]>
gitweb.michael.orlitzky.com - sage.d.git/log
Michael Orlitzky [Sun, 20 Sep 2015 02:34:52 +0000 (22:34 -0400)]
Add examples for the is_lyapunov_like() function.
Michael Orlitzky [Sun, 20 Sep 2015 00:57:39 +0000 (20:57 -0400)]
Update usage of matrix_simplify_full (now part of Sage).
Michael Orlitzky [Sat, 19 Sep 2015 01:50:10 +0000 (21:50 -0400)]
Remove matrix_simplify_full(), this was merged in trac #12162.
Michael Orlitzky [Sat, 19 Sep 2015 01:49:26 +0000 (21:49 -0400)]
Remove matrix_subs_expr() now that subs() absorbed its features.
Michael Orlitzky [Thu, 17 Sep 2015 05:12:40 +0000 (01:12 -0400)]
Add the is_lyapunov_like() function to the cone.cone module.
Michael Orlitzky [Fri, 11 Sep 2015 15:06:50 +0000 (11:06 -0400)]
Begin work on the rearrangement cone.
Michael Orlitzky [Thu, 20 Aug 2015 20:15:44 +0000 (16:15 -0400)]
Replace _rho with _restrict_to_space in cone/tests.py.
Michael Orlitzky [Thu, 20 Aug 2015 18:18:20 +0000 (14:18 -0400)]
Update discrete_complementarity_set() docs.
Michael Orlitzky [Sun, 16 Aug 2015 15:51:26 +0000 (11:51 -0400)]
Rename _rho to _restrict_to_space and make it work on subspaces instead.
Michael Orlitzky [Tue, 11 Aug 2015 04:41:51 +0000 (00:41 -0400)]
Slightly simplify LL() code.
Michael Orlitzky [Wed, 5 Aug 2015 01:18:41 +0000 (21:18 -0400)]
Update some comments for lyapunov_rank().
Michael Orlitzky [Wed, 5 Aug 2015 01:18:01 +0000 (21:18 -0400)]
Move some excessive tests into a new module, cone.tests.
Michael Orlitzky [Tue, 4 Aug 2015 15:56:25 +0000 (11:56 -0400)]
Update max_ambient_dim parameter name for random_cone().
Michael Orlitzky [Mon, 27 Jul 2015 14:36:07 +0000 (10:36 -0400)]
Update docs for cone.discrete_complementarity_set().
Michael Orlitzky [Mon, 27 Jul 2015 14:33:55 +0000 (10:33 -0400)]
mjo/interpolation.py: fix typo.
Michael Orlitzky [Mon, 15 Jun 2015 18:54:10 +0000 (14:54 -0400)]
Make basically_the_same() and rho() functions private.
Add examples/tests for basically_the_same().
Add linear isomorphism test for lyapunov_rank().
Michael Orlitzky [Mon, 15 Jun 2015 17:29:00 +0000 (13:29 -0400)]
Add a new test for a theorem in the improper paper.
Michael Orlitzky [Sun, 14 Jun 2015 16:25:55 +0000 (12:25 -0400)]
Remove the lineality() and codim() functions (now depends on branch u/mjo/ticket/18701).
Michael Orlitzky [Sat, 13 Jun 2015 01:39:12 +0000 (21:39 -0400)]
Finish test cleanup, notation updates, and dead code removal.
Michael Orlitzky [Fri, 12 Jun 2015 23:13:40 +0000 (19:13 -0400)]
Revert "Remove unused codim() function."
This reverts commit
4418c497a443fb1f5cb068ced5a2ddd5a9a0ad05 .
Michael Orlitzky [Fri, 12 Jun 2015 23:13:31 +0000 (19:13 -0400)]
In the middle of mangling things.
Michael Orlitzky [Fri, 12 Jun 2015 22:27:33 +0000 (18:27 -0400)]
Aaaand the test that was using that "unused" function.
Michael Orlitzky [Fri, 12 Jun 2015 22:27:07 +0000 (18:27 -0400)]
Remove unused unrestrict_span() function.
Michael Orlitzky [Fri, 12 Jun 2015 22:25:00 +0000 (18:25 -0400)]
Remove unused codim() function.
Michael Orlitzky [Sun, 7 Jun 2015 04:42:39 +0000 (00:42 -0400)]
Add back the drop_dependent() function.
It turns out that it doesn't matter if our basis is orthogonal, so we
don't need to do Gram-Schmidt. Since this will be relied upon in the
paper, we go back to using (a subset of) the rays of the cone as our
basis.
Michael Orlitzky [Sun, 7 Jun 2015 01:16:27 +0000 (21:16 -0400)]
Begin to clear up the mysteries of why the restriction doesn't work.
Everything's still a mess, but at least the tests are passing. No more
do we check for exact equality between K_W and K_star_W_star; instead
we just check that many properties agree. In the paper we have an
isomorphism, and equality holds. However, my isomorphism isn't an
isomorphism when using a coordinate system -- and we have to use a
non-normal one in Sage because we can't normalize vectors over
QQ. Huh.
Michael Orlitzky [Sat, 6 Jun 2015 03:40:29 +0000 (23:40 -0400)]
Use built-in Gram-Schmidt to make things a little better.
Michael Orlitzky [Fri, 5 Jun 2015 22:49:52 +0000 (18:49 -0400)]
Commit a big fucking mess while I refactor the span restriction.
Michael Orlitzky [Thu, 4 Jun 2015 18:11:29 +0000 (14:11 -0400)]
Update existing tests to use the codim() function.
Michael Orlitzky [Thu, 4 Jun 2015 17:33:07 +0000 (13:33 -0400)]
Add the cone codim() function.
Michael Orlitzky [Thu, 4 Jun 2015 16:52:22 +0000 (12:52 -0400)]
Switch existing tests to use the lineality() method.
Michael Orlitzky [Thu, 4 Jun 2015 16:51:34 +0000 (12:51 -0400)]
Add a cone lineality function.
Michael Orlitzky [Wed, 3 Jun 2015 20:25:52 +0000 (16:25 -0400)]
Remove unused code and implement the improved Lyapunov rank algorithm.
Michael Orlitzky [Wed, 3 Jun 2015 19:48:02 +0000 (15:48 -0400)]
Add the project_span() function.
Michael Orlitzky [Mon, 1 Jun 2015 15:20:01 +0000 (11:20 -0400)]
Add some more reduction theorem test cases.
Michael Orlitzky [Mon, 1 Jun 2015 05:39:33 +0000 (01:39 -0400)]
Remove the broken isomorphism tests leaving the reduction test working.
Michael Orlitzky [Mon, 1 Jun 2015 05:33:58 +0000 (01:33 -0400)]
Try the span_iso approach to fix my isomorphism tests.
Michael Orlitzky [Mon, 1 Jun 2015 00:33:11 +0000 (20:33 -0400)]
Commit the good (?) version of cone.py.
Michael Orlitzky [Mon, 1 Jun 2015 00:32:51 +0000 (20:32 -0400)]
Commit the busted version of cone.py.
Michael Orlitzky [Sun, 31 May 2015 00:05:32 +0000 (20:05 -0400)]
Add a few tests for LL(K).
Michael Orlitzky [Sat, 30 May 2015 22:41:01 +0000 (18:41 -0400)]
Fix the LL(K) code.
Implement lyapunov_rank(K) in terms of LL(K).
Michael Orlitzky [Sun, 24 May 2015 16:42:59 +0000 (12:42 -0400)]
First attempt at implementing LL(K); it doesn't work.
Michael Orlitzky [Sun, 24 May 2015 15:59:27 +0000 (11:59 -0400)]
Add citations for Lyapunov rank examples/tests.
Stub out LL(K).
Michael Orlitzky [Sun, 24 May 2015 14:50:25 +0000 (10:50 -0400)]
Remove is_full_space() and random_cone(); see Sage trac #18454.
Michael Orlitzky [Tue, 19 May 2015 19:09:02 +0000 (15:09 -0400)]
Factor out the is_full_space() function.
Michael Orlitzky [Tue, 19 May 2015 18:15:23 +0000 (14:15 -0400)]
Don't count on 2*dim(V) rays generating V.
Michael Orlitzky [Tue, 19 May 2015 15:02:07 +0000 (11:02 -0400)]
Ensure that we generate min_rays generators; add more cone tests.
Michael Orlitzky [Tue, 19 May 2015 03:41:41 +0000 (23:41 -0400)]
Simplify the random_cone() code by defaulting to lower bounds of zero.
Michael Orlitzky [Mon, 18 May 2015 01:49:27 +0000 (21:49 -0400)]
Factor out the discrete_complementarity_set() function.
Michael Orlitzky [Thu, 14 May 2015 20:18:17 +0000 (16:18 -0400)]
Add a random_cone() function and use it in two tests.
Michael Orlitzky [Mon, 11 May 2015 14:59:20 +0000 (10:59 -0400)]
Add basic Lyapunov rank implementation for polyhedral cones.
Michael Orlitzky [Thu, 13 Nov 2014 00:36:03 +0000 (19:36 -0500)]
Add the matrix_of_transformation function to matrix_vector.py.
Michael Orlitzky [Tue, 4 Nov 2014 17:54:34 +0000 (12:54 -0500)]
Import mjo.cone.* by default.
Michael Orlitzky [Tue, 4 Nov 2014 17:21:22 +0000 (12:21 -0500)]
Add random_doubly_nonnegative() and random_extreme_doubly_nonnegative() functions.
Michael Orlitzky [Tue, 4 Nov 2014 16:04:04 +0000 (11:04 -0500)]
Add is_admissible_extreme_rank() and use it to implement has_admissible_extreme_rank().
Michael Orlitzky [Tue, 4 Nov 2014 14:57:53 +0000 (09:57 -0500)]
Add the random_psd() function.
Michael Orlitzky [Mon, 3 Nov 2014 14:49:02 +0000 (09:49 -0500)]
Implement the is(_extreme)_completely_positive functions.
Michael Orlitzky [Mon, 3 Nov 2014 03:04:52 +0000 (22:04 -0500)]
Implement the is_extreme_doubly_nonnegative() function, first shot.
Michael Orlitzky [Mon, 3 Nov 2014 03:03:52 +0000 (22:03 -0500)]
Add the E() basis matrix function.
Michael Orlitzky [Mon, 3 Nov 2014 03:03:08 +0000 (22:03 -0500)]
Convert "n" to an integer explicitly in has_admissible_extreme_rank().
Michael Orlitzky [Mon, 3 Nov 2014 03:00:27 +0000 (22:00 -0500)]
Add the matrix_vector module with only the isomorphism() function for now.
Michael Orlitzky [Mon, 3 Nov 2014 01:24:45 +0000 (20:24 -0500)]
Comment/whitespace cleanup.
Michael Orlitzky [Thu, 30 Oct 2014 18:24:00 +0000 (14:24 -0400)]
doubly_nonnegative.py: Begin implementing the extreme vector algorithms.
Michael Orlitzky [Thu, 30 Oct 2014 18:23:32 +0000 (14:23 -0400)]
Whitespace fix.
Michael Orlitzky [Thu, 30 Oct 2014 18:23:16 +0000 (14:23 -0400)]
Whitespace fix.
Michael Orlitzky [Thu, 30 Oct 2014 18:04:33 +0000 (14:04 -0400)]
Refactor symmetric_pds/doubly_nonnegative.
Ensure that we're working over an amenable base ring in factor_psd.
New module: completely_positive (cone).
Michael Orlitzky [Thu, 30 Oct 2014 12:37:04 +0000 (08:37 -0400)]
Add __init__.py for mjo.cone.
Michael Orlitzky [Thu, 30 Oct 2014 12:36:49 +0000 (08:36 -0400)]
New module: cone.doubly_nonnegative.
Michael Orlitzky [Thu, 30 Oct 2014 12:34:20 +0000 (08:34 -0400)]
New module: cone.symmetric_psd.
Michael Orlitzky [Wed, 29 Oct 2014 22:25:24 +0000 (18:25 -0400)]
Disable terminal colors in init.sage.
Michael Orlitzky [Wed, 29 Oct 2014 22:24:52 +0000 (18:24 -0400)]
Add the matrix_simplify_full() function to symbolic.py.
Michael Orlitzky [Fri, 24 Oct 2014 21:23:59 +0000 (17:23 -0400)]
Update SymbolSequence to use underscores.
Michael Orlitzky [Fri, 24 Oct 2014 16:45:55 +0000 (12:45 -0400)]
Fix a comment.
Michael Orlitzky [Fri, 24 Oct 2014 15:36:16 +0000 (11:36 -0400)]
Add a note to the README about `sage -t` not loading init.sage.
Michael Orlitzky [Fri, 24 Oct 2014 15:21:39 +0000 (11:21 -0400)]
Fix a (harmless) doctest failure in mjo.interpolation.
Michael Orlitzky [Thu, 16 Oct 2014 19:27:01 +0000 (15:27 -0400)]
Use ~/src/sage.d as library path.
Michael Orlitzky [Fri, 9 May 2014 07:10:38 +0000 (03:10 -0400)]
Add the matrix_subs_expr function.
Michael Orlitzky [Tue, 15 Apr 2014 01:08:56 +0000 (21:08 -0400)]
Override list_plot (in addition to the already-overriden plot()).
Michael Orlitzky [Tue, 15 Apr 2014 01:08:34 +0000 (21:08 -0400)]
Don't override matplotlib's LaTeX installation.
Michael Orlitzky [Fri, 16 Aug 2013 20:51:47 +0000 (16:51 -0400)]
Remove medium_simplify() now that simplify_radical() was removed from full_simplify().
Michael Orlitzky [Mon, 4 Mar 2013 23:58:59 +0000 (18:58 -0500)]
Add doctests for set_simplification_domain().
Michael Orlitzky [Wed, 5 Dec 2012 00:15:32 +0000 (19:15 -0500)]
Define product = prod, which is provided by Sage.
Michael Orlitzky [Tue, 4 Dec 2012 23:50:41 +0000 (18:50 -0500)]
Add a test for the empty product.
Add some backticks.
Michael Orlitzky [Tue, 4 Dec 2012 23:44:46 +0000 (18:44 -0500)]
Use SR.var() instead of var() everywhere.
Michael Orlitzky [Wed, 28 Nov 2012 02:54:39 +0000 (21:54 -0500)]
Add the lagrange_interpolate() function, which lets you avoid evaluating your function (if you have one) before calling lagrange_polynomial().
Michael Orlitzky [Sun, 18 Nov 2012 22:53:21 +0000 (17:53 -0500)]
Add an example of a least squares problem.
Michael Orlitzky [Sun, 18 Nov 2012 07:23:54 +0000 (02:23 -0500)]
Fix legendre_p for finite field elements.
Michael Orlitzky [Sun, 18 Nov 2012 05:26:11 +0000 (00:26 -0500)]
Add more tests and examples.
Fix the orthogonality test.
Michael Orlitzky [Sun, 18 Nov 2012 01:33:45 +0000 (20:33 -0500)]
Add a few tests for the shifted Legendre polynomials.
Michael Orlitzky [Sun, 18 Nov 2012 01:15:26 +0000 (20:15 -0500)]
Combine the standard/general Legendre polynomial functions.
Michael Orlitzky [Sat, 17 Nov 2012 20:53:14 +0000 (15:53 -0500)]
Add the orthogonal_polynomials module with the Legendre functions.
Michael Orlitzky [Tue, 6 Nov 2012 03:42:52 +0000 (22:42 -0500)]
Remove the __call__() code and refactor everything into subcalls of __getitem__().
Michael Orlitzky [Tue, 6 Nov 2012 01:50:22 +0000 (20:50 -0500)]
Update all doctests to use index[] notation instead of __call__.
Michael Orlitzky [Mon, 5 Nov 2012 21:45:01 +0000 (16:45 -0500)]
Allow tuples to be passed in square brackets.
Michael Orlitzky [Sun, 4 Nov 2012 21:32:50 +0000 (16:32 -0500)]
Clean up a little bit of the interpolation code.
Michael Orlitzky [Sun, 4 Nov 2012 21:12:49 +0000 (16:12 -0500)]
Add the SymbolSequence class; eventually headed upstream.
Michael Orlitzky [Sun, 4 Nov 2012 21:12:17 +0000 (16:12 -0500)]
Add the lagrange_psi function.
Michael Orlitzky [Fri, 26 Oct 2012 19:20:25 +0000 (15:20 -0400)]
Add the Hermite functions to interpolation.py.
Michael Orlitzky [Fri, 26 Oct 2012 18:22:56 +0000 (14:22 -0400)]
Make the divided_difference() function rely on divided_difference_coefficients().
Make divided_difference() and newton_polynomial() take a list of function values instead of the function itself.
Michael Orlitzky [Fri, 26 Oct 2012 18:13:03 +0000 (14:13 -0400)]
Add the divided_difference_coefficients() function.