2649598c64
FIX Enforcing y to be numeric in GPR and fixing python3 issue with maps
2015-10-19 18:07:14 +02:00
3957fc3b34
ENH GPR.log_marginal_likelihood() returns the current log-likelihood if no theta vector is provided
2015-10-19 18:07:14 +02:00
8f77ced609
MISC Changing default for copy_X_train to True in GaussianProcessClassifier
2015-10-19 18:07:14 +02:00
e2a340a4ee
REF More meaningful names for hyperparameters of GP kernels
2015-10-19 18:07:14 +02:00
2dd1f285b9
PEP8 Making pep8 and pyflakes happy
2015-10-19 18:07:14 +02:00
1c4b23f47b
MISC copy_X_train defaults to True in GPR and GPC
2015-10-19 18:07:14 +02:00
e87fad31fb
REF Renaming max_iter to max_iter_predict in GaussianProcessClassifier
2015-10-19 18:07:14 +02:00
c6db24e56f
REF Several small changes based on @eickenberg's suggestions
2015-10-19 18:07:13 +02:00
d34b7ddbfc
REF Realizing multi-class GPC bei delegating to OneVsRest or OneVsOne
...
Delegation is more flexible than subclassing.
2015-10-19 18:07:13 +02:00
34506ad705
COS Making BinaryGaussianProcessClassifierLaplace private
2015-10-19 18:07:13 +02:00
032c2a763b
FIX Added missing import of ConstantKernel in gpc.py
2015-10-19 18:07:13 +02:00
83bc6c2440
ENH Adding max_iter parameter for Newton iteration in GPC
2015-10-19 18:07:13 +02:00
c57aa64312
ENH X_train_ can be copied if requested
2015-10-19 18:07:13 +02:00
adc7451201
PEP8 Fixing PEP8 issues in gaussian_process package
2015-10-19 18:07:13 +02:00
59d754ba90
FIX Hiding method decision_function in GaussianProcessClassifier
2015-10-19 18:07:13 +02:00
768c81fd1b
TST Further GP-related fixes
2015-10-19 18:07:13 +02:00
92a6354aaf
TST Fix GPC and GPR such that all tests are ok
2015-10-19 18:07:13 +02:00
27b30166c3
FIX Several minor changes/fixes in gaussian_process module
2015-10-19 18:07:12 +02:00
3a312625f5
ENH Support for multi-class GPC added based on OneVsRestClassifier
2015-10-19 18:07:12 +02:00
07cd4d4e05
ENH GPC can use an externally defined optimizer for hyperparameter tuning
2015-10-19 18:07:12 +02:00
bcd4b72897
REFACTOR Remove theta_ attribute from GPR and GPC
...
It can be accessed instead via kernel_.theta
2015-10-19 18:07:11 +02:00
4db5142bb8
REFACTOR GP kernel hyperparameters theta and their bounds are log-tranformed
...
This representation of the search space is more amenable for
hyperparameter search, as hyperparameters like
length-scales naturally live on a log-scale.
2015-10-19 18:07:11 +02:00
22f049b6bc
ENH GaussianProcessClassifier supports multiple restarts of the optimizer
2015-10-19 18:07:11 +02:00
caa4196c3a
DOC Revised docstrings of kernels.py
2015-10-19 18:07:10 +02:00
7773707e0a
DOC gpc module fully documented
2015-10-19 18:07:10 +02:00
1c70e52dd6
ENH Adding diag(X) method to GP kernels.
...
Using this method rather than np.diag(kernel(X)) or np.apply_along_axis
reduces memory-consumption and speeds up computation considerably
2015-10-19 18:04:56 +02:00
fb4eded5ee
ENH GPC can use warm-starts in _posterior_mode()
2015-10-19 18:04:56 +02:00
fd2cfc868c
ENH Reducing memory-consumption of GPC.predict_proba()
...
We compute the diagonal of var_f_star directly and not the whole matrix.
2015-10-19 18:04:56 +02:00
3043a126a4
FIX GPC uses LabelEncoder in fit() and returns (n_samples, n_classes) array in predict_proba()
2015-10-19 18:04:56 +02:00
e3423b989a
FIX Minor bugfixes and downward-compatibility/python3 issue
2015-10-19 18:04:56 +02:00
2feb6e02cf
REFACTOR Replacing all calls to deepcopy() by clone()
2015-10-19 18:04:55 +02:00
403b5f1412
REFACTOR REnamed kernel property params to theta
2015-10-19 18:04:55 +02:00
d0bc5dc35c
FIX GPs do not modify kernel attribute but store optimized kernels in kernel_
2015-10-19 18:04:55 +02:00
d8405c2403
REFACTOR Hyperparameter optimization in GPs enabled explicitly by optimizer parameter
2015-10-19 18:04:55 +02:00
e05daaec52
ENH Avoid to create square matrix W_sr in all but one place (TODO)
2015-10-19 18:04:55 +02:00
e165fed1cc
REFACTOR GaussianProcessClassification/Regression -> ...Classifier/Regressor
2015-10-19 18:04:55 +02:00
03e35758f8
REFACTOR Merging kernel methods auto() and cross() in __call__()
2015-10-19 18:04:54 +02:00
2ab097f0d3
TST Dealing with failing tests from test_common
2015-10-19 18:04:54 +02:00
a5dc3491ad
MISC Revised gaussian_process package based on @eickenberg's suggestions
2015-10-19 18:04:54 +02:00
515defe7a4
REFACTOR auto_correlation() -> auto(), cross_correlation() -> cross()
2015-10-19 18:04:54 +02:00
b76ca723fc
ADD Gaussian Process classification based on Laplace approximation
2015-10-19 18:04:54 +02:00
Jan Hendrik Metzen