Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-26 Thread Kalevi Suominen
On Thursday, March 26, 2015 at 12:27:13 AM UTC+2, Luv Agarwal wrote: What reconsideration are you talking about? I was just pondering if you really have time to complete also isolate_roots in week 1 already. Even proj might take some time... Kalevi -- You received this message

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-26 Thread Luv Agarwal
On Thursday, March 26, 2015 at 1:02:09 PM UTC+5:30, Kalevi Suominen wrote: On Thursday, March 26, 2015 at 12:27:13 AM UTC+2, Luv Agarwal wrote: What reconsideration are you talking about? I was just pondering if you really have time to complete also isolate_roots in week 1

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-25 Thread Kalevi Suominen
On Wednesday, March 25, 2015 at 1:54:24 AM UTC+2, Luv Agarwal wrote: Is primitive_element doing something similar to simple https://github.com/sympy/sympy/wiki/GSoC-2015-Application-Luv-Agarwal:-Cylindrical-Algebraic-Decomposition#2-simple?. If yes, will there be difference in

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-25 Thread Luv Agarwal
On Tuesday, March 24, 2015 at 5:56:59 PM UTC+5:30, Kalevi Suominen wrote: On Tuesday, March 24, 2015 at 2:44:03 AM UTC+2, Luv Agarwal wrote I have updated my proposal https://github.com/sympy/sympy/wiki/GSoC-2015-Application-Luv-Agarwal:-Cylindrical-Algebraic-Decomposition . Please

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-24 Thread Luv Agarwal
On Wednesday, March 25, 2015 at 3:24:23 AM UTC+5:30, Kalevi Suominen wrote: On Tuesday, March 24, 2015 at 11:28:11 PM UTC+2, Aaron Meurer wrote: The current algebraic extension code in the polys (like Poly(extension=[sqrt(x)])) is quite weak. Will this project require expanding those

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-24 Thread Kalevi Suominen
On Tuesday, March 24, 2015 at 11:28:11 PM UTC+2, Aaron Meurer wrote: The current algebraic extension code in the polys (like Poly(extension=[sqrt(x)])) is quite weak. Will this project require expanding those abilities? Aaron, I have not delved deeper into the matter, but it looks like

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-24 Thread Aaron Meurer
There was also a GSoC project on this a couple of years ago (https://github.com/sympy/sympy/wiki/GSoC-2013-Application-Katja-Sophie-Hotz:-Faster-Algorithms-for-Polynomials-over-Algebraic-Number-Fields), and I am not fully caught up on what still needs to be implemented. IIRC, though, there were

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-24 Thread Luv Agarwal
On Wednesday, March 25, 2015 at 3:28:34 AM UTC+5:30, Aaron Meurer wrote: There was also a GSoC project on this a couple of years ago ( https://github.com/sympy/sympy/wiki/GSoC-2013-Application-Katja-Sophie-Hotz:-Faster-Algorithms-for-Polynomials-over-Algebraic-Number-Fields), and I am

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-23 Thread Luv Agarwal
On Sunday, March 22, 2015 at 4:46:40 PM UTC+5:30, Kalevi Suominen wrote: On Sunday, March 22, 2015 at 11:33:43 AM UTC+2, Luv Agarwal wrote: Thanks, I looked at them. How can I implement dup_isolate_real_roots_sqf for algebraic domains? It seems that the algorithm in rootisolation

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-22 Thread Kalevi Suominen
On Sunday, March 22, 2015 at 11:33:43 AM UTC+2, Luv Agarwal wrote: Thanks, I looked at them. How can I implement dup_isolate_real_roots_sqf for algebraic domains? It seems that the algorithm in rootisolation should work for real algebraic numbers as well. The critical points are the

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-22 Thread Luv Agarwal
On Sunday, March 22, 2015 at 2:49:37 PM UTC+5:30, Luv Agarwal wrote: On Saturday, March 21, 2015 at 12:15:34 PM UTC+5:30, Kalevi Suominen wrote: On Saturday, March 21, 2015 at 3:40:07 AM UTC+2, Luv Agarwal wrote: I have added the implementation details of few functions in the draft

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-22 Thread Luv Agarwal
On Saturday, March 21, 2015 at 12:15:34 PM UTC+5:30, Kalevi Suominen wrote: On Saturday, March 21, 2015 at 3:40:07 AM UTC+2, Luv Agarwal wrote: I have added the implementation details of few functions in the draft

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-21 Thread Kalevi Suominen
On Saturday, March 21, 2015 at 3:40:07 AM UTC+2, Luv Agarwal wrote: I have added the implementation details of few functions in the draft https://github.com/sympy/sympy/wiki/GSoC-2015-Application-Luv-Agarwal:-Cylindrical-Algebraic-Decomposition . How can I implement refine_root (and

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-20 Thread Luv Agarwal
I have added the implementation details of few functions in the draft https://github.com/sympy/sympy/wiki/GSoC-2015-Application-Luv-Agarwal:-Cylindrical-Algebraic-Decomposition . How can I implement refine_root (and dup_isolate_real_roots_sqf) for polynomials over algebraic numbers? -- You

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-16 Thread Luv Agarwal
I have made an initial draft at [this](https://github.com/sympy/sympy/wiki/GSoC-2015-Application-Luv-Agarwal:-Cylindrical-Algebraic-Decomposition). I will be adding more implementing details asap. It would be great if I can get feedbacks. Thanks -- You received this message because you are

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-16 Thread Aaron Meurer
This looks like a good start. I would add in the pseudocode (maybe in a new section if it is too much) in each part, what functions will be needed to do that step, and if those functions are implemented or not (or if something not implemented will need to be implemented). These should be specific

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-16 Thread Aaron Meurer
Does the variable order to CAD matter? In Python, {x, y} is a set, so {x, y} == {y, x}. Should cylindrically_decompose(x**2 + y**2 1, {x, y}) really be cylindrically_decompose(x**2 + y**2 1, [x, y])? Aaron Meurer On Mon, Mar 16, 2015 at 4:07 PM, Aaron Meurer asmeu...@gmail.com wrote: This

Re: [sympy] Re: CAD Implementation - GSoC 2015

2015-03-07 Thread Aaron Meurer
A suggestion is to show a fake session where you show what the module will do (like a doctest for the final function showing inputs and outputs). This shows that you at least are thinking about the API, and that you understand the problem. For something as complicated as CAD, I would also make

[sympy] Re: CAD Implementation - GSoC 2015

2015-03-06 Thread Luv Agarwal
What should be the level of abstraction of implementation details in the gsoc application?. Is it good to include heavy implementation details (that I have thought of yet)?. Should I include the details of as many algorithms as I can?. If not then where should I discuss those details with