Simbotic Container

Docker (and VSCode DevContainer) for running computer vision and simulation tools.

Supports

• SimboticEngine (UnrealEngine 4.0) and plugins
• SimboticTorch (LibTorch GPU)

Features

• Stable Rust 1.55.0
• GStreamer 1.16.2 (with WebRTC and Data Channels)
• LibTorch 1.9.0 - GPU
• CUDA 11.4
• cuDNN 8.2.2
• nvidia/cudagl:11.4.2-devel-ubuntu20.04
• Inherits UID/GID from host user
• ssh keys added to container agent
• VSCode Remote Containers

Pre-requisites

Audio setup

On host, create pulseaudio socket:

pactl load-module module-native-protocol-unix socket=/tmp/pulseaudio.socket

NVidia Docker 2 setup

Setup NVidia docker: https://github.com/NVIDIA/nvidia-docker

Install:

distribution=$(. /etc/os-release;echo $ID$VERSION_ID) \
   && curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add - \
   && curl -s -L https://nvidia.github.io/nvidia-docker/$distribution/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
sudo apt-get update
sudo apt-get install -y nvidia-docker2
sudo systemctl restart docker

Test nvidia-docker installation:

docker run --rm --gpus all nvidia/cudagl:11.4.2-devel-ubuntu20.04 nvidia-smi

Setup Docker container

./docker_build.sh
./docker_run.sh

Or Setup for VSCode Remote Containers

Move to the root directory of your project and run:

git submodule add [email protected]:Simbotic/SimboticContainer.git .devcontainer

VSCode:

• Install the VSCode extension Remote - Containers
• Press F1 to bring up the Command Palette and type in remote-containers for a full list of commands
• Run the Remote-Containers: Reopen in Container command or run Remote-Containers: Open Folder in Container... command and select the local folder

Note: If you don't want this as a Git Submodule, you may also choose to download this repository as a zip, extract its content, and paste it in a .devcontainer directory at the root of your project.

Test using SimboticTorch

https://github.com/Simbotic/SimboticTorch