apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-08 03:22:38 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge pull request #321 from SheffieldML/limit=3

Browse source 
[chaching] changing all chacher limits to 3
This commit is contained in:
Max Zwiessele 2016-03-07 13:32:54 +00:00
commit 462e23ab2b
48 changed files with 72 additions and 72 deletions
Download patch file Download diff file Expand all files Collapse all files

2
GPy/core/sparse_gp.py
View file

@ -44,7 +44,7 @@ class SparseGP(GP):
#pick a sensible inference method #pick a sensible inference method
if inference_method is None: if inference_method is None:
if isinstance(likelihood, likelihoods.Gaussian): if isinstance(likelihood, likelihoods.Gaussian):
inference_method = var_dtc.VarDTC(limit=1) inference_method = var_dtc.VarDTC(limit=3)
else: else:
#inference_method = ?? #inference_method = ??
raise NotImplementedError("what to do what to do?") raise NotImplementedError("what to do what to do?")

2
GPy/inference/latent_function_inference/var_dtc.py
View file

@ -22,7 +22,7 @@ class VarDTC(LatentFunctionInference):
""" """
const_jitter = 1e-8 const_jitter = 1e-8
def __init__(self, limit=1): def __init__(self, limit=3):
from paramz.caching import Cacher from paramz.caching import Cacher
self.limit = limit self.limit = limit
self.get_trYYT = Cacher(self._get_trYYT, limit) self.get_trYYT = Cacher(self._get_trYYT, limit)

2
GPy/inference/latent_function_inference/var_dtc_parallel.py
View file

@ -21,7 +21,7 @@ class VarDTC_minibatch(LatentFunctionInference):
""" """
const_jitter = 1e-8 const_jitter = 1e-8
def __init__(self, batchsize=None, limit=1, mpi_comm=None): def __init__(self, batchsize=None, limit=3, mpi_comm=None):
self.batchsize = batchsize self.batchsize = batchsize
self.mpi_comm = mpi_comm self.mpi_comm = mpi_comm

12
GPy/kern/src/add.py
View file

@ -37,7 +37,7 @@ class Add(CombinationKernel):
else: else:
return False return False
@Cache_this(limit=2, force_kwargs=['which_parts']) @Cache_this(limit=3, force_kwargs=['which_parts'])
def K(self, X, X2=None, which_parts=None): def K(self, X, X2=None, which_parts=None):
""" """
Add all kernels together. Add all kernels together.
@ -51,7 +51,7 @@ class Add(CombinationKernel):
which_parts = [which_parts] which_parts = [which_parts]
return reduce(np.add, (p.K(X, X2) for p in which_parts)) return reduce(np.add, (p.K(X, X2) for p in which_parts))
@Cache_this(limit=2, force_kwargs=['which_parts']) @Cache_this(limit=3, force_kwargs=['which_parts'])
def Kdiag(self, X, which_parts=None): def Kdiag(self, X, which_parts=None):
if which_parts is None: if which_parts is None:
which_parts = self.parts which_parts = self.parts
@ -98,17 +98,17 @@ class Add(CombinationKernel):
[target.__iadd__(p.gradients_XX_diag(dL_dKdiag, X)) for p in self.parts] [target.__iadd__(p.gradients_XX_diag(dL_dKdiag, X)) for p in self.parts]
return target return target
@Cache_this(limit=1, force_kwargs=['which_parts']) @Cache_this(limit=3, force_kwargs=['which_parts'])
def psi0(self, Z, variational_posterior): def psi0(self, Z, variational_posterior):
if not self._exact_psicomp: return Kern.psi0(self,Z,variational_posterior) if not self._exact_psicomp: return Kern.psi0(self,Z,variational_posterior)
return reduce(np.add, (p.psi0(Z, variational_posterior) for p in self.parts)) return reduce(np.add, (p.psi0(Z, variational_posterior) for p in self.parts))
@Cache_this(limit=1, force_kwargs=['which_parts']) @Cache_this(limit=3, force_kwargs=['which_parts'])
def psi1(self, Z, variational_posterior): def psi1(self, Z, variational_posterior):
if not self._exact_psicomp: return Kern.psi1(self,Z,variational_posterior) if not self._exact_psicomp: return Kern.psi1(self,Z,variational_posterior)
return reduce(np.add, (p.psi1(Z, variational_posterior) for p in self.parts)) return reduce(np.add, (p.psi1(Z, variational_posterior) for p in self.parts))
@Cache_this(limit=1, force_kwargs=['which_parts']) @Cache_this(limit=3, force_kwargs=['which_parts'])
def psi2(self, Z, variational_posterior): def psi2(self, Z, variational_posterior):
if not self._exact_psicomp: return Kern.psi2(self,Z,variational_posterior) if not self._exact_psicomp: return Kern.psi2(self,Z,variational_posterior)
psi2 = reduce(np.add, (p.psi2(Z, variational_posterior) for p in self.parts)) psi2 = reduce(np.add, (p.psi2(Z, variational_posterior) for p in self.parts))
@ -144,7 +144,7 @@ class Add(CombinationKernel):
raise NotImplementedError("psi2 cannot be computed for this kernel") raise NotImplementedError("psi2 cannot be computed for this kernel")
return psi2 return psi2
@Cache_this(limit=1, force_kwargs=['which_parts']) @Cache_this(limit=3, force_kwargs=['which_parts'])
def psi2n(self, Z, variational_posterior): def psi2n(self, Z, variational_posterior):
if not self._exact_psicomp: return Kern.psi2n(self, Z, variational_posterior) if not self._exact_psicomp: return Kern.psi2n(self, Z, variational_posterior)
psi2 = reduce(np.add, (p.psi2n(Z, variational_posterior) for p in self.parts)) psi2 = reduce(np.add, (p.psi2n(Z, variational_posterior) for p in self.parts))

2
GPy/kern/src/eq_ode2.py
View file

@ -64,7 +64,7 @@ class EQ_ODE2(Kern):
self.W = Param('W', W) self.W = Param('W', W)
self.link_parameters(self.lengthscale, self.C, self.B, self.W) self.link_parameters(self.lengthscale, self.C, self.B, self.W)
@Cache_this(limit=2) @Cache_this(limit=3)
def K(self, X, X2=None): def K(self, X, X2=None):
#This way is not working, indexes are lost after using k._slice_X #This way is not working, indexes are lost after using k._slice_X
#index = np.asarray(X, dtype=np.int) #index = np.asarray(X, dtype=np.int)

2
GPy/kern/src/kern.py
View file

@ -68,7 +68,7 @@ class Kern(Parameterized):
def _effective_input_dim(self): def _effective_input_dim(self):
return np.size(self._all_dims_active) return np.size(self._all_dims_active)
@Cache_this(limit=20) @Cache_this(limit=3)
def _slice_X(self, X): def _slice_X(self, X):
try: try:
return X[:, self._all_dims_active].astype('float') return X[:, self._all_dims_active].astype('float')

4
GPy/kern/src/linear.py
View file

@ -51,7 +51,7 @@ class Linear(Kern):
self.link_parameter(self.variances) self.link_parameter(self.variances)
self.psicomp = PSICOMP_Linear() self.psicomp = PSICOMP_Linear()
@Cache_this(limit=2) @Cache_this(limit=3)
def K(self, X, X2=None): def K(self, X, X2=None):
if self.ARD: if self.ARD:
if X2 is None: if X2 is None:
@ -62,7 +62,7 @@ class Linear(Kern):
else: else:
return self._dot_product(X, X2) * self.variances return self._dot_product(X, X2) * self.variances
@Cache_this(limit=1, ignore_args=(0,)) @Cache_this(limit=3, ignore_args=(0,))
def _dot_product(self, X, X2=None): def _dot_product(self, X, X2=None):
if X2 is None: if X2 is None:
return tdot(X) return tdot(X)

10
GPy/kern/src/mlp.py
View file

@ -45,7 +45,7 @@ class MLP(Kern):
self.link_parameters(self.variance, self.weight_variance, self.bias_variance) self.link_parameters(self.variance, self.weight_variance, self.bias_variance)
@Cache_this(limit=20, ignore_args=()) @Cache_this(limit=3, ignore_args=())
def K(self, X, X2=None): def K(self, X, X2=None):
if X2 is None: if X2 is None:
X_denom = np.sqrt(self._comp_prod(X)+1.) X_denom = np.sqrt(self._comp_prod(X)+1.)
@ -57,7 +57,7 @@ class MLP(Kern):
XTX = self._comp_prod(X,X2)/X_denom[:,None]/X2_denom[None,:] XTX = self._comp_prod(X,X2)/X_denom[:,None]/X2_denom[None,:]
return self.variance*four_over_tau*np.arcsin(XTX) return self.variance*four_over_tau*np.arcsin(XTX)
@Cache_this(limit=20, ignore_args=()) @Cache_this(limit=3, ignore_args=())
def Kdiag(self, X): def Kdiag(self, X):
"""Compute the diagonal of the covariance matrix for X.""" """Compute the diagonal of the covariance matrix for X."""
X_prod = self._comp_prod(X) X_prod = self._comp_prod(X)
@ -88,14 +88,14 @@ class MLP(Kern):
"""Gradient of diagonal of covariance with respect to X""" """Gradient of diagonal of covariance with respect to X"""
return self._comp_grads_diag(dL_dKdiag, X)[3] return self._comp_grads_diag(dL_dKdiag, X)[3]
@Cache_this(limit=50, ignore_args=()) @Cache_this(limit=3, ignore_args=())
def _comp_prod(self, X, X2=None): def _comp_prod(self, X, X2=None):
if X2 is None: if X2 is None:
return (np.square(X)*self.weight_variance).sum(axis=1)+self.bias_variance return (np.square(X)*self.weight_variance).sum(axis=1)+self.bias_variance
else: else:
return (X*self.weight_variance).dot(X2.T)+self.bias_variance return (X*self.weight_variance).dot(X2.T)+self.bias_variance
@Cache_this(limit=20, ignore_args=(1,)) @Cache_this(limit=3, ignore_args=(1,))
def _comp_grads(self, dL_dK, X, X2=None): def _comp_grads(self, dL_dK, X, X2=None):
var,w,b = self.variance, self.weight_variance, self.bias_variance var,w,b = self.variance, self.weight_variance, self.bias_variance
K = self.K(X, X2) K = self.K(X, X2)
@ -130,7 +130,7 @@ class MLP(Kern):
dX2 = common.T.dot(X)*w-((common*XTX).sum(axis=0)/(X2_prod+1.))[:,None]*X2*w dX2 = common.T.dot(X)*w-((common*XTX).sum(axis=0)/(X2_prod+1.))[:,None]*X2*w
return dvar, dw, db, dX, dX2 return dvar, dw, db, dX, dX2
@Cache_this(limit=20, ignore_args=(1,)) @Cache_this(limit=3, ignore_args=(1,))
def _comp_grads_diag(self, dL_dKdiag, X): def _comp_grads_diag(self, dL_dKdiag, X):
var,w,b = self.variance, self.weight_variance, self.bias_variance var,w,b = self.variance, self.weight_variance, self.bias_variance
K = self.Kdiag(X) K = self.Kdiag(X)

2
GPy/kern/src/poly.py
View file

@ -27,7 +27,7 @@ class Poly(Kern):
_, _, B = self._AB(X, X2) _, _, B = self._AB(X, X2)
return B * self.variance return B * self.variance
@Cache_this(limit=2) @Cache_this(limit=3)
def _AB(self, X, X2=None): def _AB(self, X, X2=None):
if X2 is None: if X2 is None:
dot_prod = np.dot(X, X.T) dot_prod = np.dot(X, X.T)

4
GPy/kern/src/prod.py
View file

@ -39,7 +39,7 @@ class Prod(CombinationKernel):
kernels.insert(i, part) kernels.insert(i, part)
super(Prod, self).__init__(kernels, name) super(Prod, self).__init__(kernels, name)
@Cache_this(limit=2, force_kwargs=['which_parts']) @Cache_this(limit=3, force_kwargs=['which_parts'])
def K(self, X, X2=None, which_parts=None): def K(self, X, X2=None, which_parts=None):
if which_parts is None: if which_parts is None:
which_parts = self.parts which_parts = self.parts
@ -48,7 +48,7 @@ class Prod(CombinationKernel):
which_parts = [which_parts] which_parts = [which_parts]
return reduce(np.multiply, (p.K(X, X2) for p in which_parts)) return reduce(np.multiply, (p.K(X, X2) for p in which_parts))
@Cache_this(limit=2, force_kwargs=['which_parts']) @Cache_this(limit=3, force_kwargs=['which_parts'])
def Kdiag(self, X, which_parts=None): def Kdiag(self, X, which_parts=None):
if which_parts is None: if which_parts is None:
which_parts = self.parts which_parts = self.parts

8
GPy/kern/src/psi_comp/__init__.py
View file

@ -21,7 +21,7 @@ from .gaussherm import PSICOMP_GH
from . import rbf_psi_comp, linear_psi_comp, ssrbf_psi_comp, sslinear_psi_comp from . import rbf_psi_comp, linear_psi_comp, ssrbf_psi_comp, sslinear_psi_comp
class PSICOMP_RBF(PSICOMP): class PSICOMP_RBF(PSICOMP):
@Cache_this(limit=10, ignore_args=(0,)) @Cache_this(limit=3, ignore_args=(0,))
def psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False): def psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False):
variance, lengthscale = kern.variance, kern.lengthscale variance, lengthscale = kern.variance, kern.lengthscale
if isinstance(variational_posterior, variational.NormalPosterior): if isinstance(variational_posterior, variational.NormalPosterior):
@ -31,7 +31,7 @@ class PSICOMP_RBF(PSICOMP):
else: else:
raise ValueError("unknown distriubtion received for psi-statistics") raise ValueError("unknown distriubtion received for psi-statistics")
@Cache_this(limit=10, ignore_args=(0,2,3,4)) @Cache_this(limit=3, ignore_args=(0,2,3,4))
def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior): def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
variance, lengthscale = kern.variance, kern.lengthscale variance, lengthscale = kern.variance, kern.lengthscale
if isinstance(variational_posterior, variational.NormalPosterior): if isinstance(variational_posterior, variational.NormalPosterior):
@ -43,7 +43,7 @@ class PSICOMP_RBF(PSICOMP):
class PSICOMP_Linear(PSICOMP): class PSICOMP_Linear(PSICOMP):
@Cache_this(limit=10, ignore_args=(0,)) @Cache_this(limit=3, ignore_args=(0,))
def psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False): def psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False):
variances = kern.variances variances = kern.variances
if isinstance(variational_posterior, variational.NormalPosterior): if isinstance(variational_posterior, variational.NormalPosterior):
@ -53,7 +53,7 @@ class PSICOMP_Linear(PSICOMP):
else: else:
raise ValueError("unknown distriubtion received for psi-statistics") raise ValueError("unknown distriubtion received for psi-statistics")
@Cache_this(limit=10, ignore_args=(0,2,3,4)) @Cache_this(limit=3, ignore_args=(0,2,3,4))
def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior): def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
variances = kern.variances variances = kern.variances
if isinstance(variational_posterior, variational.NormalPosterior): if isinstance(variational_posterior, variational.NormalPosterior):

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

@ -27,7 +27,7 @@ class PSICOMP_GH(PSICOMP):
def _setup_observers(self): def _setup_observers(self):
pass pass
@Cache_this(limit=10, ignore_args=(0,)) @Cache_this(limit=3, ignore_args=(0,))
def comp_K(self, Z, qX): def comp_K(self, Z, qX):
if self.Xs is None or self.Xs.shape != qX.mean.shape: if self.Xs is None or self.Xs.shape != qX.mean.shape:
from paramz import ObsAr from paramz import ObsAr
@ -38,7 +38,7 @@ class PSICOMP_GH(PSICOMP):
self.Xs[i] = self.locs[i]*S_sq+mu self.Xs[i] = self.locs[i]*S_sq+mu
return self.Xs return self.Xs
@Cache_this(limit=10, ignore_args=(0,)) @Cache_this(limit=3, ignore_args=(0,))
def psicomputations(self, kern, Z, qX, return_psi2_n=False): def psicomputations(self, kern, Z, qX, return_psi2_n=False):
mu, S = qX.mean.values, qX.variance.values mu, S = qX.mean.values, qX.variance.values
N,M,Q = mu.shape[0],Z.shape[0],mu.shape[1] N,M,Q = mu.shape[0],Z.shape[0],mu.shape[1]
@ -62,7 +62,7 @@ class PSICOMP_GH(PSICOMP):
psi2 += self.weights[i]* tdot(Kfu.T) psi2 += self.weights[i]* tdot(Kfu.T)
return psi0, psi1, psi2 return psi0, psi1, psi2
@Cache_this(limit=10, ignore_args=(0, 2,3,4)) @Cache_this(limit=3, ignore_args=(0, 2,3,4))
def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, qX): def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, qX):
mu, S = qX.mean.values, qX.variance.values mu, S = qX.mean.values, qX.variance.values
if self.cache_K: Xs = self.comp_K(Z, qX) if self.cache_K: Xs = self.comp_K(Z, qX)

4
GPy/kern/src/psi_comp/rbf_psi_comp.py
View file

@ -132,5 +132,5 @@ def _psi2compDer(dL_dpsi2, variance, lengthscale, Z, mu, S):
return _dL_dvar, _dL_dl, _dL_dZ, _dL_dmu, _dL_dS return _dL_dvar, _dL_dl, _dL_dZ, _dL_dmu, _dL_dS
_psi1computations = Cacher(__psi1computations, limit=5) _psi1computations = Cacher(__psi1computations, limit=3)
_psi2computations = Cacher(__psi2computations, limit=5) _psi2computations = Cacher(__psi2computations, limit=3)

4
GPy/kern/src/psi_comp/rbf_psi_gpucomp.py
View file

@ -326,7 +326,7 @@ class PSICOMP_RBF_GPU(PSICOMP_RBF):
except: except:
return self.fall_back.psicomputations(kern, Z, variational_posterior, return_psi2_n) return self.fall_back.psicomputations(kern, Z, variational_posterior, return_psi2_n)
@Cache_this(limit=10, ignore_args=(0,)) @Cache_this(limit=3, ignore_args=(0,))
def _psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False): def _psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False):
""" """
Z - MxQ Z - MxQ
@ -371,7 +371,7 @@ class PSICOMP_RBF_GPU(PSICOMP_RBF):
except: except:
return self.fall_back.psiDerivativecomputations(kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior) return self.fall_back.psiDerivativecomputations(kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior)
@Cache_this(limit=10, ignore_args=(0,2,3,4)) @Cache_this(limit=3, ignore_args=(0,2,3,4))
def _psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior): def _psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
# resolve the requirement of dL_dpsi2 to be symmetric # resolve the requirement of dL_dpsi2 to be symmetric
if len(dL_dpsi2.shape)==2: dL_dpsi2 = (dL_dpsi2+dL_dpsi2.T)/2 if len(dL_dpsi2.shape)==2: dL_dpsi2 = (dL_dpsi2+dL_dpsi2.T)/2

2
GPy/kern/src/psi_comp/ssrbf_psi_comp.py
View file

@ -88,7 +88,7 @@ try:
return psi0,psi1,psi2,psi2n return psi0,psi1,psi2,psi2n
from GPy.util.caching import Cacher from GPy.util.caching import Cacher
psicomputations = Cacher(_psicomputations, limit=1) psicomputations = Cacher(_psicomputations, limit=3)
def psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, variance, lengthscale, Z, variational_posterior): def psiDerivativecomputations(dL_dpsi0, dL_dpsi1, dL_dpsi2, variance, lengthscale, Z, variational_posterior):
ARD = (len(lengthscale)!=1) ARD = (len(lengthscale)!=1)

4
GPy/kern/src/psi_comp/ssrbf_psi_gpucomp.py
View file

@ -375,7 +375,7 @@ class PSICOMP_SSRBF_GPU(PSICOMP_RBF):
def get_dimensions(self, Z, variational_posterior): def get_dimensions(self, Z, variational_posterior):
return variational_posterior.mean.shape[0], Z.shape[0], Z.shape[1] return variational_posterior.mean.shape[0], Z.shape[0], Z.shape[1]
@Cache_this(limit=1, ignore_args=(0,)) @Cache_this(limit=3, ignore_args=(0,))
def psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False): def psicomputations(self, kern, Z, variational_posterior, return_psi2_n=False):
""" """
Z - MxQ Z - MxQ
@ -409,7 +409,7 @@ class PSICOMP_SSRBF_GPU(PSICOMP_RBF):
else: else:
return psi0, psi1_gpu.get(), psi2_gpu.get() return psi0, psi1_gpu.get(), psi2_gpu.get()
@Cache_this(limit=1, ignore_args=(0,2,3,4)) @Cache_this(limit=3, ignore_args=(0,2,3,4))
def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior): def psiDerivativecomputations(self, kern, dL_dpsi0, dL_dpsi1, dL_dpsi2, Z, variational_posterior):
variance, lengthscale = kern.variance, kern.lengthscale variance, lengthscale = kern.variance, kern.lengthscale
from ....util.linalg_gpu import sum_axis from ....util.linalg_gpu import sum_axis

4
GPy/kern/src/stationary.py
View file

@ -81,11 +81,11 @@ class Stationary(Kern):
def dK_dr(self, r): def dK_dr(self, r):
raise NotImplementedError("implement derivative of the covariance function wrt r to use this class") raise NotImplementedError("implement derivative of the covariance function wrt r to use this class")
@Cache_this(limit=20, ignore_args=()) @Cache_this(limit=3, ignore_args=())
def dK2_drdr(self, r): def dK2_drdr(self, r):
raise NotImplementedError("implement second derivative of covariance wrt r to use this method") raise NotImplementedError("implement second derivative of covariance wrt r to use this method")
@Cache_this(limit=5, ignore_args=()) @Cache_this(limit=3, ignore_args=())
def K(self, X, X2=None): def K(self, X, X2=None):
""" """
Kernel function applied on inputs X and X2. Kernel function applied on inputs X and X2.

8
GPy/kern/src/trunclinear.py
View file

@ -54,12 +54,12 @@ class TruncLinear(Kern):
self.add_parameter(self.variances) self.add_parameter(self.variances)
self.add_parameter(self.delta) self.add_parameter(self.delta)
@Cache_this(limit=2) @Cache_this(limit=3)
def K(self, X, X2=None): def K(self, X, X2=None):
XX = self.variances*self._product(X, X2) XX = self.variances*self._product(X, X2)
return XX.sum(axis=-1) return XX.sum(axis=-1)
@Cache_this(limit=2) @Cache_this(limit=3)
def _product(self, X, X2=None): def _product(self, X, X2=None):
if X2 is None: if X2 is None:
X2 = X X2 = X
@ -149,12 +149,12 @@ class TruncLinear_inf(Kern):
self.add_parameter(self.variances) self.add_parameter(self.variances)
# @Cache_this(limit=2) # @Cache_this(limit=3)
def K(self, X, X2=None): def K(self, X, X2=None):
tmp = self._product(X, X2) tmp = self._product(X, X2)
return (self.variances*tmp).sum(axis=-1) return (self.variances*tmp).sum(axis=-1)
# @Cache_this(limit=2) # @Cache_this(limit=3)
def _product(self, X, X2=None): def _product(self, X, X2=None):
if X2 is None: if X2 is None:
X2 = X X2 = X

2
GPy/models/bayesian_gplvm.py
View file

@ -61,7 +61,7 @@ class BayesianGPLVM(SparseGP_MPI):
else: else:
from ..inference.latent_function_inference.var_dtc import VarDTC from ..inference.latent_function_inference.var_dtc import VarDTC
self.logger.debug("creating inference_method var_dtc") self.logger.debug("creating inference_method var_dtc")
inference_method = VarDTC(limit=1 if not missing_data else Y.shape[1]) inference_method = VarDTC(limit=3 if not missing_data else Y.shape[1])
if isinstance(inference_method,VarDTC_minibatch): if isinstance(inference_method,VarDTC_minibatch):
inference_method.mpi_comm = mpi_comm inference_method.mpi_comm = mpi_comm

2
GPy/models/bayesian_gplvm_minibatch.py
View file

@ -61,7 +61,7 @@ class BayesianGPLVMMiniBatch(SparseGPMiniBatch):
if inference_method is None: if inference_method is None:
from ..inference.latent_function_inference.var_dtc import VarDTC from ..inference.latent_function_inference.var_dtc import VarDTC
self.logger.debug("creating inference_method var_dtc") self.logger.debug("creating inference_method var_dtc")
inference_method = VarDTC(limit=1 if not missing_data else Y.shape[1]) inference_method = VarDTC(limit=3 if not missing_data else Y.shape[1])
super(BayesianGPLVMMiniBatch,self).__init__(X, Y, Z, kernel, likelihood=likelihood, super(BayesianGPLVMMiniBatch,self).__init__(X, Y, Z, kernel, likelihood=likelihood,
name=name, inference_method=inference_method, name=name, inference_method=inference_method,

2
GPy/models/sparse_gp_minibatch.py
View file

@ -45,7 +45,7 @@ class SparseGPMiniBatch(SparseGP):
# pick a sensible inference method # pick a sensible inference method
if inference_method is None: if inference_method is None:
if isinstance(likelihood, likelihoods.Gaussian): if isinstance(likelihood, likelihoods.Gaussian):
inference_method = var_dtc.VarDTC(limit=1 if not missing_data else Y.shape[1]) inference_method = var_dtc.VarDTC(limit=3 if not missing_data else Y.shape[1])
else: else:
#inference_method = ?? #inference_method = ??
raise NotImplementedError("what to do what to do?") raise NotImplementedError("what to do what to do?")

2
GPy/util/linalg.py
View file

@ -78,7 +78,7 @@ def jitchol(A, maxtries=5):
try: raise try: raise
except: except:
logging.warning('\n'.join(['Added jitter of {:.10e}'.format(jitter), logging.warning('\n'.join(['Added jitter of {:.10e}'.format(jitter),
' in '+traceback.format_list(traceback.extract_stack(limit=2)[-2:-1])[0][2:]])) ' in '+traceback.format_list(traceback.extract_stack(limit=3)[-2:-1])[0][2:]]))
return L return L
# def dtrtri(L, lower=1): # def dtrtri(L, lower=1):