Commit graph

4405 commits

Author SHA1 Message Date
Max Zwiessele
d789e55481 stupid kern stash merge 2013-04-23 16:40:09 +01:00
Max Zwiessele
2b3b7350cd kern conflict 2013-04-23 16:37:13 +01:00
Max Zwiessele
264f0d21b6 kern stash conflict 2013-04-23 16:34:31 +01:00
Max Zwiessele
ebc3b8756e psi_stat_test stash 2013-04-23 16:34:01 +01:00
Max Zwiessele
68dd416640 Merge branch 'devel' into mrd 2013-04-23 16:21:57 +01:00
Max Zwiessele
389a04d2b5 bugfix: cross term psi1 bias + linear 2013-04-23 16:21:41 +01:00
Max Zwiessele
743112c448 psi1 not working (strange transposes) 2013-04-23 15:52:43 +01:00
Max Zwiessele
f01be172be moved *2. of psi2 statistics into kern and corrected bias+linear cross term 2013-04-23 15:22:30 +01:00
Max Zwiessele
d4e0194cf7 Merge branch 'devel' into mrd 2013-04-23 14:11:12 +01:00
Max Zwiessele
a71b4d80c0 Merge branch 'mrd' of github.com:SheffieldML/GPy into mrd 2013-04-23 14:10:53 +01:00
Max Zwiessele
2c3a53b174 psi stat tests done and failing gracefully 2013-04-23 14:10:38 +01:00
Max Zwiessele
0c8b83454f Revert "merge devel mrd"
This reverts commit 3f625a9347, reversing
changes made to dc6faeb303.
2013-04-23 14:02:15 +01:00
Max Zwiessele
3f625a9347 merge devel mrd 2013-04-23 13:51:17 +01:00
Max Zwiessele
dc6faeb303 psi stat tests 2013-04-23 13:44:31 +01:00
James Hensman
2205c333b2 fixed a weird regular expression bug in ensure_def_constraints 2013-04-23 12:19:41 +01:00
James Hensman
d402047ff3 more minor bugs 2013-04-23 12:01:10 +01:00
James Hensman
9109d451ab fixing small bug in independent outputs kern 2013-04-23 11:59:00 +01:00
James Hensman
f35578804a prod_orthogonal now caches the K matrices 2013-04-23 10:56:10 +01:00
James Hensman
f145141923 added a kernel for independent outputs 2013-04-23 10:02:12 +01:00
James Hensman
8bd017466d Nparam_transformed work better now
Before, counted the number of fixes, which failed when a fix fixed more
than one parameter...
2013-04-22 13:37:59 +01:00
James Hensman
698f52e5e3 GPy now fails silently if sympy is not present 2013-04-22 13:15:39 +01:00
James Hensman
56ecd4782a made the basic GP class use dtrtrs where possible 2013-04-22 11:59:32 +01:00
James Hensman
7fbcae2503 Merge branch 'devel' of github.com:SheffieldML/GPy into devel 2013-04-22 11:58:17 +01:00
Max Zwiessele
d0512f92b6 xticklabels improved 2013-04-21 10:58:46 +01:00
Alan Saul
267a8e427c Adding gradients, shapes starting to make sense 2013-04-19 17:41:01 +01:00
Alan Saul
1420aa532c Attempted to introduce gradient methods, won't work yet I doubt 2013-04-19 12:23:00 +01:00
Max Zwiessele
e79f94d74a Merge branch 'devel' into mrd 2013-04-19 10:02:32 +01:00
Max Zwiessele
10703e4774 demo changed, bgplvm still broken 2013-04-18 17:59:01 +01:00
andreas
25063fd53d pdinv passes extra args to jitchol 2013-04-18 16:39:55 +01:00
Max Zwiessele
865e9df255 BGPLVM still failing, doesn't seem to be numerical : ( 2013-04-17 15:45:20 +01:00
Alan Saul
7b44a4cb53 Merge remote-tracking branch 'upstream/devel' 2013-04-16 16:36:15 +01:00
Alan Saul
01671b6c57 Merged likelihood functions 2013-04-16 16:34:26 +01:00
Max Zwiessele
0ed7f8dfdd Merge branch 'devel' into mrd 2013-04-16 15:05:23 +01:00
Max Zwiessele
009b7314bf added bgplvm_simulation on same simulation 2013-04-16 15:04:25 +01:00
Nicolo Fusi
c5a3052f9f Merge branch 'devel' of github.com:SheffieldML/GPy into devel 2013-04-16 15:03:26 +01:00
Nicolo Fusi
06e811b2c3 small changes 2013-04-16 15:03:18 +01:00
Max Zwiessele
3baeeb1e35 Merge branch 'devel' into mrd 2013-04-16 12:38:37 +01:00
Max Zwiessele
aaf51f0e19 simulation data changes 2013-04-16 12:37:31 +01:00
James Hensman
ab50dc7cec a litle more stability in svigp
Another instance of dpotrs instead of dot
2013-04-16 12:36:15 +01:00
Max Zwiessele
9b182eb3f8 Merge branch 'mrd' into devel 2013-04-16 11:30:07 +01:00
Max Zwiessele
350497c726 adjusted plotting behaviour in X1d 2013-04-16 11:25:51 +01:00
Max Zwiessele
9acc6e9723 changed printing behaviour in cholesky to kill last line 2013-04-16 11:13:53 +01:00
James Hensman
445e6c1d30 comments only 2013-04-16 11:09:33 +01:00
James Hensman
403dc42462 improved stability of sparse GP computations
Specifically in computing dL_dKmm
2013-04-16 11:06:21 +01:00
James Hensman
dd27b285d6 rbf computation of psi2 now works if there's only one datum 2013-04-15 17:53:26 +01:00
Nicolo Fusi
d3ae9546f8 Merge branch 'mrd' into devel 2013-04-15 16:28:45 +01:00
Nicolo Fusi
7c63d7e346 removed useless _set_params() 2013-04-15 16:27:56 +01:00
Max Zwiessele
2b2edae083 new gradcheck for more stability in mrd_tests 2013-04-15 16:26:03 +01:00
Nicolo Fusi
4c38eea49e Merge branch 'mrd' into devel 2013-04-15 16:02:32 +01:00
Nicolo Fusi
671ae41256 small changes 2013-04-15 16:02:02 +01:00