Skip to content

Latest commit

 

History

History
186 lines (98 loc) · 5.33 KB

GravMachine_Install_Ubuntu_18_04_CUDA.md

File metadata and controls

186 lines (98 loc) · 5.33 KB
Raw

Installation of Gravity Machine on Ubuntu 18.04 LTS with CUDA.

These installations and dependencies correspond to the legacy version (1.0.0) of the code-base.

1. Preliminaries

Install pip

sudo apt-get install python3-pip

Install numpy

pip3 install numpy

Install Scipy

pip3 install scipy

Install pandas

pip3 install pandas

Install Python Image Library

pip3 install pillow

Install imutils

pip3 install imutils

Install Serial

pip3 install pyserial

Import cmocean

sudo pip3 install cmocean

Install matplotlib

sudo pip3 install matplotlib

2. CUDA install:

Important note: Need to ensure that UEFI secure boot is configured correctly. In particular, this needs to be enabled during Ubuntu installation and the key should be 'Enrolled' during the first reboot. Not doing this can cause lots of downstream issues installing NVIDIA drivers.

sudo add-apt-repository ppa:graphics-drivers/ppa
sudo apt update
sudo ubuntu-drivers autoinstall

Reboot.

sudo apt install nvidia-cuda-toolkit gcc-6

Once this completes

nvcc --version

This should show the version of the drivers installed.

##3. Install dependencies for TIS camera (Need to install before OpenCV) Build Dependency sudo apt-get install git g++ cmake pkg-config libudev-dev libudev1 libtinyxml-dev libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev libglib2.0-dev libgirepository1.0-dev libusb-1.0-0-dev libzip-dev uvcdynctrl python-setuptools libxml2-dev libpcap-dev libaudit-dev libnotify-dev autoconf intltool gtk-doc-tools python3-setuptools

Run time dependency

sudo apt-get install gstreamer1.0-tools gstreamer1.0-x gstreamer1.0-plugins-base gstreamer1.0-plugins-good gstreamer1.0-plugins-bad gstreamer1.0-plugins-ugly libxml2 libpcap0.8 libaudit1 libnotify4 python3-pyqt5 python3-gi

3. Install camera drivers (TIScamera)

git clone --recursive https://github.com/TheImagingSource/tiscamera.git
cd tiscamera
mkdir build
cd build

# With ARAVIS:
cmake -DBUILD_ARAVIS=ON -DBUILD_GST_1_0=ON -DBUILD_TOOLS=ON -DBUILD_V4L2=ON -DCMAKE_INSTALL_PREFIX=/usr ..
# Without ARAVIS
cmake -DBUILD_ARAVIS=OFF -DBUILD_GST_1_0=ON -DBUILD_TOOLS=ON -DBUILD_V4L2=ON -DCMAKE_INSTALL_PREFIX=/usr ..

make
sudo make install

4. OpenCV installation (from source)

sources: 1. https://www.pyimagesearch.com/2018/05/28/ubuntu-18-04-how-to-install-opencv/ 2. https://gist.github.com/raulqf/a3caa97db3f8760af33266a1475d0e5e Against Pyimagesearch's advice, this installation will not use virtual environments since that causes issues while using the TIS camera module.

sudo apt-get update

sudo apt-get install build-essential cmake unzip pkg-config

sudo apt-get install build-essential cmake unzip pkg-config

sudo apt-get install libjpeg-dev libpng-dev libtiff-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

Get the official OpenCV release

wget -O opencv.zip https://github.com/opencv/opencv/archive/3.4.4.zip

Get the OpenCV-contrib

wget -O opencv_contrib.zip https://github.com/opencv/opencv_contrib/archive/3.4.4.zip

Unzip the two files

unzip opencv.zip
unzip opencv_contrib.zip

cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D INSTALL_C_EXAMPLES=OFF -D INSTALL_PYTHON_EXAMPLES=ON -D OPENCV_EXTRA_MODULES_PATH=~/opencv_contrib/modules -D WITH_CUDA=ON -D WITH_TBB=ON -D ENABLE_FAST_MATH=1 -D CUDA_FAST_MATH=1 -D WITH_CUBLAS=1 -D BUILD_EXAMPLES=ON -D WITH_GSTREAMER=ON ..

Build fails since the compiler used is greater than gcc-6. Trying to recompile using a specified compiler.

cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D INSTALL_C_EXAMPLES=OFF -D INSTALL_PYTHON_EXAMPLES=ON -D OPENCV_EXTRA_MODULES_PATH=~/opencv_contrib/modules -D WITH_CUDA=ON -D WITH_TBB=ON -D ENABLE_FAST_MATH=1 -D CUDA_FAST_MATH=1 -D WITH_CUBLAS=1 -D BUILD_EXAMPLES=ON -D WITH_GSTREAMER=ON -D CMAKE_C_COMPILER=/usr/bin/gcc-6 -D CMAKE_CXX_COMPILER=/usr/bin/g++-6 ..

The above command with the compiler specified works!

5. Install OpenCV (Direct). This is alternative to above step.

pip3 install opencv-python
pip3 install opencv-contrib-python

Install pyqtgraph

pip3 install pyqtgraph

Install gi

sudo apt-get install python3-gi

Install Tk (for python3)

sudo apt-get install python3-tk

6. GUI dependencies (There are specific dependencies to make sure the Qt GUI works)

Make sure that only PyQt5 is installed and not alongside earlier versions

Install QtPy as a wrapper over PyQt5

sudo pip3 install qtpy

Install opengl for 3D graphics

sudo apt-get install python3-pyqt5.qtopengl

(OPTIONAL) Install the Qtsvg library for exporting vector graphics from Qt

sudo apt-get install python3-pyqt5.qtsvg

7. Other dependencies (for specific OpenCV trackers etc.)

install pytorch

Instructions

sudo pip3 install http://download.py)torch.org/whl/cu91/torch-0.4.0-cp36-cp36m-linux_x86_64.whl 

sudo pip3 install torchvision

Install v4l for help with debugging camera issues (Optional)

$ sudo apt-get update $ sudo apt-get install v4l-utils