apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-06-08 15:05:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

chancges to where gradients are computed in laplace

Browse source
This commit is contained in:
James Hensman 2014-03-13 12:13:00 +00:00
parent 4d00b9db03
commit ccf5167b75
2 changed files with 11 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

27
GPy/core/gp.py
View file

@ -27,7 +27,7 @@ class GP(Model):
""" """
def __init__(self, X, Y, kernel, likelihood, inference_method=None, name='gp', **Y_metadata): def __init__(self, X, Y, kernel, likelihood, inference_method=None, name='gp', Y_metadata=None):
super(GP, self).__init__(name) super(GP, self).__init__(name)
assert X.ndim == 2 assert X.ndim == 2
@ -42,10 +42,7 @@ class GP(Model):
assert Y.shape[0] == self.num_data assert Y.shape[0] == self.num_data
_, self.output_dim = self.Y.shape _, self.output_dim = self.Y.shape
if Y_metadata is not None: self.Y_metadata = Y_metadata or {}
self.Y_metadata = Y_metadata
else:
self.Y_metadata = None
assert isinstance(kernel, kern.Kern) assert isinstance(kernel, kern.Kern)
#assert self.input_dim == kernel.input_dim #assert self.input_dim == kernel.input_dim
@ -67,8 +64,8 @@ class GP(Model):
self.add_parameter(self.likelihood) self.add_parameter(self.likelihood)
def parameters_changed(self): def parameters_changed(self):
self.posterior, self._log_marginal_likelihood, self.grad_dict = self.inference_method.inference(self.kern, self.X, self.likelihood, self.Y, **self.Y_metadata) self.posterior, self._log_marginal_likelihood, self.grad_dict = self.inference_method.inference(self.kern, self.X, self.likelihood, self.Y, self.Y_metadata)
self.likelihood.update_gradients(np.diag(self.grad_dict['dL_dK']), **self.Y_metadata) self.likelihood.update_gradients(self.grad_dict['dL_dthetaL'])
self.kern.update_gradients_full(self.grad_dict['dL_dK'], self.X) self.kern.update_gradients_full(self.grad_dict['dL_dK'], self.X)
def log_likelihood(self): def log_likelihood(self):
@ -99,7 +96,7 @@ class GP(Model):
var = var.reshape(-1, 1) var = var.reshape(-1, 1)
return mu, var return mu, var
def predict(self, Xnew, full_cov=False, **likelihood_args): def predict(self, Xnew, full_cov=False, Y_metadata=None):
""" """
Predict the function(s) at the new point(s) Xnew. Predict the function(s) at the new point(s) Xnew.
@ -123,7 +120,7 @@ class GP(Model):
mu, var = self._raw_predict(Xnew, full_cov=full_cov) mu, var = self._raw_predict(Xnew, full_cov=full_cov)
# now push through likelihood # now push through likelihood
mean, var, _025pm, _975pm = self.likelihood.predictive_values(mu, var, full_cov, **likelihood_args) mean, var, _025pm, _975pm = self.likelihood.predictive_values(mu, var, full_cov, Y_metadata)
return mean, var, _025pm, _975pm return mean, var, _025pm, _975pm
def posterior_samples_f(self,X,size=10, full_cov=True): def posterior_samples_f(self,X,size=10, full_cov=True):
@ -147,7 +144,7 @@ class GP(Model):
return Ysim return Ysim
def posterior_samples(self,X,size=10, full_cov=True,noise_model=None): def posterior_samples(self,X,size=10, full_cov=True, Y_metadata=None):
""" """
Samples the posterior GP at the points X. Samples the posterior GP at the points X.
@ -162,15 +159,7 @@ class GP(Model):
:returns: Ysim: set of simulations, a Numpy array (N x samples). :returns: Ysim: set of simulations, a Numpy array (N x samples).
""" """
Ysim = self.posterior_samples_f(X, size, full_cov=full_cov) Ysim = self.posterior_samples_f(X, size, full_cov=full_cov)
if isinstance(self.likelihood, Gaussian): Ysim = self.likelihood.noise_model.samples(Ysim, Y_metadata)
noise_std = np.sqrt(self.likelihood._get_params())
Ysim += np.random.normal(0,noise_std,Ysim.shape)
elif isinstance(self.likelihood, Gaussian_Mixed_Noise):
assert noise_model is not None, "A noise model must be specified."
noise_std = np.sqrt(self.likelihood._get_params()[noise_model])
Ysim += np.random.normal(0,noise_std,Ysim.shape)
else:
Ysim = self.likelihood.noise_model.samples(Ysim)
return Ysim return Ysim

6
GPy/likelihoods/student_t.py
View file

@ -37,13 +37,13 @@ class StudentT(Likelihood):
def parameters_changed(self): def parameters_changed(self):
self.variance = (self.v / float(self.v - 2)) * self.sigma2 self.variance = (self.v / float(self.v - 2)) * self.sigma2
def update_gradients(self, partial): def update_gradients(self, grads):
""" """
Pull out the gradients, be careful as the order must match the order Pull out the gradients, be careful as the order must match the order
in which the parameters are added in which the parameters are added
""" """
self.sigma2.gradient = partial[0] self.sigma2.gradient = grads[0]
self.v.gradient = partial[1] self.v.gradient = grads[1]
def pdf_link(self, link_f, y, extra_data=None): def pdf_link(self, link_f, y, extra_data=None):
""" """