38 commits

Author	SHA1	Message	Date
James Hensman	52ba8e4ba3	remo0ved slices from models ... slices are now handles by special indexing kern parts, such as coregionalisation, independent_outputs. The old slicing functionality has been removed simply to clean up the code a little. Now that input_slices still exist (and will continue to be useful) in kern.py. They do need a little work though, for the psi-statistics	2013-04-28 17:22:04 +01:00
Nicolo Fusi	cc158d64a9	fixed import error	2013-02-15 13:39:01 +00:00
Nicolo Fusi	2abba8cf14	merged with master	2013-02-15 11:56:37 +00:00
James Hensman	5447d6fbfc	Assorted work on combining the EP and sparse methods	2013-02-01 17:12:45 +00:00
James Hensman	7dfbcebb87	some tidying in the likelihood classes	2013-02-01 09:47:30 +00:00
Nicolo Fusi	be8417315c	minor changes	2013-01-31 13:38:05 +00:00
Nicolo Fusi	49d2e4e4f6	merged master	2013-01-31 09:57:40 +00:00
Nicolo Fusi	2493e2d336	integrated sparse GP regression and BGPLVM classes	2013-01-30 15:51:36 +00:00
Nicolo Fusi	8a3e10700d	massive merge of the debug branch	2013-01-30 15:30:54 +00:00
Nicolo Fusi	e1edb062ed	minor changes	2013-01-30 13:20:01 +00:00
Nicolo Fusi	ece4e2442c	working on cross terms	2013-01-30 11:18:22 +00:00
Nicolo Fusi	10c774e84e	new shape for psi2	2013-01-29 16:10:12 +00:00
Nicolo Fusi	936d08723e	BGPLVM working with rbf+white	2013-01-29 14:02:41 +00:00
Nicolo Fusi	8b6e244cf1	decent gradients for most parameters	2013-01-28 15:55:40 +00:00
James Hensman	0fb8ab91bb	various hackday stuff, including scale factor in sparse GP	2013-01-18 17:47:50 +00:00
James Hensman	126b45673b	Scale factor added to sparse_GP_regression ... and sparse_GP_demo ammended to be less annoying (m1)	2013-01-18 15:09:48 +00:00
Neil Lawrence	b43db3f8da	Modified log_like_gradients to make it _log_like_gradients and moved extract_gradient to _log_like_gradients_transformed.	2013-01-18 13:47:37 +00:00
Neil Lawrence	1674bc529b	expand_param and extract_param replaced with set_params_transformed and get_params_transformed	2013-01-18 13:37:17 +00:00
Neil Lawrence	3c700dfa50	Changed get_param and set_param to _get_params and _set_params	2013-01-18 12:31:37 +00:00
James Hensman	9f05744374	general changes to bebugging code	2013-01-16 11:21:55 +00:00
James Hensman	4241a7c24f	simplified the debug classes	2013-01-15 09:34:47 +00:00
James Hensman	1a135ca9f7	some code to debug the sprase GP gradients with.	2013-01-11 17:54:10 +00:00
James Hensman	57934d82a6	sphinx configuratino for readthedocs.org	2013-01-09 15:45:39 +00:00
James Hensman	aae006411a	GP_regression and sparse_GP_regression now only return the full ... posterior covariance matrix when requested.	2012-12-21 11:42:39 +00:00
James Hensman	cc3ccf1ddc	reduced the memory requirements of the sparse GP ... by a factor of M!	2012-12-18 15:51:39 +00:00
James Hensman	5f2c2a04f8	chenged a little of the notation in the sparse GP ... This should allow easier implementation of het. noise	2012-12-08 15:15:31 -08:00
James Hensman	ab56f3f9b7	small tweak to the gradients in sparse GP	2012-12-07 23:03:53 -08:00
James Hensman	c72d1b1a5b	changed the behaviour of pdinv: now returns L and Li as well as the inverse	2012-12-06 09:51:13 -08:00
James Hensman	c003b2f34d	Minor edits	2012-12-05 09:10:27 -08:00
James Hensman	5f22bd00e6	made uncertain inputs a simple swith in the sparse GP class. This simplifies the inherritance structure	2012-12-03 18:35:05 +00:00
James Hensman	c4761bedf5	minor commenting	2012-12-02 15:12:14 +00:00
James Hensman	1d8104bcc2	general tidying in sparse_GP_regression ... Have also made a small ampount of headway in enabling heteroscedatic noise.	2012-12-02 13:40:19 +00:00
James Hensman	365d069ae8	saved a little computation by exploiting the symmetry of a matrix	2012-12-02 12:35:54 +00:00
James Hensman	1c60e50fed	I think the gradients bug in the sparse GP model is due to Kmm being unstable to invert. REducing M in some of the examples really helps	2012-12-02 12:32:20 +00:00
James Hensman	821701142a	tidied up some commented code from sparse_GP_regression	2012-11-30 12:43:14 +00:00
James Hensman	ba84a43ea3	sparse GP regression now working on this branch	2012-11-30 11:47:35 +00:00
Nicolo Fusi	92adeecb6c	added copyright notice and license at the top	2012-11-29 16:39:20 +00:00
Nicolo Fusi	61984274dd	models	2012-11-29 16:32:48 +00:00