‪On Tue, 16 Feb 2021 at 17:01, ‫אוריאל מליחי‬‎ <[email protected]> wrote:‬
>
> For your first question, I intend to create a new function that when given a 
> set of linear inequalities and a target inequality it would output True if 
> the target is implied by this set, False if it is not, and Unknown otherwise. 
> I wrote doctests to make it all more clear.
>
> You can see it here –
>
> https://github.com/orielmalihi/Final-Project/blob/main/iset%20tests.py

That would be useful in sympy.

> For your second question, I intended to use Scipy LP solver, but as far as I 
> can tell it only works with floats. I could convert every coefficient to 
> float using the 'evalf()' function but then we would get 99.9% accurate 
> results and not 100%.

We can't use scipy as part of sympy because sympy does not "depend" on
scipy. In sympy we want to handle the symbolic case. The *minimum* we
would want for an implementation in sympy is something that handles
systems with rational coefficients without *any* rounding error. The
more interesting cases would be irrational coefficients like sqrt(2)
or the symbolic case where the coefficients are something like a
symbol x. If the implementation just uses scipy then there's no reason
to include it in sympy because users could just use it directly from
scipy.


Oscar

-- 
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 [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/sympy/CAHVvXxQb4TEb7tPov6df5Vv2P-d4XYSKpi3xqUHY5w%2BuTCx2pQ%40mail.gmail.com.

Reply via email to