4405 commits

This branch

This branch

All branches

Author	SHA1	Message	Date
Ricardo	68e37e8684	The next step is to optimize the noise models' parameters	2013-07-10 19:33:43 +01:00
Max Zwiessele	be5a43a2c2	improved docstring for MRD	2013-07-10 16:52:20 +01:00
Max Zwiessele	266ec23f8a	GPLVM get/setstate added	2013-07-10 16:51:41 +01:00
Max Zwiessele	e8f2bec195	SCG optimizer now greatly improved in printing	2013-07-10 16:49:51 +01:00
Max Zwiessele	69e31f3dc1	SCG optimizer now greatly improved in printing	2013-07-10 16:49:45 +01:00
Alan Saul	57001851c4	Trying to debug kernel parameters learning (fails even when noise fixed) ... may be some instablility, seems like it can get it if it starts close	2013-07-09 20:05:03 +01:00
Ricardo	06ffb884ab	files re-organized	2013-07-09 17:54:56 +01:00
Alan Saul	ee980227ac	Fixed 2*variance plotting instead of 2*std plotting, tidied up	2013-07-09 11:35:42 +01:00
Ricardo	b3fd843d33	confidence interval fixed	2013-07-08 18:33:44 +01:00
Alan Saul	2a366619b3	Changed incorrect naming	2013-07-08 16:09:20 +01:00
James Hensman	2ab1ac3376	added link to user mailing list	2013-07-08 16:06:36 +01:00
Alan Saul	4e5cefb4b5	Reparameratised in terms of sigma2	2013-07-08 15:48:53 +01:00
James Hensman	3ad48534c8	changes to the hierarchical kernpart. ... Looks to work now.	2013-07-08 15:29:37 +01:00
Nicolo Fusi	f19a26a006	added constructor for hierarchical kernel	2013-07-08 14:50:13 +01:00
Nicolo Fusi	d887e1ceda	merged	2013-07-08 14:28:54 +01:00
Nicolo Fusi	074529c1c0	changes to multiout constructor	2013-07-08 14:27:32 +01:00
James Hensman	58a59dcab0	created a hierarchical kernel	2013-07-08 14:20:42 +01:00
James Hensman	9c85911b87	Merge branch 'devel' of github.com:SheffieldML/GPy into devel	2013-07-08 13:06:11 +01:00
James Hensman	db7485b906	fixed a bug in constructor of periodic_matern52	2013-07-08 13:06:02 +01:00
Ricardo	b856c60d30	massive changes	2013-07-02 18:18:11 +01:00
Alan Saul	ab6a3a571e	Playing trying to find what makes it want to go so low	2013-07-02 11:14:48 +01:00
Ricardo	4054442462	more files	2013-07-01 20:31:16 +01:00
Alan Saul	a7169ab1ab	Fixed bug where B wasn't refering to current f location	2013-07-01 15:21:47 +01:00
Alan Saul	26b3855af5	Everything seems to be gradchecking again	2013-07-01 10:06:20 +01:00
Ricardo	7361d311c1	further corrections	2013-06-30 23:36:37 +01:00
Max Zwiessele	a8eb7eb5f7	minor adjustments	2013-06-28 11:13:20 +01:00
Max Zwiessele	96829862cc	minor printing improvements	2013-06-28 11:13:12 +01:00
Max Zwiessele	29746cd8be	added value restoring if unpickling objects	2013-06-28 11:02:59 +01:00
Max Zwiessele	12640a2d17	plot ard ticks improved	2013-06-28 11:02:15 +01:00
Max Zwiessele	c0d514b6c0	optional plotting of inducing inputs added	2013-06-28 11:01:31 +01:00
Max Zwiessele	7325e319b4	plot_latent enhancements	2013-06-28 11:00:42 +01:00
Max Zwiessele	08a902ed6c	massively improved printing	2013-06-28 11:00:09 +01:00
Max Zwiessele	66a6bde715	more robust gradient clippinggit stat	2013-06-28 10:59:29 +01:00
Alan Saul	c90b1f0c99	Added minimizer for finding f, doesn't help	2013-06-27 15:04:57 +01:00
Max Zwiessele	3b35d6d321	Merge branch 'devel' of github.com:SheffieldML/GPy into devel	2013-06-27 11:18:10 +01:00
Max Zwiessele	9e9131b90d	plot_latent bug-fix of creating no figure	2013-06-27 11:17:58 +01:00
Max Zwiessele	8cdedf2edb	plot_latent bug-fix in mrd	2013-06-27 11:11:41 +01:00
Max Zwiessele	26260c355c	added docstring to domains	2013-06-27 10:40:23 +01:00
Ricardo	efa782c636	massive changes	2013-06-27 01:03:32 +01:00
Nicolas	b9ca93e8ab	Fixed bug in constructors	2013-06-26 18:41:05 +01:00
Nicolas	9deb1fc1c2	Merge branch 'devel' of github.com:SheffieldML/GPy into devel	2013-06-26 17:32:41 +01:00
Nicolas	97acb36b55	new constrain_negative negative_logexp (selected by default)	2013-06-26 17:29:37 +01:00
Max Zwiessele	761f7b02fc	_get/set_params into parameterized	2013-06-26 17:05:28 +01:00
Max Zwiessele	146dff7cbf	_get/set_params into parameterized	2013-06-26 17:04:10 +01:00
Max Zwiessele	d978d3e390	added tutorial for creating models	2013-06-26 17:02:48 +01:00
Max Zwiessele	c773672b69	params left in __get/setitem__	2013-06-26 17:02:35 +01:00
Max Zwiessele	f5effb8cb6	added robust pickling, switches to old behaviour, if get/setstate not implemented	2013-06-26 16:48:48 +01:00
Alan Saul	617d73ca32	Now checkgrads a lot more of the time, but still fails in optimisation, seems also odd that when parameter is fixed kernel parameters go to infinity	2013-06-26 15:44:26 +01:00
Ricardo	da108cc6d1	predictive mean done	2013-06-25 18:20:09 +01:00
Alan Saul	064efd5535	Added another optimisation which doesn't use gradients. Seems like F is almost always found, but Y can be off, suggesting that Wi__Ki_W is wrong, maybe W?	2013-06-25 18:20:00 +01:00