James Hensman
|
4d1feb9d9d
|
first draft of DTC
|
2014-02-11 16:27:08 +00:00 |
|
James Hensman
|
103aaebb29
|
dumb merge conflict in a comment
|
2014-02-11 13:44:18 +00:00 |
|
Max Zwiessele
|
b944427733
|
rename dK_dtheta > gradients_X
|
2014-02-11 12:20:57 +00:00 |
|
Max Zwiessele
|
6a1c700c03
|
psi stat and kernel tests new parameterization
|
2014-02-11 12:20:16 +00:00 |
|
Max Zwiessele
|
9e1546524e
|
linear kern variational updates
|
2014-02-11 12:17:07 +00:00 |
|
Max Zwiessele
|
ed87e6bbd2
|
final touches to gradchecker
|
2014-02-10 17:00:50 +00:00 |
|
Max Zwiessele
|
32491abaff
|
gradchecker now with fixed inputs
|
2014-02-10 16:42:00 +00:00 |
|
Max Zwiessele
|
9eef4ebded
|
fixes added for gradchecking
|
2014-02-10 16:08:11 +00:00 |
|
Max Zwiessele
|
a13d6ca894
|
Merge branch 'params' of github.com:SheffieldML/GPy into params
Conflicts:
GPy/core/parameterization/param.py
|
2014-02-10 16:02:57 +00:00 |
|
Max Zwiessele
|
826d2f04ff
|
checkgrad now global and callable from any parameter
|
2014-02-10 16:01:55 +00:00 |
|
Alan Saul
|
a2b986dc5c
|
Fixed a couple of small params bugs
|
2014-02-10 15:49:50 +00:00 |
|
Alan Saul
|
b72d319909
|
Merged parameterized fixing
|
2014-02-10 15:42:47 +00:00 |
|
Alan Saul
|
6cbf810856
|
Laplace now appears to be grad checking again
|
2014-02-10 15:40:06 +00:00 |
|
Alan Saul
|
c6d466e72d
|
Moved fix parameter to constrainable
|
2014-02-10 15:39:54 +00:00 |
|
Max Zwiessele
|
6a068775f5
|
Merge branch 'params' of github.com:SheffieldML/GPy into params
Conflicts:
GPy/core/parameterization/param.py
GPy/core/parameterization/parameter_core.py
GPy/core/parameterization/parameterized.py
|
2014-02-10 15:21:09 +00:00 |
|
Max Zwiessele
|
e0c68d5eb3
|
_highest_parent_ now follows the tree, dK_dX > gradient_X, added update_grads_variational to linear, bgplvm for new framework
|
2014-02-10 15:12:49 +00:00 |
|
Alan Saul
|
d2a0e4a265
|
Stupid error, needed to actually USE the gradients in student t... Looks like s2 of rasm's may have an extra -? dW_df == -d2logpdf_df not just d2logpdf_df?
|
2014-02-10 12:54:49 +00:00 |
|
Alan Saul
|
fc44478ed2
|
Have most of the likelihood testing working, laplace likelihood parameters need fixing, some of the signs are wrong I believe
|
2014-02-10 12:29:09 +00:00 |
|
Alan Saul
|
0f263d2ff2
|
Have most of the likelihood testing working, laplace likelihood parameters need fixing, some of the signs are wrong I believe
|
2014-02-10 12:28:24 +00:00 |
|
Alan Saul
|
625943ef27
|
Merge branch 'params' of github.com:SheffieldML/GPy into params
|
2014-02-07 15:16:55 +00:00 |
|
Alan Saul
|
186feb45a1
|
Fixed likelihood tests for new parameters structure
|
2014-02-07 15:16:52 +00:00 |
|
Alan Saul
|
c28f11f291
|
Fixed parameter bugs
|
2014-02-07 15:16:05 +00:00 |
|
James Hensman
|
cf879a39a4
|
Merge branch 'params' of github.com:SheffieldML/GPy into params
|
2014-02-06 18:17:09 +00:00 |
|
James Hensman
|
7d2295d854
|
some messing with fitc
|
2014-02-06 18:17:04 +00:00 |
|
Max Zwiessele
|
87dab55fe1
|
adjusted periodic exponential to the new parameterization
|
2014-02-06 16:22:45 +00:00 |
|
Alan Saul
|
648609c3b1
|
Merge branch 'params' of github.com:SheffieldML/GPy into params
|
2014-02-06 16:22:18 +00:00 |
|
Alan Saul
|
b12fb6a2a8
|
Fixed parameterized oddity where it was updating all constrained parameters as soon as any were constrained rather than after all are constrained@
@
|
2014-02-06 16:22:08 +00:00 |
|
James Hensman
|
f5f2c56f7c
|
spelling
|
2014-02-06 14:36:24 +00:00 |
|
Max Zwiessele
|
1c7949a61c
|
small changes to parameterization init
|
2014-02-06 14:03:40 +00:00 |
|
Max Zwiessele
|
0d52430ffb
|
fixed product kernel copy error
|
2014-02-06 14:03:15 +00:00 |
|
James Hensman
|
1c9151a7d0
|
added caching framework
|
2014-02-06 12:46:56 +00:00 |
|
James Hensman
|
36456af5d4
|
Merge branch 'params' of github.com:SheffieldML/GPy into params
|
2014-02-06 12:33:04 +00:00 |
|
James Hensman
|
54272d22bb
|
first crack at a caching object
|
2014-02-06 12:33:00 +00:00 |
|
Max Zwiessele
|
ef3bfa4676
|
highest parent fix
|
2014-02-06 12:00:43 +00:00 |
|
James Hensman
|
6346af8764
|
assorted fixes
|
2014-02-06 09:28:27 +00:00 |
|
James Hensman
|
75241ecf89
|
predictino working nicely for laplace
|
2014-02-05 17:52:17 +00:00 |
|
James Hensman
|
80629e00b6
|
fiddling with plotting
|
2014-02-05 17:12:52 +00:00 |
|
James Hensman
|
432d9668a6
|
merged simple conflict£
|
2014-02-05 16:26:10 +00:00 |
|
James Hensman
|
f653bc430e
|
an afternoon's work on the laplace approximation
|
2014-02-05 16:23:35 +00:00 |
|
James Hensman
|
2ff7e286ee
|
more owrk on the Laplace approx
|
2014-02-05 10:56:48 +00:00 |
|
Max Zwiessele
|
b2328c4f47
|
starting varDTC with uncertain inputs [not working]
|
2014-02-05 10:48:23 +00:00 |
|
Zhenwen Dai
|
70e7d72bf2
|
add spike-and-slab gplvm kernel [unfinished].]
|
2014-02-03 09:12:43 +00:00 |
|
Max Zwiessele
|
1b49f7ab30
|
not calling self.parameters_changed explicitly anymore -> not needed
|
2014-01-31 17:01:50 +00:00 |
|
James Hensman
|
399adb1b00
|
some documenting, and fiddling with the laplace approx
|
2014-01-31 16:59:06 +00:00 |
|
James Hensman
|
9f40ab0f83
|
sparse GP now working nicely
|
2014-01-31 10:15:30 +00:00 |
|
James Hensman
|
e0dbfbe148
|
sparse GP now checkgrads, optimises sensibly. Predicitno still not working
|
2014-01-30 12:03:02 +00:00 |
|
James Hensman
|
a93312c306
|
Merge branch 'params' of github.com:SheffieldML/GPy into params
|
2014-01-29 17:02:56 +00:00 |
|
James Hensman
|
a31ff3acc3
|
some hacking on sparse_gp inference
|
2014-01-29 17:02:44 +00:00 |
|
Zhenwen Dai
|
9a7d9fa5b7
|
adapter laplace inference into the param framework
|
2014-01-29 10:28:39 +00:00 |
|
James Hensman
|
ae03b63afb
|
tidied up sparse_gp_regression
|
2014-01-29 09:40:22 +00:00 |
|