apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-30 14:35:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merging

Browse source
This commit is contained in:
Alan Saul 2013-05-06 10:15:39 +01:00
parent bd346643db
commit f95666a8f9
2 changed files with 6 additions and 10 deletions
Download patch file Download diff file Expand all files Collapse all files

1
GPy/likelihoods/Laplace.py
View file

@ -98,6 +98,7 @@ class Laplace(likelihood):
A = I + C A = I + C
#plt.imshow(A) #plt.imshow(A)
#plt.show() #plt.show()
ki, _, _, _ = pdinv(self.K)
I_KW_i, _, _, _ = pdinv(A) I_KW_i, _, _, _ = pdinv(A)
#FIXME: Careful dK_dthetaK is not the derivative with respect to the marginal just prior K! #FIXME: Careful dK_dthetaK is not the derivative with respect to the marginal just prior K!

13
GPy/models/GP.py
View file

@ -6,15 +6,9 @@ import numpy as np
import pylab as pb import pylab as pb
from .. import kern from .. import kern
from ..core import model from ..core import model
<<<<<<< HEAD
from ..util.linalg import pdinv, mdot from ..util.linalg import pdinv, mdot
from ..util.plot import gpplot, x_frame1D, x_frame2D, Tango from ..util.plot import gpplot, x_frame1D, x_frame2D, Tango
from ..likelihoods import EP, Laplace from ..likelihoods import EP, Laplace
=======
from ..util.linalg import pdinv, mdot
from ..util.plot import gpplot, x_frame1D, x_frame2D, Tango
from ..likelihoods import EP
>>>>>>> upstream/devel
class GP(model): class GP(model):
""" """
@ -34,6 +28,7 @@ class GP(model):
""" """
def __init__(self, X, likelihood, kernel, normalize_X=False): def __init__(self, X, likelihood, kernel, normalize_X=False):
self.has_uncertain_inputs=False
# parse arguments # parse arguments
self.X = X self.X = X
@ -128,12 +123,12 @@ class GP(model):
Note, we use the chain rule: dL_dtheta = dL_dK * d_K_dtheta Note, we use the chain rule: dL_dtheta = dL_dK * d_K_dtheta
""" """
dL_dthetaK = self.kern.dK_dtheta(dL_dK=self.dL_dK, X=self.X, slices1=self.Xslices, slices2=self.Xslices) dL_dthetaK = self.kern.dK_dtheta(dL_dK=self.dL_dK, X=self.X)
if isinstance(self.likelihood, Laplace): if isinstance(self.likelihood, Laplace):
dL_dthetaK_explicit = dL_dthetaK dL_dthetaK_explicit = dL_dthetaK
#Need to pass in a matrix of ones to get access to raw dK_dthetaK values without being chained #Need to pass in a matrix of ones to get access to raw dK_dthetaK values without being chained
fake_dL_dKs = np.ones(self.dL_dK.shape) fake_dL_dKs = np.ones(self.dL_dK.shape)
dK_dthetaK = self.kern.dK_dtheta(dL_dK=fake_dL_dKs, X=self.X, slices1=self.Xslices, slices2=self.Xslices) dK_dthetaK = self.kern.dK_dtheta(dL_dK=fake_dL_dKs, X=self.X)
dL_dthetaK_implicit = self.likelihood._Kgradients(self.dL_dK, dK_dthetaK) dL_dthetaK_implicit = self.likelihood._Kgradients(self.dL_dK, dK_dthetaK)
dL_dthetaK = dL_dthetaK_explicit + dL_dthetaK_implicit dL_dthetaK = dL_dthetaK_explicit + dL_dthetaK_implicit
@ -251,7 +246,7 @@ class GP(model):
else: else:
raise NotImplementedError, "Cannot define a frame with more than two input dimensions" raise NotImplementedError, "Cannot define a frame with more than two input dimensions"
def plot(self, samples=0, plot_limits=None, which_data='all', which_functions='all', resolution=None, levels=20): def plot(self, samples=0, plot_limits=None, which_data='all', which_functions='all', which_parts='all', resolution=None, levels=20):
""" """
TODO: Docstrings! TODO: Docstrings!
:param levels: for 2D plotting, the number of contour levels to use :param levels: for 2D plotting, the number of contour levels to use