symengine/symengine

Fast symbolic manipulation library, written in C++

Marcello

@torshind_gitlab

Output:

x + y
x + y
x + 4.0*y
0.0 + x + 4.0*y + 4.0*z
x + 4*y + 4*z
0.0 + 4.0*y + 4.0*z
4*y + 4*z

Do you have any idea how to eliminate the spurious 0.0 when the coefficients are double?

Isuru Fernando

@isuruf

Its a tough one. We want expand(4.0*(x + y - (x+y))) to give 0.0 instead of 0 to make it clear that the expression is inexact (floating point vs integer)

You can eliminate 0.0 in the final expression by checking if the expression is an Add and if the coefficient is 0.0, remove it

But that doesn't fix 0.0 in intermediate expressions

On second thought, returning 0 for expand(4.0*(x + y - (x+y))) is totally fine

Isuru Fernando

@isuruf

Currently, in symengine 0 * 4.0 gives 0.0, but it should probably return 0. @certik, what do you think?

Isuru Fernando

@isuruf

Opened symengine/symengine#1559

Marcello

@torshind_gitlab

I deleted my previous message because I verified that the workaround that I've proposed doesn't work. Sorry for the confusion.

Isuru Fernando

@isuruf

@torshind_gitlab, you need to use integer(0) as the coefficient in the code you posted previously

Marcello

@torshind_gitlab

yep, if get_coef()->is_zero() is true, forcing an exact coefficient does the trick

thanks again @isuruf

Isuru Fernando

@isuruf

@richardotis, @bocklund, I've released 0.4.1 of symengine. With conda you can use python-symengine 0.4.0 and symengine 0.4.1

I could use some help in fixing the tests of symengine.py at symengine/symengine.py#289

Steven Lee

@stevenleeS0ht

Is there versioning compatibility requirement between symengine and python-symengine?

Isuru Fernando

@isuruf

@stevenleeS0ht, yes, we usually only support one version of symengine recorded at https://github.com/symengine/symengine.py/blob/master/symengine_version.txt

Carl Poppa

@tgn3000

@isuruf Hi Isuru, long time no see! Was wondering if the following behavior is expected. Thanks!! (master, ed7479baf59a36636061acc35b676bcf9073a932)

int main() {
  RCP<const Basic> a = symbol("a"), b = symbol("b"), c = symbol("c");
  map_basic_basic d {{mul(a,b), c}};
  RCP<const Basic> r1 = mul(SymEngine::integer(2), mul(a, b)),
  r2 = add(SymEngine::integer(1), r1);
  std::cout << *r1 << " -> " << *r1->subs(d) << std::endl;
  std::cout << *r2 << " -> " << *r2->subs(d) << std::endl;
  return 0;
}

produces

2*a*b -> 2*a*b
1 + 2*a*b -> 1 + 2*c

Isuru Fernando

@isuruf

@tgn3000, looks like an oversight

Note that, we have

2*a*b*c -> 2*a*b*c
1 + 2*a*b*c -> 1+2*a*b*c

Richard Otis

@richardotis

Hi @isuruf , do you have a sense what the blockers are for the next release of symengine.py? The new as_ctypes API enables some really nice solver accuracy improvements for us downstream, and also lets pycalphad resolve an install-time bug related to finding libsymengine.

Isuru Fernando

@isuruf

@richardotis, I'd like to get #1596, #1582, #1592 done

The new as_ctypes API enables some really nice solver accuracy improvements for us downstream

What's the difference between this and the cython one?

Richard Otis

@richardotis

Two things: it supports the lambda backend with nogil calling (current release doesn't), and it eliminates symengine as a build-time dependency

Richard Otis

@richardotis

@isuruf Thanks for your hard work getting the new release out there. Our downstream CI picked it up without skipping a beat.

Sylvain Corlay

@SylvainCorlay

:+1:

Jogi Miglani

@jmig5776

Can anybody please guide me on this symengine/symengine#1617?

Ondřej Čertík

@certik

@jmig5776 I just replied there.

Jogi Miglani

@jmig5776

Yaa I am on it.Thanks

Dario Izzo

@darioizzo

Is there a user documentation for symengine? I am struggling to find one ...

Isuru Fernando

@isuruf

@darioizzo, it's at https://symengine.github.io/classSymEngine_1_1Add.html , but it has close to zero documentation

Björn Dahlgren

@bjodah

@darioizzo even though it's not a 1:1 mapping, SymPy's documentation is a good place to look since many functions are essentially a reimplementation in C++. Also, I find the tests in SymEngine to be quite approachable for examples of usage.

Jogi Miglani

@jmig5776

Btw can anyone tell me that what's the plan for documentation for symengine, because I myself facing very much problems while implementing cancel for apart ....if there is predefined plan then I would like to help in that

Where communities thrive

People

Repo info

Activity