-
Notifications
You must be signed in to change notification settings - Fork 0
GSoC 2012 Report Stefan Krastanov: Functional Differential Geometry (Vector Analysis)
(This document is work in progress)
My name is Stefan Krastanov (github). At the moment I am starting my final year as an undergraduate (integrated BSc and MSc in Physics) at ENS Lyon, France. Before GSoC I have already contributed to SymPy in some minor ways and liked the possibility to work more extensively on a tool that I enjoy using.
This is a brief report outlining the project and the functionality that was implemented as part of my GSoC 2012 work.
For more details about the project check the initial proposal page. Documentation of my progress and details about the implemented functionality can be found on the dedicated blog page.
My mentor for the project was Matthew Rocklin. The whole community helped extensively especially in the issues more detached from the module itself like changes to the SymPy core, ode solvers and matrices module. Prof. Comer Duncan helped by keeping me in touch with what actual researchers would like from the module.
There were a number of issues encountered during the development of the project that relate to other parts of SymPy:
- the use of Basic in args
- simplification routines
- the assumptions subsystem
- the ode solver for systems of equations
- implementation of Jordan forms in
sympy.matrices
I will finish the additions that I made for these modules and commit them in the next month.
Obviously, I will also continue the extensive work that is still to be done for the diffgeom module itself.
The SymPy community (and in particular my mentor Matthew) helped me greatly during code reviews and when I needed help in modifying parts of SymPy not directly related to the new module. Besides the diffgeom module they helped me improve some other parts of SymPy (the ode solvers and the matrices module). While none of this is completely merged I am happy that I was able to contribute in so many different parts of SymPy even if the changes were minor. I will continue working on the diffgeom module and I hope that it will become a useful tool for researchers in General Relativity.