Hi Community,I'm Tieming, a first year graduate student in University of Southern California, majoring in Electrical Engineering.
I am very good in python, C++, for me, the longest programming language experience is MATLAB, since my bachelor is in Physics and Semiconductor area, I have done a lot of computational simulation jobs.
I have implemented a open source software in Python, trying to automatically do data cleaning, and provide a one - stop refinement of raw data, so I'm also good at open source community. For C++, I have practiced many kinds of data structures and algorithms.
I haven't used sympy before, but I'm very excited to make a great contribution to this package and attend GSoC 2020!
Basically I am very interested in Classical Mechanics: Efficient Equation of Motion Generation with C++, which is still in a beginning point. Because I would like to see the great speed improvement in Symengine. So I roughly separate into several starting steps, based on works in 2017 https://groups.google.com/forum/?fromgroups=#!topic/sympy/KFyfC4gQUSQ:
1. list some (as much as possible) slow functions in sympy.physics.vector and sympy.physics.mechanics.
2. check these slow functions, whether they could use symengine to speed up, using existed functinons (such as matrix.cpp) in symengine.
3. implement possible functions or algorithms in cpp, for example for Kane's method or Lagrangian's method, to speed up the rest slow functions.
These are my simple ideas, I feel sorry because I didn't know about this project earier, so please kindly help me for the whole ideas ( or I misunderstand the project ) so that I could catch up the proposal deadline and start this project in a few days.
Best,
Tieming Sun