Commit messages

July 12, 2014

jpritikin committed r3645 in OpenMx at 03:12:
Refactor omxInitialMatrixAlgebraCompute
jpritikin committed r3644 in OpenMx at 03:12:
Repair constraint processing; add test
in /trunk:

July 11, 2014

jpritikin committed r3638 in OpenMx at 02:19:
Extensive rework of the conformability checking pass

July 7, 2014

jpritikin committed r3619 in OpenMx at 03:35:
Increase sanity of omxMarkClean/omxMarkDirty
jpritikin committed r3617 in OpenMx at 03:35:
Add a conformability checking step to the backend 2/2
in /trunk:
jpritikin committed r3616 in OpenMx at 03:34:
Add a conformability checking step to the backend 1/2
in /trunk:

June 27, 2014

jpritikin committed r3592 in OpenMx at 06:28:
Add matrix logarithm algebra operator
in /trunk:

June 16, 2014

jpritikin committed r3565 in OpenMx at 03:40:
CRAN prohibits variable length arrays

And that's probably a good thing too because declaring variable length
arrays on the stack can contribute to stack overflows.
in /trunk:

May 16, 2014

jpritikin committed r3472 in OpenMx at 23:17:
Add packageStartupMessage if compiled without OpenMP
in /trunk:
jpritikin committed r3468 in OpenMx at 09:02:
Remove extra copy of # of evaluations

Date: Fri, 16 May 2014 09:22:12 +0000
From: "Brick, Timothy" <brick@mpib-berlin.mpg.de>

I'm looking back at the logs, and I'm not sure exactly why that was.
The thing that springs to mind was that the second value was originally
a placeholder intended to eventually keep a separate number for row-wise
computations.  There was a reason somebody wanted a report of row-wise
computations, but I think we found an alternative or realized we weren't
that interested in it as a benchmark, so the idea got shelved.

I can't think of any other reason this would have been done, and I
imagine it's no longer important to keep.
in /trunk/src: glue.cpp (modified, previous: r3461)
Syndicate content