If there any mentors for the solvers module please review this, so that it might be easier to submit a better proposal.
On Monday, March 22, 2021 at 12:41:48 PM UTC+5:30 quadiro...@gmail.com wrote: > Hi everyone! > > My name is Muhammed Abdul Quadir Owais (GIT-LINK > <https://github.com/MaqOwais>). I am interested in taking part in GSoC > this year. I have been contributing to sympy since last year (was active > around 5 to 6 months) and for the same reason, I’ve pretty nice command > over solveset module. I would like to discuss the solveset and solvers > proposal for improving solvers more specifically the solveset.py module. > > The one-line description of this project in my view is, > > “To make `solveset` function able to solve the equations as much as > `solve`, 'linsolve' functions with its special features/attributes (like > giving a complete and definite set of solutions along with the complex > solutions which could return an infinite number of solutions,..) and this > should be ultimately easily maintainable (with better structure and > documentation).” > > My step by step plan is, > > > 1. Improving lambert for solveset along with solving transcendental > equations(like lambert, exponential, etc..) in the complex domain. (On > which I’m currently working and could be finished prior to GSoC). (PR > #14972 <https://github.com/sympy/sympy/pull/14972> → lambert pr > <https://github.com/sympy/sympy/pull/21025>, pr for exp > <https://github.com/sympy/sympy/pull/21130>) > > 2. Improving nonlinsolve for trigonometric (PR#18489 > <https://github.com/sympy/sympy/pull/18489>) > > 3. Integration of helper solvers with solveset. So that solveset may give > solutions for multiple variables also set infrastructure to be developed > along the way! > > (helper solvers → linsolve, nonlinsolve, and solve_decomposition) > > > > > > Further issues to be addressed are: > > 1. > > Improving `invert` function in case of the complex domain. > > Currently, a lot of complex domain equations are not able to give > solutions( whereas the real domain is able to give solutions) due to less > development of the `invert` function > > 1. > > Solvify gives solutions ( as of solve function) with the help of > solveset but does cover only fewer cases when compared to solveset. > > This is due to the fact that solvify doesn’t return solutions for > intersection set so to cover this solvify should include cases for > intersection set. > > > > The major reference PR’s which are concerned with this GSoC: > > 1. > > https://github.com/sympy/sympy/issues/10006 → This is Replacing > solve() with solveset() > 2. > > https://github.com/sympy/sympy/pull/18489 → for nonlinsolve for > trigonometric and Improving its set infrastructure > 3. > > https://github.com/sympy/sympy/issues/16861 → Proper solution of > linear equations with symbolic coefficients. > > > CREDITS: OSCAR BENJAMIN helped me a lot in understanding the solveset > better also in current PR’s, in documentations even though I was a bit > frustrating :) > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to sympy+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/fa592ddc-a401-41da-846e-8ef6e6165f3en%40googlegroups.com.
