apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-30 14:35:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

gradient check and debug options

Browse source
This commit is contained in:
Max Zwiessele 2014-03-12 12:06:21 +00:00
parent 2200c5c30b
commit 53e071b892
2 changed files with 9 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

6
GPy/core/gp.py
View file

@ -70,7 +70,7 @@ class GP(Model):
self.posterior, self._log_marginal_likelihood, grad_dict = self.inference_method.inference(self.kern, self.X, self.likelihood, self.Y, Y_metadata=self.Y_metadata) self.posterior, self._log_marginal_likelihood, grad_dict = self.inference_method.inference(self.kern, self.X, self.likelihood, self.Y, Y_metadata=self.Y_metadata)
self.likelihood.update_gradients(np.diag(grad_dict['dL_dK'])) self.likelihood.update_gradients(np.diag(grad_dict['dL_dK']))
self.kern.update_gradients_full(grad_dict['dL_dK'], self.X) self.kern.update_gradients_full(grad_dict['dL_dK'], self.X)
def log_likelihood(self): def log_likelihood(self):
return self._log_marginal_likelihood return self._log_marginal_likelihood
@ -186,7 +186,7 @@ class GP(Model):
""" """
assert "matplotlib" in sys.modules, "matplotlib package has not been imported." assert "matplotlib" in sys.modules, "matplotlib package has not been imported."
from ..plotting.matplot_dep import models_plots from ..plotting.matplot_dep import models_plots
models_plots.plot_fit_f(self,*args,**kwargs) return models_plots.plot_fit_f(self,*args,**kwargs)
def plot(self, *args, **kwargs): def plot(self, *args, **kwargs):
""" """
@ -207,7 +207,7 @@ class GP(Model):
""" """
assert "matplotlib" in sys.modules, "matplotlib package has not been imported." assert "matplotlib" in sys.modules, "matplotlib package has not been imported."
from ..plotting.matplot_dep import models_plots from ..plotting.matplot_dep import models_plots
models_plots.plot_fit(self,*args,**kwargs) return models_plots.plot_fit(self,*args,**kwargs)
def _getstate(self): def _getstate(self):
""" """

7
GPy/core/model.py
View file

@ -339,7 +339,7 @@ class Model(Parameterized):
print "No free parameters to check" print "No free parameters to check"
return return
gradient = self.objective_function_gradients(x) gradient = self.objective_function_gradients(x).copy()
np.where(gradient == 0, 1e-312, gradient) np.where(gradient == 0, 1e-312, gradient)
ret = True ret = True
for nind, xind in itertools.izip(param_index, transformed_index): for nind, xind in itertools.izip(param_index, transformed_index):
@ -350,7 +350,12 @@ class Model(Parameterized):
f2 = self.objective_function(xx) f2 = self.objective_function(xx)
numerical_gradient = (f1 - f2) / (2 * step) numerical_gradient = (f1 - f2) / (2 * step)
if _debug: if _debug:
for p in self.kern.flattened_parameters:
p._parent_._debug=True
self.gradient[xind] = numerical_gradient self.gradient[xind] = numerical_gradient
self._set_params_transformed(x)
for p in self.kern.flattened_parameters:
p._parent_._debug=False
if np.all(gradient[xind]==0): ratio = (f1-f2) == gradient[xind] if np.all(gradient[xind]==0): ratio = (f1-f2) == gradient[xind]
else: ratio = (f1 - f2) / (2 * step * gradient[xind]) else: ratio = (f1 - f2) / (2 * step * gradient[xind])
difference = np.abs((f1 - f2) / 2 / step - gradient[xind]) difference = np.abs((f1 - f2) / 2 / step - gradient[xind])