couple comments: 1) the log should be informative, currently it contains lots of "upd" stuff
2) the lines -> rows changes should be in a separate patch otherwise it looks ok to me. Ondrej On Mon, Aug 24, 2009 at 2:54 PM, Luke<hazelnu...@gmail.com> wrote: > Ondrej and I cooked this patch up together while we were at the Scipy > conference. It implements the following: > > 1) Matrix.get_diag_blocks() -- returns a list of square matrices > taken from the main diagonal of self. > 2) block_diag(list_of_matrices) -- given a list of square matrices, > return a new Matrix instance with each of the matrices in the list on > the main diagonal > 3) Keword argument "try_block_diag"=False added to Matrix.inv() so > that instead of calling the appropriate inversion method on the whole > matrix, the method is instead called on each block diagonal sub matrix > and then the complete inverse is assembled from the individual > sub-inverse matrices. > 4) Added an empty line to the end of each docstring with a doctest so > that both "bin/doctest sympy/matrices" and "bin/doctest > sympy/matrices/matrices.py" pass all doctests. > > If anybody is inverting symbolic matrices with entries which are sympy > expressions, I have found that using "try_block_diag"=True results in > drastically reduced amount of time to invert the matrix. > > Thanks, > ~Luke > > ---------- Forwarded message ---------- > From: Luke <hazelnu...@gmail.com> > Date: Mon, Aug 24, 2009 at 2:43 PM > Subject: Re: reno > To: Ondrej Certik <ond...@certik.cz> > > > I fixed all the doctests in matrices.py to have one line at the end > and squashed it into a single patch that implements all of the block > diagonal stuff we worked on. I'll send the patch to the mailing list > for review, but I have also attached it on this email. > > ~Luke > > On Mon, Aug 24, 2009 at 2:15 PM, Ondrej Certik<ond...@certik.cz> wrote: >> You are right, I am getting failures too, if I execute the >> matrices.py. This should work --- could you please report it in the >> issue? >> >> Thanks, >> O. >> >> On Mon, Aug 24, 2009 at 1:36 PM, Luke<hazelnu...@gmail.com> wrote: >>> No, it is fine. I was doing: >>> $ bin/doctest sympy/matrices/matrices.py >>> instead of >>> bin/doctest sympy/matrices/ >>> >>> ~Luke >>> >>> >>> On Mon, Aug 24, 2009 at 1:21 PM, Ondrej Certik<ond...@certik.cz> wrote: >>>> in master or in the blocks branch? I am not getting any failures in either: >>>> >>>> $ bin/doctest sympy/matrices/ >>>> ============================= test process starts >>>> ============================== >>>> executable: /usr/bin/python (2.6.2-final-0) >>>> >>>> /var/lib/python-support/python2.6/IPython/Magic.py:38: >>>> DeprecationWarning: the sets module is deprecated >>>> from sets import Set >>>> sympy/matrices/__init__.py[0] >>>> [OK] >>>> sympy/matrices/matrices.py[21] ..................... >>>> [OK] >>>> >>>> ================== tests finished: 21 passed in 0.08 seconds >>>> =================== >>>> >>>> >>>> >>>> Ondrej >>>> >>> >> > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "sympy-patches" group. To post to this group, send email to sympy-patches@googlegroups.com To unsubscribe from this group, send email to sympy-patches+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sympy-patches?hl=en -~----------~----~----~----~------~----~------~--~---
