GSoC with Flint – Week 1
My Google summer of code project was accepted, so this summer I’m working on implementing efficient methods to find the Hermite normal form of a matrix in Flint. The official coding period began this week and so I have been starting to code.
It has been a slow first week for me as I still have several university exams on, so I haven’t had as much time to work on the project this week as I will in future weeks. So there is not much to report so far but I’ll do so anyway to get in to the swing of things.
I originally wrote some sample code when I applied for GSoC which was a very simple HNF algorithm following Henri Cohen’s book. This used a column Hermite normal form convention that placed zero columns at the start, however this is not the convention I am using for the actual project where the row Hermite form makes more sense (additionally this convention is used by two of the papers I am trying to implement). So I have been changing the code and tests over to the new convention. This has not been quite as straightforward as I would have liked and it has thrown up a few bugs, seemingly in my test code rather than in the HNF algorithm itself however. Hopefully I can sort these testing issues out and get onto implementing some more interesting algorithms soon.