deploy

Bonnetal! Deployment/Inference (C++/ROS)

Description

• Contains C++ libraries and inference nodes to all tasks in order to allow for:
  • Deploy on a robot using ROS
  • Build libraries against existing C++ applications that can do DL inference for all tasks.

---

Tasks

• Full-image classification.
• Semantic Segmentation.
• Instance Segmentation.
• Object Detection.
• CNN Keypoint/Feature Extraction
• Counting

---

Dependencies

If your system is still not running after installing all these dependencies, have a look at the docker files we include. We put an effort into making sure the pipeline works there, and constantly update the installed dependencies, but if we forget one, please submit a PR to this readme to help the rest.

System dependencies

First you need to install the nvidia driver and CUDA, so have fun!

• CUDA Installation guide: link

• Then you can do the other dependencies:

  $ sudo apt-get update 
  $ sudo apt-get install -yqq  build-essential ninja-build \
    python3-dev python3-pip apt-utils curl git cmake unzip autoconf autogen \
    libtool mlocate zlib1g-dev python3-numpy python3-wheel wget \
    software-properties-common openjdk-8-jdk libpng-dev  \
    libxft-dev ffmpeg libboost-all-dev \
    libyaml-cpp-dev
  $ sudo updatedb

• Then follow the steps to install ROS: Link.

Python dependencies

• Then install the Python packages needed:

  $ sudo apt install python-empy
  $ sudo pip install catkin_tools trollius # for build

Pytorch

• From source: When building the Bonnetal C++ deployment interface, we had some problems between the provided libtorch binaries and the Dual ABI introduced in GCC 5.1, so to avoid installing twice with 2 different methods, we recommend building PyTorch from source here too (the version on my GitHub is in sync with the last one supported by our framework).

  $ git clone --recursive https://github.com/tano297/pytorch
  $ cd pytorch
  $ sudo python3 setup.py install
  $ sudo cp -r torch/include/* /usr/local/include
  $ sudo cp -r torch/lib/* /usr/local/lib
  $ sudo cp -r torch/share/* /usr/local/share

  The last cp steps are optional, the same behavior can be achieved in a similar fashion by modifiying a flag before each build:

  $ CMAKE_PREFIX_PATH=/absolute/path/to/libtorch:$CMAKE_PREFIX_PATH
  

TensorRT

In order to infer with TensorRT during inference with the C++ libraries:

• Install TensorRT: Link.

---

Building

• For standalone use (if no ROS is installed):

  $ cd bonnetal/deploy
  $ git clone https://github.com/ros/catkin.git src/catkin
  $ CMAKE_PREFIX_PATH=""
  $ catkin init
  $ catkin build bonnetal_classification_standalone #example package for classification task

  If you get an error, and your system default is python3 (3.5/3.6) you may need to change the default to avoid breaking the catkin-pkg package:

  $ sudo update-alternatives --install /usr/bin/python python /usr/bin/python2.7 1
  $ sudp update-alternatives --install /usr/bin/python python /usr/bin/python3.6 2

• Extending an existing catkin workspace (ROS installed and workspace sourced):

  $ cd bonnetal/deploy
  $ catkin init
  $ catkin build
  $ source devel/setup.bash