get install -y procps
apt-get install -y iputils-ping
apt-get install -y net-tools
apt-get install -y iproute iproute-doc
apt-get install -y p7zip-full
libarchive.13.so
apt install -y libarchive-dev
apt-get install -y lsb-release
glxinfo: command not found .... nvidia debian
apt-get install -y mesa-utils
apt-get install -y zip unzip
apt-get install -y bzip2
How to Extract RAR Files in Ubuntu Linux
sudo apt install unrar
apt-get install -y apache2-utils
apt-get install -y uni2ascii
Before installing cmake
, it's required to install openssl
.
apt-get install -y cmake
To install newer version of cmake
, follows How do I install the latest version of cmake from the command line?
apt-get install -y cmake-curses-gui
apt-get install -y build-essential
Use gcc -v
, g++ --version
and make -v
to check their versions.
This solves:
No CMAKE_CXX_COMPILER could be found.
install gcc-9 on Ubuntu 18.04?
Or using:
apt install gcc-9 g++-9
to install gcc and g++ of specific version.
apt-get install -y dlocate
apt-get install python3
Anaconda documentation : Installing on Linux
How to Install Anaconda on Ubuntu 18.04 and 20.04
How to Install JAVA 8 on Ubuntu 18.04/16.04, Linux Mint 19/18
apt install -y openjdk-8-jdk openjdk-8-jre
Set environment variable, in ~/.bashrc
:
export JAVA_HOME="/usr/lib/jvm/java-1.8.0-openjdk-amd64"
export PATH=$PATH:$JAVA_HOME/bin
apt install python3-pip
apt-get install -y libgoogle-glog-dev
Cartexius/install_gtest_ubuntu.md
# this only download google/googletest's source code, not install it!
sudo apt-get install libgtest-dev
# the source code is put here
cd /usr/src/gtest
mkdir build && cd build
sudo cmake ..
sudo make
sudo cp *.a /usr/lib
mkdir /usr/local/lib/gtest
sudo ln -s /usr/lib/libgtest.a /usr/local/lib/gtest/libgtest.a
sudo ln -s /usr/lib/libgtest_main.a /usr/local/lib/gtest/libgtest_main.a
apt-get install -y libgflags-dev
apt-get install pciutils
How To Install "libffi-dev" Package on Ubuntu
apt install libffi-dev -y
How do I install GNU Readline?
sudo apt install libreadline8 libreadline-dev -y
Package MagickCore was not found in the pkg-config search path
sudo apt install libmagickwand-dev -y
apt-get install cpio
How do I install the OpenSSL libraries on Ubuntu?
apt-get install libssl-dev
Check version:
openssl version -a
Installing curl.h library [duplicate]
apt-get install libcurl4-openssl-dev
apt-get install libconfig++8-dev
apt-get install -y pkg-config
CMAKE Could NOT find PkgConfig (missing: PKG_CONFIG_EXECUTABLE): This can solve Could NOT find PkgConfig (missing: PKG_CONFIG_EXECUTABLE)
when running cmake.
apt-get install -y libmysqlclient-dev
mysql.h will be located at /usr/include/mysql/.
How To Install and Secure Redis on Ubuntu 20.04
apt-get install libboost-all-dev
How to install the NVIDIA drivers on Ubuntu 20.04 Focal Fossa Linux
sudo apt list --installed | grep nvidia
You will find something like:
nvidia-utils-460/focal-updates,focal-security,now 460.73.01-0ubuntu0.20.04.1 amd64 [installed,automatic]
Then install nvidia-driver with corresponding version:
sudo apt install nvidia-driver-460
sudo reboot # to solve “NVIDIA-SMI has failed because it couldn’t communicate with the NVIDIA driver” when using nvidia-smi
How to install CUDA on Ubuntu 20.04 Focal Fossa Linux
apt install nvidia-cuda-toolkit -y
How to uninstall CUDA Toolkit and cuDNN under Linux?
How to uninstall Cuda7.5 from ubuntu?
Add the following:
# sudo apt-get --purge -y remove 'cuda*'
sudo apt autoremove --purge nvidia-cuda-toolkit
sudo reboot
docker pull nvidia/cuda:11.3.0-cudnn8-devel-ubuntu18.04
cuda installation path: /usr/local/cuda-11.3
.
cudnn installation path: /usr/include
, /usr/include/x86_64-linux-gnu
and /usr/lib/x86_64-linux-gnu
.
apt-get install libgtk-3-dev
apt-get install -y ffmpeg
apt-get install -y mediainfo
How to Install Latest VLC in Ubuntu Linux
snap install vlc
systemctl: command not found on ubuntu 16.04
apt-get -y install systemd
Ubuntu Linux install OpenSSH server
Start SSH automatically on boot
apt-get install -y openssh-server
Verify that ssh service running:
systemctl status ssh
To start it automatically after reboot:
sudo systemctl enable ssh
apt-get install intel-gpu-tools
curl -s https://packagecloud.io/install/repositories/github/git-lfs/script.deb.sh | sudo bash
sudo apt-get install git-lfs
git lfs install
Configure error: could not find the zlib library
apt-get install zlib1g-dev
How to check if HDF5 is installed?
apt-get install -y libhdf5-dev
How to install LaTex on Ubuntu 18.04 Bionic Beaver Linux
apt-get install -y texlive-latex-extra
CMake could not find OpenGL in Ubuntu
apt-get install -y libgl1-mesa-dev
how to install x11_xt_lib when configure VTK?
apt-get install -y libxt-dev
How to install Yarn on Ubuntu 20.04 LTS
curl -sL https://dl.yarnpkg.com/debian/pubkey.gpg | sudo apt-key add -
echo "deb https://dl.yarnpkg.com/debian/ stable main" | sudo tee /etc/apt/sources.list.d/yarn.list
sudo apt update
sudo apt install yarn
yarn --version
How to install GLEW on ubuntu?
apt install libglew-dev -y
使用cmake编译出现missing: JPEG_LIBRARY JPEG_INCLUDE_DIR?
apt install libjpeg-dev -y
How to Install libtiff-dev in Ubuntu 18.04
apt install libtiff-dev -y
error while loading shared libraries: libpng12.so.0
wget http://archive.ubuntu.com/ubuntu/pool/main/libp/libpng/libpng_1.2.54.orig.tar.xz
tar xvf libpng_1.2.54.orig.tar.xz
cd libpng-1.2.54
./autogen.sh
./configure
make -j8
sudo make install
sudo ldconfig
Note: png++-0.2.9
requires libpng12.so.0
, but the following command will install libpng16.
sudo apt-get install libpng-dev
PNG++ Installation on Ubuntu 16.04
wget http://download.savannah.nongnu.org/releases/pngpp/png++-0.2.9.tar.gz
sudo tar -xzvf png++-0.2.9.tar.gz -C /usr/src
cd /usr/src/png++-0.2.9/
make
make test
sudo make install
apt install -y libtool
Download Qt installer from Qt Downloads, and then:
mkdir qt5.14.2-install && cd qt5.14.2-install
wget http://download.qt-project.org/official_releases/qt/5.14/5.14.2/qt-opensource-linux-x64-5.14.2.run
chmod +x qt-opensource-linux-x64-5.14.2.run
./qt-opensource-linux-x64-5.14.2.run
A dialog window will pop out, and you can choose the directory(must exist and empty) to install to, say its Qt5.14.2
.
Download VTK-8.2.0.tar.gz
from VTK download page and then untar it.
Configure:
mkdir VTK-Release-build
cd VTK-Release-build
cmake -DCMAKE_BUILD_TYPE:STRING=Release /path/to/VTK-8.2.0
Then download and install Qt5.
Configure VTK with the following options:
cd /path/to/VTK-Release-build
cmake -DVTK_QT_VERSION:STRING=5 \
-DQT_QMAKE_EXECUTABLE:PATH=/path/to/Qt5.14.2/5.14.2/gcc_64/bin/qmake \
-DVTK_Group_Qt:BOOL=ON \
-DCMAKE_PREFIX_PATH:PATH=/path/to/Qt5.14.2/5.14.2/gcc_64/lib/cmake \
-DBUILD_SHARED_LIBS:BOOL=ON
/path/to/VTK
Finally build:
make -j$(nproc)
And install:
make install
apt-get install -y libopencv-dev # this will install older version: 2.4.9.1
The installation path is: /usr/share/OpenCV/
.
Follow the instruction here: OpenCV Tutorials/Introduction to OpenCV/Installation in Linux
Prerequistie: python3 and numpy
apt-get install -y python3-dev
apt-get install -y python3-pip
python3 -m pip install numpy
Prerequisite for highgui
:
apt-get install -y libgtk2.0-dev pkg-config
First download opencv-x.x.x.zip from OpenCV - Releases and opencv_contrib-x.x.x.zip from opencv/opencv_contrib - Releases, and then unzip.
Create directory:
cd ~/opencv
mkdir build
cd build
Configure:
cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local OPENCV_EXTRA_MODULES_PATH=/xxx/opencv_contrib-3.4.1/modules ..
From Build specific modules OpenCV, to only install some modules, one can specify -DBUILD_LIST=...
:
-DBUILD_LIST=core,improc,imgcodecs,dnn,videoio,cudev,highgui
To determine which modules are required, one should check one's source code and find the opencv headers included, for me, they are:
#include <opencv2/opencv.hpp>
#include <opencv2/core/types.hpp>
#include <opencv2/imgproc.hpp>
#include <opencv2/imgcodecs.hpp>
#include <opencv2/dnn/dnn.hpp>
#include <opencv2/videoio.hpp>
Note that cudev
and highgui
are not directly included, but they are required because cudev
is for gpu programming and highgui
is for the user interface window.
After specifying -DBUILD_LIST=...
, part of cmake
's output would be:
-- Disabled by dependency: calib3d cudaarithm cudabgsegm cudacodec cudafeatures2d cudafilters cudaimgproc cudalegacy cudaobjdetect cudaoptflow cudastereo cudawarping features2d flann ml objdetect photo python2 python_bindings_generator shape stitching superres ts video videostab viz
Add OPENCV_EXTRA_MODULES_PATH
to build with modules from opencv_contrib.
Build:
make -j$(nproc)
Install libraries:
sudo make install
Setting environment variables:
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib
To uninstall, go to the build directory, and then:
sudo make uninstall
One still needs to delete the following directories:
rm `ls /usr/local/lib/*opencv*so*` #remove dead softlinks
rm -r /usr/local/lib/opencv4
rm -r /usr/local/include/opencv4
rm -r /usr/local/share/opencv4
Installing TensorRT: Using The NVIDIA Machine Learning Network Repo For Debian Installation
Setup repo:
# Install the NVIDIA Machine Learning network repository installation package
wget https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64/nvidia-machine-learning-repo-ubuntu1604_1.0.0-1_amd64.deb
dpkg -i nvidia-machine-learning-repo-*.deb
If it outputs the following error message:
gpg: no valid OpenPGP data found.
Failed to add GPGKEY at http://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1604/x86_64/7fa2af80.pub to apt keys.
Then try:
wget https://developer.download.nvidia.cn/compute/machine-learning/repos/ubuntu1604/x86_64/7fa2af80.pub -O 7fa2af80.pub
chmod 755 7fa2af80.pub
Continue to install TensorRT:
apt-get update -y
# Install the TensorRT package that fits your particular needs
apt-get install libnvinfer5=5.1.5-1+cuda9.0 libnvinfer-dev=5.1.5-1+cuda9.0
# So TensorRT won't upgrade to a new version
apt-mark hold libnvinfer5 libnvinfer-dev
To check TensorRT version:
dpkg -l | grep nvinfer
or
dpkg -l | grep TensorRT
First download TensorRT tar.gz file from NVIDIA TensorRT.
tar xzvf TensorRT-5.1.x.x.<os>.<arch>-gnu.cuda-x.x.cudnn7.x.tar.gz
Add the following line into ~/.bashrc
:
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:<eg:TensorRT-5.1.x.x/lib>
And then:
source ~/.bashrc
Test, follow the instruction here:
cd <TensorRT root directory>/samples && make
# download 0.pgm ~ 9.pgm
cd <TensorRT root directory>/data/mnist && python3 download_pgms.py
cd <TensorRT root directory>/bin && ./sample_mnist
Follow the instruction here: Install Intel® Distribution of OpenVINO™ toolkit for Linux*
Download l_openvino_toolkit_p_2019.3.376.tgz
from https://software.intel.com/en-us/openvino-toolkit/choose-download/free-download-linux.
tar -xzf l_openvino_toolkit_p_2019.3.376.tgz
cd l_openvino_toolkit_p_2019.3.376
apt-get update -y
# these are the packages needed during installation
apt-get install -y pciutils cpio python3 sudo cmake
./install.sh
cd /opt/intel/openvino/install_dependencies
sudo -E ./install_openvino_dependencies.sh
This is required if you want to run on Intel GPU!
cd /opt/intel/openvino/install_dependencies
./install_NEO_OCL_driver.sh
sudo usermod -a -G video USERNAME
If this is not installed, trying to run on GPU will generate the following error!
[ ERROR ] Failed to create plugin /opt/intel/openvino_2019.3.376/deployment_tools/inference_engine/lib/intel64/libclDNNPlugin.so for device GPU
Please, check your environment
clGetPlatformIDs error -1001
add the following into ~/.bashrc
:
source /opt/intel/openvino/bin/setupvars.sh
and then source ~/.bashrc
.
# first install python tensorflow
python3 -m pip install -i https://pypi.tuna.tsinghua.edu.cn/simple tensorflow
cd /opt/intel/openvino/deployment_tools/model_optimizer/install_prerequisites
sudo ./install_prerequisites_tf.sh
cd /opt/intel/openvino/deployment_tools/demo
# Image Classification verification script
./demo_squeezenet_download_convert_run.sh
# Inference Pipeline verification script
./demo_security_barrier_camera.sh
Uninstall OpenVINO completely from Ubuntu machine
cd /opt/intel/openvino/openvino_toolkit_uninstaller
sudo ./uninstall.sh
How to install Android SDK on Ubuntu?
Download sdk-tools-linux-4333796.zip
from android studio - Command line tools only
Unzip it to an appropiate location, the resulting directory is tools
.
mkdir android-sdk
mv tools android-sdk
Download the zip file from https://developer.android.com/ndk/downloads.
Unzip it to an appropiate location.
Update environment variables, in /etc/profile
, add:
export ANDROID_NDK_HOME=/home/jian/Documents/installation/android-ndk-r20
export PATH=$ANDROID_NDK_HOME:$PATH
First download the required version from https://github.com/protocolbuffers/protobuf/tags (here I download v2.6.1 because TensorRT uses proto2).
tar -xzf v2.6.1.tar.gz
And you will get a folder protobuf-2.6.1
.
Since the link of googletest in ./autogen.sh
is broken, we need to download it ourselves.
Download googletest from https://github.com/google/googletest/releases (here I download release-1.5.0 because proto buffers v2.6.1 requires it).
tar -xzf release-1.5.0.tar.gz
# got folder: googletest-release-1.5.0
mv googletest-release-1.5.0 protobuf-2.6.1/gtest
Then following the official instruction:
./autogen.sh
./configure
make
make check
sudo make install
sudo ldconfig # refresh shared library cache.
To generate c++ code:
protoc --cpp_out=`pwd` <your_file>.proto
It will generate <your_file>.pb.cc
and <your_file>.pb.h
.
For more: check the developer guide.
How to Install KDE Plasma in Linux Desktop
sudo apt install kde-plasma-desktop
And select sddm
.
How to take a screenshot in KDE Plasma Ubuntu?
sudo apt install kde-spectacle
sudo apt install fcitx5*
im-config
select fcitx5 and then restart.
What is the ECM used in building fcitx?
sudo apt install extra-cmake-modules
According to Fcitx5: Any Traditional Chinese input users know why Cangjie is simplified-only now?, one needs to install it manually:
如何編譯「fcitx5-table-extra」並且簡易打包成「Debian Package」
Prerequisite:
sudo apt install libime-bin # libime_tabledict
sudo apt install extra-cmake-modules # CMakeLists.txt: ECMConfig.cmake, ecm-config.cmake
sudo apt install libimetable-dev libimecore-dev
sudo apt install libfcitx5utils-dev libfcitx5core-dev libfcitx5config-dev fcitx5-chinese-addons
git clone https://github.com/fcitx/fcitx5-table-extra.git
cd fcitx5-table-extra/
mkdir build
cd build/
cmake -DCMAKE_INSTALL_PREFIX=/usr -DCMAKE_INSTALL_LIBDIR=/usr/lib ..
make
sudo make install
When cmake --build .
, the following errors may happen:
/bin/sh: 1: LibIME::tabledict: not found
make[2]: *** [tables/CMakeFiles/table_data.dir/build.make:92:tables/cangjie3.main.dict] 錯誤 127
make[1]: *** [CMakeFiles/Makefile2:733:tables/CMakeFiles/table_data.dir/all] 錯誤 2
make: *** [Makefile:130:all] 錯誤 2
Edit tables/CMakeFiles/table_data.dir/build.make
, replace LibIME::tabledict
with libime_tabledict
and then retry.
And there may be following error:
terminate called after throwing an instance of 'std::invalid_argument'
what(): invalid rule entry
Aborted (core dumped)
make[2]: *** [tables/CMakeFiles/table_data.dir/build.make:156:tables/zhengma-large.main.dict] 錯誤 134
make[1]: *** [CMakeFiles/Makefile2:733:tables/CMakeFiles/table_data.dir/all] 錯誤 2
make: *** [Makefile:130:all] 錯誤 2
Since I don't use zhengma, so just edit tables/CMakeFiles/table_data.dir/build.make
comment out line 156:
cd /xxx/fcitx5-table-extra/build/tables && libime_tabledict /xxx/fcitx5-table-extra/tables/zhengma-large.txt /xxx/fcitx5-table-extra/build/tables/zhengma-large.main.dict
After comment this line, when sudo make install
, another error may occur:
CMake Error at tables/cmake_install.cmake:86 (file):
file INSTALL cannot find
"/home/ubt/Documents/installation/fcitx5-table-extra/build/tables/zhengma-large.main.dict":
No such file or directory.
Call Stack (most recent call first):
cmake_install.cmake:47 (include)
make: *** [Makefile:86:install] 錯誤 1
Edit tables/cmake_install.cmake
and comment out line 105:
"/xxx/fcitx5-table-extra/build/tables/zhengma-large.main.dict"
yum install -y facter
yum install -y mlocate
yum install -y iproute iproute-doc
yum install -y bzip2
yum install -y httpd-tools