Dear Thomas,
the suggestion of Mateusz appears to work fine for my needs. I don't mind
if some optimization opportunities are left out... the common factors are
even too many for me as of now :-)
one more little question though:
after collecting the factors i have to do some C code generation for them,
and i am doing some string magic to have a form i like
for my C++ FE program. Nothing too clever really, but i would need to have
access somehow to the name of the symbol.
i thought at first that __str__() would do, but no
imagine i have a symbol
a = symbol('a')
a = sqrt(2)
i would like something that returns me "a"
while
a.__str__() will return sqrt(2)
is there a method to give this? (surely yes, but i can not find it in the
documentation
cheers
Riccardo
On Sunday, November 29, 2015 at 10:08:21 AM UTC+1, Thomas Hisch wrote:
>
> Take a look at https://github.com/sympy/sympy/pull/7318. I remember that
> this PR didn't work for all matrices (I guess matices including expressions
> with sqrt(2)). If we find a better way to determine the common factor, I'll
> update the PR.
>
> On Saturday, November 28, 2015 at 8:08:52 PM UTC+1, Riccardo Rossi wrote:
>>
>> Dear Mateusz,
>>
>> what you suggest is exactly what i was hoping for and could not find by
>> googling
>>
>> i'll definitely try that out :-)
>>
>> Dzienkuje Bardzo! (hope spelling is correct and...that i guessed
>> correctly your nationalty)
>>
>> cheers
>> Riccardo
>>
>> On Saturday, November 28, 2015 at 11:07:20 AM UTC+1, Mateusz Paprocki
>> wrote:
>>>
>>> Hi,
>>>
>>> On 27 November 2015 at 19:34, Riccardo Rossi <[email protected]> wrote:
>>> > Dear list,
>>> >
>>> > i am a newby to sympy, and i should say that i liked what i found, so
>>> ...
>>> > first of all kudos to the developers.
>>> >
>>> > as of now i can succesfully generate my finite element matrices using
>>> sympy,
>>> > which saves me quite a lot of work.
>>> >
>>> > the point is that now i would like to optimize a bit what i did, and i
>>> would
>>> > like to collect some common factors between the entries of a matrix.
>>> >
>>> > for example imagine that i have (pseudocode and just an example, no
>>> physics
>>> > behind)
>>> >
>>> > a,b = symbols('a b')
>>> >
>>> > A = Matrix(2,1)
>>> > A[0] = a*(exp(a+b)+exp(b^2))
>>> > A[1] = b*(exp(a+b)+exp(b^2))
>>> >
>>> > i would like a way to detect that the term
>>> > (exp(a+b)+exp(b^2))
>>> >
>>> > is common to the different entries and eventually later on do
>>> something of
>>> > the type
>>> >
>>> > aux = (exp(a+b)+exp(b^2))
>>> > A[0] = a*aux
>>> > A[1] = b*aux
>>> >
>>> > note that later on for me it would be still interesting to do
>>> something
>>> > similar on SOME of the entries of the matrix
>>> >
>>> > for example if i had
>>> >
>>> > A = Matrix(3,1)
>>> > A[0] = a*(exp(a+b)+exp(b^2))
>>> > A[1] = b*(exp(a+b)+exp(b^2))
>>> > A[2] = a+b
>>> >
>>> > i would still love to have
>>> >
>>> >
>>> > aux = (exp(a+b)+exp(b^2))
>>> > A[0] = a*aux
>>> > A[1] = b*aux
>>> > A[2] = a+b
>>> >
>>>
>>> you could use cse() (common subexpression elimination) for this, e.g.:
>>>
>>> In [1]: from sympy import *
>>>
>>> In [2]: var('a,b')
>>> Out[2]: (a, b)
>>>
>>> In [3]: aux = exp(a + b) + exp(b**2)
>>>
>>> In [4]: Matrix([a*aux, b*aux, a + b])
>>> Out[4]:
>>> Matrix([
>>> [a*(exp(b**2) + exp(a + b))],
>>> [b*(exp(b**2) + exp(a + b))],
>>> [ a + b]])
>>>
>>> In [5]: replacements, (M,) = cse(_)
>>>
>>> In [6]: M
>>> Out[6]:
>>> Matrix([
>>> [a*x1],
>>> [b*x1],
>>> [ x0]])
>>>
>>> In [7]: replacements
>>> Out[7]: [(x0, a + b), (x1, exp(b**2) + exp(x0))]
>>>
>>> In [8]: M.subs(list(reversed(replacements)))
>>> Out[8]:
>>> Matrix([
>>> [a*(exp(b**2) + exp(a + b))],
>>> [b*(exp(b**2) + exp(a + b))],
>>> [ a + b]])
>>>
>>> However, this may not be exactly what you want, because it eliminates
>>> `a + b` as well.
>>>
>>> Mateusz
>>>
>>> >
>>> >
>>> > thanks in advance for any suggestion.
>>> >
>>> > cheers
>>> > Riccardo
>>> >
>>> > --
>>> > 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 post to this group, send email to [email protected].
>>> > Visit this group at http://groups.google.com/group/sympy.
>>> > To view this discussion on the web visit
>>> >
>>> https://groups.google.com/d/msgid/sympy/a72b7481-c0c8-49a2-9a1f-3e88ae8f5f3d%40googlegroups.com.
>>>
>>>
>>> > For more options, visit https://groups.google.com/d/optout.
>>>
>>
--
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 post to this group, send email to [email protected].
Visit this group at http://groups.google.com/group/sympy.
To view this discussion on the web visit
https://groups.google.com/d/msgid/sympy/1c8e9ddf-94fa-46ea-88a6-b79d77deec02%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
