Title: Prototyping Inverse Problems using Jupyter and Docker
Date: Wednesday January 16th, 14:30-16:30
Place: DIM-CMM Beauchef 851, North Building, 5th floor, Seminar Room
Instructor: Reidmen Aróstica
Free Registration:
Link to the Hands on Guide (you can print it before the session)
Abstract:
Prototyping is an important aspect for the vast majority of fields. Oriented from a developers perspective, tools such as Docker containers have become standard practice to test and share results avoiding the conflict with packages versions and instal- lations, which in combination with Jupyter Notebooks becomes a useful tool in scientific environments.
The objective of the hand on session will be to introduce tools such as Dockers, Jupyter Notebooks in combination with FEniCS framework to study some classical inverse problems in Python and show some of the main functionalities to offer new tools in a dynamic environment.
The exposition will be a two hour hands-on session divided into two blocks with live coding and case study, requiring Docker and FEniCS software installed on a working laptop.
References:
- Merkel, Dirk (2014) et al. Docker: Lightweight Linux Containers for Consistent Development and Deployment (Linux J.)
- Martin S. Alnæs and Jan Blechta and Johan Hake (2015) et. al. The FEniCS Project Version 1.5 (Archive of Numerical Software)
- Anders Logg and Kent-Andre Mardal and Garth N. Wells and others (2012) et al. Automated Solution of Differential Equations by the Finite Element Method (Springer)
Required preinstallation in your laptop:
In order to assist to the hands on session, you will need a personal computer with the Docker and FEniCS software pre-installed as follows on any OS:
- Install Docker software, freely available at https://www.docker.com . Check if Docker was correctly installed by entering:
- On Linux/Unix, entering on the terminal and typing: docker
- On Windows, starting the Docker application and entering to powershell and typing: docker
- Install FEniCS software by typing on the terminal:
- docker pull quay.io/fenicsproject/stable:2017.2.0
- (not necessary but try it) Install hippylib by typing:
- docker run -ti -p 127.0.0.1:8000:8000 -v $(pwd):/home/fenics/shared -w /home/fenics/shared quay.io/fenicsproject/stable:2017.2.0
- pip3 install hippylib[notebook] –user