Ryan Curtin
|
c0f44eeacd
|
Fix syntax.
|
2015-02-06 16:38:34 -05:00 |
|
Ryan Curtin
|
da3c955207
|
Standardize name of test suite.
|
2015-02-06 16:37:56 -05:00 |
|
Ryan Curtin
|
62def05b67
|
Backport operator-(sp_mat) for Armadillo <= 4.000.
|
2015-02-06 16:18:15 -05:00 |
|
Ryan Curtin
|
22ae38ac0c
|
Remove tab character.
|
2015-02-06 16:06:14 -05:00 |
|
Ryan Curtin
|
ea795f1dee
|
Revert "older versions of arma do not support unary minus on sparse matrices"
This reverts commit a8ecf093989cbe3d58c36c4e21e772c99d7856e4.
|
2015-02-06 15:56:05 -05:00 |
|
Stephen Tu
|
cc6199a55e
|
Comment out SDP test which is failing randomly
|
2015-02-04 11:05:29 -08:00 |
|
Stephen Tu
|
2bfe20de84
|
cov matrices should be symmetric- fixes error in gaussian test case
|
2015-02-04 10:59:34 -08:00 |
|
Stephen Tu
|
4b4d22bdee
|
older versions of arma do not support unary minus on sparse matrices
|
2015-02-04 10:59:34 -08:00 |
|
Ryan Curtin
|
d48cf446ea
|
Fix documentation, thanks to Squalluca.
|
2015-02-03 19:44:28 -05:00 |
|
Ryan Curtin
|
bb79a6af95
|
Merge branch 'jaskaran1-master'
|
2015-02-03 09:50:27 -05:00 |
|
Ryan Curtin
|
5b313de0e4
|
Merge branch 'master' of https://github.com/jaskaran1/mlpack into jaskaran1-master
Conflicts:
README.md
|
2015-02-03 09:50:10 -05:00 |
|
Jaskaran Singh
|
59c7595950
|
Updated path of install in Readme
|
2015-02-03 14:46:02 +05:30 |
|
Ryan Curtin
|
a6e5686f94
|
Add new contributor.
|
2015-02-02 17:16:34 -05:00 |
|
Ryan Curtin
|
9984629087
|
Refactor notice about LD_LIBRARY_PATH.
|
2015-02-02 17:09:59 -05:00 |
|
Ryan Curtin
|
41d34cb73a
|
Merge pull request #398 from jaskaran1/master
shared object file issue after make install
|
2015-02-02 17:05:56 -05:00 |
|
Ryan Curtin
|
29694386bb
|
Slight code cleanups for style.
No functionality changes.
|
2015-02-02 15:15:16 -05:00 |
|
Ryan Curtin
|
19131546bf
|
Better commenting of the SDP class.
Make the dangers of invalid SDPs documented, among other things.
|
2015-02-02 15:05:55 -05:00 |
|
Ryan Curtin
|
07b0cbc1f4
|
Refactor to access SDPs through LRSDP::SDP().
(also change everything that used direct access through LRSDP.)
|
2015-02-02 14:59:16 -05:00 |
|
Jaskaran Singh
|
1a598d3831
|
.so problem fixed
.so problem fixed after doing a make install.
|
2015-02-03 00:29:19 +05:30 |
|
Jaskaran Singh
|
8869b455bc
|
.so problem fixed
|
2015-02-03 00:18:06 +05:30 |
|
Jaskaran Singh
|
dd6ee23e80
|
.so problem fixed
Shared Object(.so) file issue after doing a make install resolved.
|
2015-02-03 00:16:21 +05:30 |
|
Ryan Curtin
|
0195f97229
|
Merge branch 'stephentu-ip'
|
2015-02-02 11:03:50 -05:00 |
|
Ryan Curtin
|
356415177c
|
Merge branch 'ip' of https://github.com/stephentu/mlpack into stephentu-ip
|
2015-02-02 11:03:38 -05:00 |
|
Marcus Edel
|
c321d54b86
|
Use the correct input to accumulate the elements. Thanks to Shangtong for pointing this out.
|
2015-01-30 11:48:41 +01:00 |
|
Stephen Tu
|
6cbf1c2bf0
|
Update naive_method.hpp
Actually fix the comment
|
2015-01-29 11:27:56 -08:00 |
|
Stephen Tu
|
9fca52090b
|
Update naive_method.hpp
Fix comment
|
2015-01-29 11:19:31 -08:00 |
|
Stephen Tu
|
9a185ef289
|
add comments for sensing SDP test + remove redundant data
can reconstruct measurement vector from the original matrix
plus the sensing matrices.
|
2015-01-29 00:32:44 -05:00 |
|
Stephen Tu
|
e0ef104613
|
Add matrix sensing SDP test case
|
2015-01-29 00:18:39 -05:00 |
|
Stephen Tu
|
88758c7959
|
add back missing edge constraint
|
2015-01-29 00:18:39 -05:00 |
|
Stephen Tu
|
0801db6bb4
|
Set the default starting penalty to 10
|
2015-01-29 00:18:39 -05:00 |
|
Stephen Tu
|
460fe44a33
|
Add a relative objective value termination criteria
|
2015-01-29 00:18:38 -05:00 |
|
Stephen Tu
|
73da878a3f
|
some comments for the algorithm implementation
|
2015-01-28 19:02:48 -08:00 |
|
Stephen Tu
|
6f0c814c9e
|
iteration=1 to allow maxIterations=0 to work
|
2015-01-28 18:34:31 -08:00 |
|
Stephen Tu
|
b29b5aef77
|
Inline alphahat into alpha
|
2015-01-28 18:34:03 -08:00 |
|
Stephen Tu
|
f1beedd837
|
use builtin diagmat
|
2015-01-28 18:27:44 -08:00 |
|
Stephen Tu
|
827c49f7c9
|
expose SDP object directly for LRSDPFunction
|
2015-01-28 18:26:24 -08:00 |
|
Stephen Tu
|
e183e8a5fb
|
remove unused variable
|
2015-01-28 18:26:15 -08:00 |
|
Ryan Curtin
|
eeae6f5b28
|
Properly handle old Armadillo versions.
Pre-4.500 doesn't have arguments to chol().
|
2015-01-26 15:25:10 -05:00 |
|
Ryan Curtin
|
cd0eae77eb
|
Merge pull request #390 from stephentu/master
Add covariance factorization caching to gaussian distribution
|
2015-01-26 20:09:05 +00:00 |
|
Ryan Curtin
|
061c1cb28e
|
Update bug tracker to Github.
|
2015-01-25 10:17:51 -05:00 |
|
Stephen Tu
|
8dc661cb9f
|
invert the lower triangular matrix the right way
thanks to rcurtin for the note
|
2015-01-22 11:59:16 -08:00 |
|
Ryan Curtin
|
8305b363f6
|
Minor syntax fixes.
|
2015-01-22 11:44:50 -05:00 |
|
Ryan Curtin
|
6e9e7731f1
|
Minor syntax fixes.
|
2015-01-22 11:41:54 -05:00 |
|
Stephen Tu
|
e7e17219b8
|
Merge remote-tracking branch 'mlpack/master' into ip
Conflicts:
src/mlpack/core/optimizers/sdp/lrsdp_function_impl.hpp
|
2015-01-21 18:11:55 -08:00 |
|
Stephen Tu
|
736f62c32d
|
fix up PrimalDualSolver::Optimize() return value
|
2015-01-21 18:09:41 -08:00 |
|
Stephen Tu
|
a76f4690c7
|
Port LR-SDP to use the new SDP problem object
|
2015-01-21 17:34:04 -08:00 |
|
Stephen Tu
|
1457061cae
|
iteration check is now iteration != maxIterations
allows for iterate until convergence
|
2015-01-21 14:52:28 -08:00 |
|
Stephen Tu
|
fca638d243
|
make the SDP objective matrix type a template parameter
|
2015-01-21 14:49:38 -08:00 |
|
Ryan Curtin
|
1677a5f9b0
|
Fix spacing for include warning.
|
2015-01-21 17:10:33 -05:00 |
|
Stephen Tu
|
6641bab361
|
use glibc constants
|
2015-01-21 13:26:50 -08:00 |
|