apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-06-11 15:15:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

unify the interface of psicomp, but the psi2n of linear kernel and gaussian qradrature still needs to be done.

Browse source
This commit is contained in:
Zhenwen Dai 2015-09-04 18:26:07 +01:00
parent 7335d4e62e
commit 23af82f851
4 changed files with 49 additions and 36 deletions
Download patch file Download diff file Expand all files Collapse all files

14
GPy/kern/_src/linear.py
View file

@ -117,29 +117,29 @@ class Linear(Kern):
#---------------------------------------#
def psi0(self, Z, variational_posterior):
return self.psicomp.psicomputations(self.variances, Z, variational_posterior)[0]
return self.psicomp.psicomputations(self, Z, variational_posterior)[0]
def psi1(self, Z, variational_posterior):
return self.psicomp.psicomputations(self.variances, Z, variational_posterior)[1]
return self.psicomp.psicomputations(self, Z, variational_posterior)[1]
def psi2(self, Z, variational_posterior):
return self.psicomp.psicomputations(self.variances, Z, variational_posterior)[2]
return self.psicomp.psicomputations(self, Z, variational_posterior)[2]
def psi2n(self, Z, variational_posterior):
return self.psicomp.psicomputations(self.variances, Z, variational_posterior, return_psi2_n=True)[2]
return self.psicomp.psicomputations(self, Z, variational_posterior, return_psi2_n=True)[2]
def update_gradients_expectations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
dL_dvar = self.psicomp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, self.variances, Z, variational_posterior)[0]
dL_dvar = self.psicomp.psiDerivativecomputations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior)[0]
if self.ARD:
self.variances.gradient = dL_dvar
else:
self.variances.gradient = dL_dvar.sum()
def gradients_Z_expectations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
return self.psicomp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, self.variances, Z, variational_posterior)[1]
return self.psicomp.psiDerivativecomputations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior)[1]
def gradients_qX_expectations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
return self.psicomp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, self.variances, Z, variational_posterior)[2:]
return self.psicomp.psiDerivativecomputations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior)[2:]
class LinearFull(Kern):
def __init__(self, input_dim, rank, W=None, kappa=None, active_dims=None, name='linear_full'):

51
GPy/kern/_src/psi_comp/__init__.py
View file

@ -8,11 +8,25 @@ from . import rbf_psi_comp
from . import ssrbf_psi_comp
from . import sslinear_psi_comp
from . import linear_psi_comp
class PSICOMP(Pickleable):
def psicomputations(self, kern, Z, qX, return_psi2_n=False):
raise NotImplementedError("Abstract method!")
def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, qX):
raise NotImplementedError("Abstract method!")
def _setup_observers(self):
pass
from .gaussherm import PSICOMP_GH
class PSICOMP_RBF(Pickleable):
@Cache_this(limit=2, ignore_args=(0,))
def psicomputations(self, variance, lengthscale, Z, variational_posterior, return_psi2_n=False):
class PSICOMP_RBF(PSICOMP):
@Cache_this(limit=5, ignore_args=(0,))
def psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False):
variance, lengthscale = kern.variance, kern.lengthscale
if isinstance(variational_posterior, variational.NormalPosterior):
return rbf_psi_comp.psicomputations(variance, lengthscale, Z, variational_posterior, return_psi2_n=return_psi2_n)
elif isinstance(variational_posterior, variational.SpikeAndSlabPosterior):
@ -20,8 +34,9 @@ class PSICOMP_RBF(Pickleable):
else:
raise ValueError("unknown distriubtion received for psi-statistics")
@Cache_this(limit=2, ignore_args=(0,1,2,3))
def psiDerivativecomputations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, variance, lengthscale, Z, variational_posterior):
@Cache_this(limit=5, ignore_args=(0,2,3,4))
def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
variance, lengthscale = kern.variance, kern.lengthscale
if isinstance(variational_posterior, variational.NormalPosterior):
return rbf_psi_comp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, variance, lengthscale, Z, variational_posterior)
elif isinstance(variational_posterior, variational.SpikeAndSlabPosterior):
@ -29,28 +44,26 @@ class PSICOMP_RBF(Pickleable):
else:
raise ValueError("unknown distriubtion received for psi-statistics")
def _setup_observers(self):
pass
class PSICOMP_Linear(PSICOMP):
class PSICOMP_Linear(Pickleable):
@Cache_this(limit=2, ignore_args=(0,))
def psicomputations(self, variance, Z, variational_posterior, return_psi2_n=False):
@Cache_this(limit=5, ignore_args=(0,))
def psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False):
variances = kern.variances
if isinstance(variational_posterior, variational.NormalPosterior):
return linear_psi_comp.psicomputations(variance, Z, variational_posterior, return_psi2_n=return_psi2_n)
return linear_psi_comp.psicomputations(variances, Z, variational_posterior, return_psi2_n=return_psi2_n)
elif isinstance(variational_posterior, variational.SpikeAndSlabPosterior):
return sslinear_psi_comp.psicomputations(variance, Z, variational_posterior)
return sslinear_psi_comp.psicomputations(variances, Z, variational_posterior)
else:
raise ValueError("unknown distriubtion received for psi-statistics")
@Cache_this(limit=2, ignore_args=(0,1,2,3))
def psiDerivativecomputations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, variance, Z, variational_posterior):
@Cache_this(limit=2, ignore_args=(0,2,3,4))
def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
variances = kern.variances
if isinstance(variational_posterior, variational.NormalPosterior):
return linear_psi_comp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, variance, Z, variational_posterior)
return linear_psi_comp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, variances, Z, variational_posterior)
elif isinstance(variational_posterior, variational.SpikeAndSlabPosterior):
return sslinear_psi_comp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, variance, Z, variational_posterior)
return sslinear_psi_comp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, variances, Z, variational_posterior)
else:
raise ValueError("unknown distriubtion received for psi-statistics")
def _setup_observers(self):
pass

6
GPy/kern/_src/psi_comp/gaussherm.py
View file

@ -9,9 +9,9 @@ import numpy as np
from GPy.util.caching import Cache_this
from ....util.linalg import tdot
from ....core.parameterization.parameter_core import Pickleable
from . import PSICOMP
class PSICOMP_GH(Pickleable):
class PSICOMP_GH(PSICOMP):
def __init__(self, degree=5, cache_K=True):
self.degree = degree
@ -36,7 +36,7 @@ class PSICOMP_GH(Pickleable):
return self.Xs
@Cache_this(limit=10, ignore_args=(0,))
def psicomputations(self, kern, Z, qX):
def psicomputations(self, kern, Z, qX, return_psi2_n=False):
mu, S = qX.mean.values, qX.variance.values
N,M,Q = mu.shape[0],Z.shape[0],mu.shape[1]
if self.cache_K: Xs = self.comp_K(Z, qX)

14
GPy/kern/_src/rbf.py
View file

@ -53,25 +53,25 @@ class RBF(Stationary):
#---------------------------------------#
def psi0(self, Z, variational_posterior):
return self.psicomp.psicomputations(self.variance, self.lengthscale, Z, variational_posterior)[0]
return self.psicomp.psicomputations(self, Z, variational_posterior)[0]
def psi1(self, Z, variational_posterior):
return self.psicomp.psicomputations(self.variance, self.lengthscale, Z, variational_posterior)[1]
return self.psicomp.psicomputations(self, Z, variational_posterior)[1]
def psi2(self, Z, variational_posterior):
return self.psicomp.psicomputations(self.variance, self.lengthscale, Z, variational_posterior, return_psi2_n=False)[2]
return self.psicomp.psicomputations(self, Z, variational_posterior, return_psi2_n=False)[2]
def psi2n(self, Z, variational_posterior):
return self.psicomp.psicomputations(self.variance, self.lengthscale, Z, variational_posterior, return_psi2_n=True)[2]
return self.psicomp.psicomputations(self, Z, variational_posterior, return_psi2_n=True)[2]
def update_gradients_expectations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
dL_dvar, dL_dlengscale = self.psicomp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, self.variance, self.lengthscale, Z, variational_posterior)[:2]
dL_dvar, dL_dlengscale = self.psicomp.psiDerivativecomputations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior)[:2]
self.variance.gradient = dL_dvar
self.lengthscale.gradient = dL_dlengscale
def gradients_Z_expectations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
return self.psicomp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, self.variance, self.lengthscale, Z, variational_posterior)[2]
return self.psicomp.psiDerivativecomputations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior)[2]
def gradients_qX_expectations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
return self.psicomp.psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, self.variance, self.lengthscale, Z, variational_posterior)[3:]
return self.psicomp.psiDerivativecomputations(self, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior)[3:]