apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-05-24 14:15:14 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

first crack at a caching object

Browse source
This commit is contained in:
James Hensman 2014-02-06 12:33:00 +00:00
parent 6346af8764
commit 54272d22bb
4 changed files with 23 additions and 26 deletions
Download patch file Download diff file Expand all files Collapse all files

41
GPy/kern/parts/prod.py
View file

@ -19,36 +19,21 @@ class Prod(Kernpart):
"""
def __init__(self,k1,k2,tensor=False):
if tensor:
super(Prod, self).__init__(k1.input_dim + k2.input_dim, '['+k1.name + '**' + k2.name +']')
super(Prod, self).__init__(k1.input_dim + k2.input_dim, k1.name + '_xx_' + k2.name)
self.slice1 = slice(0,k1.input_dim)
self.slice2 = slice(k1.input_dim,k1.input_dim+k2.input_dim)
else:
assert k1.input_dim == k2.input_dim, "Error: The input spaces of the kernels to sum don't have the same dimension."
super(Prod, self).__init__(k1.input_dim, '['+k1.name + '*' + k2.name +']')
#self.num_params = k1.num_params + k2.num_params
assert k1.input_dim == k2.input_dim, "Error: The input spaces of the kernels to multiply don't have the same dimension."
super(Prod, self).__init__(k1.input_dim, k1.name + '_x_' + k2.name)
self.slice1 = slice(0,self.input_dim)
self.slice2 = slice(0,self.input_dim)
self.k1 = k1
self.k2 = k2
# if tensor:
# self.slice1 = slice(0,self.k1.input_dim)
# self.slice2 = slice(self.k1.input_dim,self.k1.input_dim+self.k2.input_dim)
# else:
# self.slice1 = slice(0,self.input_dim)
# self.slice2 = slice(0,self.input_dim)
self._X, self._X2 = np.empty(shape=(2,1))
self.add_parameters(self.k1, self.k2)
# self._set_params(np.hstack((k1._get_params(),k2._get_params())))
# def _get_params(self):
# """return the value of the parameters."""
# return np.hstack((self.k1._get_params(), self.k2._get_params()))
#
# def _set_params(self,x):
# """set the value of the parameters."""
# self.k1._set_params(x[:self.k1.num_params])
# self.k2._set_params(x[self.k1.num_params:])
#
# def _get_param_names(self):
# """return parameter names."""
# return [self.k1.name + '_' + param_name for param_name in self.k1._get_param_names()] + [self.k2.name + '_' + param_name for param_name in self.k2._get_param_names()]
#initialize cache
self._X, self._X2 = np.empty(shape=(2,1))
self._params = None
def K(self,X,X2,target):
self._K_computations(X,X2)
@ -64,6 +49,11 @@ class Prod(Kernpart):
self._K_computations(X, X2)
return self._K2
def update_gradients_full(self, dL_dK, X):
self._K_computations(X, None)
self.k1.update_gradients_full(dL_dK*self._K2, X[:,self.slice1])
self.k2.update_gradients_full(dL_dK*self._K1, X[:,self.slice2])
def dK_dtheta(self,dL_dK,X,X2,target):
"""Derivative of the covariance matrix with respect to the parameters."""
self._K_computations(X,X2)
@ -82,6 +72,7 @@ class Prod(Kernpart):
self.k2.Kdiag(X[:,self.slice2],target2)
target += target1 * target2
def dKdiag_dtheta(self,dL_dKdiag,X,target):
K1 = np.zeros(X.shape[0])
K2 = np.zeros(X.shape[0])