This is in. Short story about this problem:
A year ago I asked my most popular question on any StackExchange ever. http://scicomp.stackexchange.com/questions/74/symbolic-software-packages-for-matrix-expressions In it I ask if there are any CASs that can handle logical programming on matrix expressions. This question was popular in the scicomp community and garnered 22 upvotes while the best answer had only 3; I believe that this imbalance is a good indicator that work should be done here. With the merger of this pull request I am finally able to boast that SymPy is capable of solving the posed problem. On Mon, Oct 22, 2012 at 4:06 PM, Matthew Rocklin <[email protected]> wrote: > I've issued a PR here implementing a couple of handlers for > MatrixExpressions. Could someone familiar with the new assumptions system > please take a look before I go much farther? I'd like to make sure that I'm > doing this correctly. There are a few parts of the new assumptions module > that I don't understand. > https://github.com/sympy/sympy/pull/1600 > > > On Mon, Oct 22, 2012 at 12:43 PM, Matthew Rocklin <[email protected]>wrote: > >> I notice that both Predicate and AppliedPredicate subclass Boolean. >> Should Predicate be a Boolean or some sort of BooleanFunction? >> >> >> On Mon, Oct 22, 2012 at 12:43 PM, Matthew Rocklin <[email protected]>wrote: >> >>> It looks like there is some code in ask.py that is specific to >>> expressions. Should I put matrix assumptions here too or separate this into >>> different files. In particular there is code like this >>> >>> known_facts = And( >>> Implies(Q.real, Q.complex), >>> Implies(Q.real, Q.hermitian), >>> .... >>> ) >>> >>> Should I just add to this in the same file? >>> >>> There is also known_facts_cnf which mentions that it might be generated. >>> Is there a process to update this code? >>> >>> On Mon, Oct 22, 2012 at 8:36 AM, Matthew Rocklin <[email protected]>wrote: >>> >>>> I'm on this topic again. Does anyone have any thoughts? >>>> >>>> >>>> On Fri, Mar 30, 2012 at 10:48 AM, Matthew Rocklin >>>> <[email protected]>wrote: >>>> >>>>> I'm interested in developing an assumptions system for matrices. This >>>>> would have predicates like symmetric, singular, orthogonal, etc.... >>>>> >>>>> I've looked through the current assumptions system (and a little bit >>>>> through pull 1162). From this I'm able to create matrix assumptions by >>>>> adding to the files in sympy/assumptions. I create a new directory for >>>>> matrix handlers (a la sets, ntheory, calculus) and I add new predicates to >>>>> Q in ask.py. >>>>> >>>>> This doesn't feel right though, I'm mixing two different assumption >>>>> systems into one set of files. For example Q now has two disjoint sets of >>>>> predicates. What is the right way to do things? Is there a clean way to >>>>> instantiate a new module specific assumptions system? >>>>> >>>>> -Matt >>>>> >>>> >>>> >>> >> > -- You received this message because you are subscribed to the Google Groups "sympy" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/sympy?hl=en.
