Docker

First, make sure you have docker installed. Follow these instructions if you don't. We provide several MUQ docker images:

Image Name	Description
`mparno/muq`	A simple image containing MUQ (c++ and python) installed in a debian environment with minimal dependencies.
`mparno/muq-jupyter`	Same as the muq image, but with jupyter lab and several other python packages installed.
`mparno/muq-hippylib`	An image building on the FEniCS `quay.io/fenicsproject/stable` image with additional installations of hIPPylib, hIPPylib2MUQ, andd MUQ.
`mparno/muq-build`	An image containing necessary dependencies for MUQ, but not MUQ itself. This image is useful for continuous integration purposes and serves as the base for the `mparno/muq` and `mparno/muq-jupyter` images.

The following command launches a container using the muq image and sharing the current directory with the /home/muq-user folder in the container:

docker pull muq

docker run -it -v ${PWD}:/home/muq-user muq bash

To launch a container using the muq-jupyter image, we should also share port 8888 so we can launch jupyter notebooks inside the container but view them in a host browser. An example doing this is

docker run -it -p 8888:8888 -v ${PWD}:/home/muq-user muq-jupyter bash