On Tue, Mar 24, 2015 at 4:36 AM, Joachim Durchholz <[email protected]> wrote:
> Am 23.03.2015 um 22:39 schrieb Aaron Meurer:
>>
>> On Mon, Mar 23, 2015 at 4:35 PM, Joachim Durchholz <[email protected]>
>> wrote:
>>>
>>> #####2 Now this is asserting Eq(1/s,1) for old and Eq(s,1) for new. These
>>> are semantically different for the case s=0 (i.e. integral of just
>>> sinh(x)):
>>> Old sort order gives a fail, new sort order gives a definite result.
>>> Now I'm wondering how this is accidentally improving Meijer heuristics.
>>> (Or
>>> is the difference mathematically irrelevant?) And I'm wondering where a
>>> changed sort order might accidentally worsen the heuristics.
>>
>>
>> If s = 0 in either case Ne(s, 1) and Ne(1/s, 1) will be False,
>
>
> Ah. I'd thought that Ne(1/s, 1) would be None (because 1/0 undefined).
>
>> so I
>>
>> think in this case they are completely the same.
>
>
> Then they are.
>
>> The version with s is obviously simpler, but we shouldn't rely on the
>> algorithm to give simple results, unless there are no other changes.
>
>
> Is Meijer heuristic? In that case, I think variation due to changed sort
> order would be expected.

Yes, it uses a lot of pattern matching, which is heuristic in nature.

Aaron Meurer

>
> Oh. I find the ordered() call in _rewrite2 isn't dependent on Basic ordering
> at all,
>
>> I would try to find some other example that this gives different
>> results for,
>
>
> I won't find anything by myself (not enough integral knowledge on this side
> of the keyboard), but I'll take a look at the other unit test failure
> related to integrals.
>
> Or, alternatively, we could start torture testing SymPy by making Basic use
> a random (but consistent-per-run) sort order, e.g. XOR PYTHONHASHROOT into
> the class names before comparing them.
> I'm obviously working on making SymPy's unit tests robust against this kind
> of change, but I'm sceptical that this will be doable for heuristic
> algorithms.
>
>> or, preferably, get ahold of Tom Bachmann, and see if he
>>
>> can shed any insight.
>
>
> I'll try to contact him.
>
> --
> 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/55113020.60804%40durchholz.org.
>
> 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/CAKgW%3D6JQwq3VfSDGShjtU9DPy6Zbv-g-J0fxtV_qp1zcJP91Dw%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to