Geometric Algebra Multivector Estimation

GAME is framework for estimation of multivectors in geometric algebra with focus on, but not limited to, the Euclidean and conformal model.

Docker

Build

docker build . -t tingelst/game

Run jupyter notebook

docker run -it -v /home/<your-user-name>/path/to/game:/home/game/game -p 8888:8888 tingelst/game:latest

Run a command line terminal within the docker container:

sudo docker run -v /home/<your-user-name>/repos/game:/home/game/repo/game -ti tingelst/game:latest /bin/bash -c "exec "${SHELL:-sh}""

If a docker container is running and you want to open a terminal

docker ps # See running containers
# Open a terminal within the container, assuming the name of container is nervous_heyrovsky
docker exec -it nervous_heyrovsky sh 

This will run jupyter notebook and show a link that you can follow in your browser to all the python notebooks.

Installation

The main external dependency of GAME is the Ceres optimization framework from Google.

On OSX we recommend building the homebrew formula from source:

brew tap homebrew/science
brew install ceres-solver --build-from-source

On Linux we recommend following the installation instructions found here: http://ceres-solver.org/building.html.

To build GAME follow the steps below:

$ git clone --recursive https://github.com/tingelst/game.git
$ cd game
$ mkdir build && cd build
$ cmake .. && make