apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-15 06:52:39 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'debug'

Browse source
This commit is contained in:
James Hensman 2013-03-12 17:05:50 +00:00
commit 9d27c920e0
1 changed files with 18 additions and 18 deletions
Download patch file Download diff file Expand all files Collapse all files

36
GPy/models/sparse_GP.py
View file

@ -153,14 +153,26 @@ class sparse_GP(GP):
#self.partial_for_likelihood += -np.diag(np.dot((self.C - 0.5 * mdot(self.C,self.psi2_beta_scaled,self.C) ) , self.psi1VVpsi1 ))*self.likelihood.precision #dD #self.partial_for_likelihood += -np.diag(np.dot((self.C - 0.5 * mdot(self.C,self.psi2_beta_scaled,self.C) ) , self.psi1VVpsi1 ))*self.likelihood.precision #dD
else: else:
#likelihood is not heterscedatic #likelihood is not heterscedatic
beta = self.likelihood.precision self.partial_for_likelihood = - 0.5 * self.N*self.D*self.likelihood.precision + 0.5 * np.sum(np.square(self.likelihood.Y))*self.likelihood.precision**2
dbeta = 0.5 * self.N*self.D/beta - 0.5 * np.sum(np.square(self.likelihood.Y)) self.partial_for_likelihood += 0.5 * self.D * (self.psi0.sum()*self.likelihood.precision**2 - np.trace(self.A)*self.likelihood.precision*sf2)
dbeta += - 0.5 * self.D * (self.psi0.sum() - np.trace(self.A)/beta*sf2) self.partial_for_likelihood += 0.5 * self.D * trace_dot(self.Bi,self.A)*self.likelihood.precision
dbeta += - 0.5 * self.D * trace_dot(self.Bi,self.A)/beta self.partial_for_likelihood += self.likelihood.precision*(0.5*trace_dot(self.psi2_beta_scaled,self.E*sf2) - np.trace(self.Cpsi1VVpsi1))
dbeta += np.trace(self.Cpsi1VVpsi1)/beta - 0.5 * trace_dot(np.dot(self.C,self.psi2_beta_scaled) , self.Cpsi1VVpsi1 )/beta
self.partial_for_likelihood = -dbeta*self.likelihood.precision**2
def log_likelihood(self):
""" Compute the (lower bound on the) log marginal likelihood """
sf2 = self.scale_factor**2
if self.likelihood.is_heteroscedastic:
A = -0.5*self.N*self.D*np.log(2.*np.pi) +0.5*np.sum(np.log(self.likelihood.precision)) -0.5*np.sum(self.V*self.likelihood.Y)
B = -0.5*self.D*(np.sum(self.likelihood.precision.flatten()*self.psi0) - np.trace(self.A)*sf2)
else:
A = -0.5*self.N*self.D*(np.log(2.*np.pi) + np.log(self.likelihood._variance)) -0.5*self.likelihood.precision*self.likelihood.trYYT
B = -0.5*self.D*(np.sum(self.likelihood.precision*self.psi0) - np.trace(self.A)*sf2)
C = -0.5*self.D * (self.B_logdet + self.M*np.log(sf2))
D = 0.5*np.trace(self.Cpsi1VVpsi1)
return A+B+C+D
def _set_params(self, p): def _set_params(self, p):
self.Z = p[:self.M*self.Q].reshape(self.M, self.Q) self.Z = p[:self.M*self.Q].reshape(self.M, self.Q)
self.kern._set_params(p[self.Z.size:self.Z.size+self.kern.Nparam]) self.kern._set_params(p[self.Z.size:self.Z.size+self.kern.Nparam])
@ -188,18 +200,6 @@ class sparse_GP(GP):
#self.likelihood.fit_FITC(self.Kmm,self.psi1,self.psi0) #self.likelihood.fit_FITC(self.Kmm,self.psi1,self.psi0)
self._set_params(self._get_params()) # update the GP self._set_params(self._get_params()) # update the GP
def log_likelihood(self):
""" Compute the (lower bound on the) log marginal likelihood """
sf2 = self.scale_factor**2
if self.likelihood.is_heteroscedastic:
A = -0.5*self.N*self.D*np.log(2.*np.pi) +0.5*np.sum(np.log(self.likelihood.precision)) -0.5*np.sum(self.V*self.likelihood.Y)
B = -0.5*self.D*(np.sum(self.likelihood.precision.flatten()*self.psi0) - np.trace(self.A)*sf2)
else:
A = -0.5*self.N*self.D*(np.log(2.*np.pi) - np.log(self.likelihood.precision)) -0.5*self.likelihood.precision*self.likelihood.trYYT
B = -0.5*self.D*(np.sum(self.likelihood.precision*self.psi0) - np.trace(self.A)*sf2)
C = -0.5*self.D * (self.B_logdet + self.M*np.log(sf2))
D = 0.5*np.trace(self.Cpsi1VVpsi1)
return A+B+C+D
def _log_likelihood_gradients(self): def _log_likelihood_gradients(self):
return np.hstack((self.dL_dZ().flatten(), self.dL_dtheta(), self.likelihood._gradients(partial=self.partial_for_likelihood))) return np.hstack((self.dL_dZ().flatten(), self.dL_dtheta(), self.likelihood._gradients(partial=self.partial_for_likelihood)))