apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-09 20:12:38 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

[mrd] fixes for updates on psi2

Browse source
This commit is contained in:
Max Zwiessele 2015-09-04 11:07:39 +01:00
parent 00f514b451
commit 79dd821424
2 changed files with 27 additions and 29 deletions
Download patch file Download diff file Expand all files Collapse all files

46
GPy/models/bayesian_gplvm_minibatch.py
View file

@ -100,22 +100,8 @@ class BayesianGPLVMMiniBatch(SparseGPMiniBatch):
dL_dpsi2=full_values['dL_dpsi2'], dL_dpsi2=full_values['dL_dpsi2'],
psi0=self.psi0, psi1=self.psi1, psi2=self.psi2) psi0=self.psi0, psi1=self.psi1, psi2=self.psi2)
kl_fctr = self.kl_factr self.X.mean.gradient = meangrad_tmp
self.X.variance.gradient = vargrad_tmp
self.X.mean.gradient[:] = 0
self.X.variance.gradient[:] = 0
self.variational_prior.update_gradients_KL(self.X)
if self.missing_data or not self.stochastics:
self.X.mean.gradient = kl_fctr*self.X.mean.gradient
self.X.variance.gradient = kl_fctr*self.X.variance.gradient
else:
d = self.output_dim
self.X.mean.gradient = kl_fctr*self.X.mean.gradient*self.stochastics.batchsize/d
self.X.variance.gradient = kl_fctr*self.X.variance.gradient*self.stochastics.batchsize/d
self.X.mean.gradient += meangrad_tmp
self.X.variance.gradient += vargrad_tmp
else: else:
self.X.gradient = self.kern.gradients_X(full_values['dL_dKnm'], self.X, self.Z) self.X.gradient = self.kern.gradients_X(full_values['dL_dKnm'], self.X, self.Z)
self.X.gradient += self.kern.gradients_X_diag(full_values['dL_dKdiag'], self.X) self.X.gradient += self.kern.gradients_X_diag(full_values['dL_dKdiag'], self.X)
@ -126,15 +112,27 @@ class BayesianGPLVMMiniBatch(SparseGPMiniBatch):
def parameters_changed(self): def parameters_changed(self):
super(BayesianGPLVMMiniBatch,self).parameters_changed() super(BayesianGPLVMMiniBatch,self).parameters_changed()
kl_fctr = self.kl_factr
if self.missing_data or not self.stochastics:
self._log_marginal_likelihood -= kl_fctr*self.variational_prior.KL_divergence(self.X)
elif self.stochastics:
d = self.output_dim
self._log_marginal_likelihood -= kl_fctr*self.variational_prior.KL_divergence(self.X)*self.stochastics.batchsize/d
if isinstance(self.inference_method, VarDTC_minibatch): kl_fctr = self.kl_factr
return if kl_fctr > 0:
Xgrad = self.X.gradient.copy()
self.X.gradient[:] = 0
self.variational_prior.update_gradients_KL(self.X)
if self.missing_data or not self.stochastics:
self.X.mean.gradient = kl_fctr*self.X.mean.gradient
self.X.variance.gradient = kl_fctr*self.X.variance.gradient
else:
d = self.output_dim
self.X.mean.gradient = kl_fctr*self.X.mean.gradient*self.stochastics.batchsize/d
self.X.variance.gradient = kl_fctr*self.X.variance.gradient*self.stochastics.batchsize/d
self.X.gradient += Xgrad
if self.missing_data or not self.stochastics:
self._log_marginal_likelihood -= kl_fctr*self.variational_prior.KL_divergence(self.X)
elif self.stochastics:
d = self.output_dim
self._log_marginal_likelihood -= kl_fctr*self.variational_prior.KL_divergence(self.X)*self.stochastics.batchsize/d
def plot_latent(self, labels=None, which_indices=None, def plot_latent(self, labels=None, which_indices=None,
resolution=50, ax=None, marker='o', s=40, resolution=50, ax=None, marker='o', s=40,

10
GPy/models/mrd.py
View file

@ -170,14 +170,14 @@ class MRD(BayesianGPLVMMiniBatch):
self._log_marginal_likelihood += b._log_marginal_likelihood self._log_marginal_likelihood += b._log_marginal_likelihood
self.logger.info('working on im <{}>'.format(hex(id(i)))) self.logger.info('working on im <{}>'.format(hex(id(i))))
self.Z.gradient[:] += b.full_values['Zgrad'] self.Z.gradient[:] += b.Z.gradient#full_values['Zgrad']
grad_dict = b.full_values #grad_dict = b.full_values
if self.has_uncertain_inputs(): if self.has_uncertain_inputs():
self.X.mean.gradient += grad_dict['meangrad'] self.X.mean.gradient += b.X.mean.gradient
self.X.variance.gradient += grad_dict['vargrad'] self.X.variance.gradient += b.X.variance.gradient
else: else:
self.X.gradient += grad_dict['Xgrad'] self.X.gradient += b.X.gradient
if self.has_uncertain_inputs(): if self.has_uncertain_inputs():
# update for the KL divergence # update for the KL divergence