apunkt/GPy
1
0
Fork 0
mirror of https://github.com/SheffieldML/GPy.git synced 2026-06-05 14:55:15 +02:00
Code Issues Projects Releases Packages Wiki Activity Actions

Some bugfixes that have affected GPLVM/sparseGPLVM since the hetero noise change

Browse source
This commit is contained in:
Nicolas 2013-02-21 15:21:37 +00:00
parent a166187231
commit 3f924bea8f
2 changed files with 6 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

5
GPy/models/sparse_GPLVM.py
View file

@ -42,7 +42,8 @@ class sparse_GPLVM(sparse_GP_regression, GPLVM):
return sparse_GP_regression.log_likelihood(self)
def dL_dX(self):
dL_dpsi1 = self.dL_dpsi1 + 2.*np.dot(self.dL_dpsi2,self.psi1)
#dL_dpsi1 = self.dL_dpsi1 + 2.*np.dot(self.dL_dpsi2,self.psi1)
dL_dpsi1 = self.dL_dpsi1 + 2.*np.dot(self.dL_dpsi2[0,:,:],self.psi1)
dL_dX = self.kern.dKdiag_dX(self.dL_dpsi0,self.X)
dL_dX += self.kern.dK_dX(dL_dpsi1.T,self.X,self.Z)
@ -55,5 +56,5 @@ class sparse_GPLVM(sparse_GP_regression, GPLVM):
def plot(self):
GPLVM.plot(self)
#passing Z without a small amout of jitter will induce the white kernel where we don;t want it!
mu, var = sparse_GP_regression.predict(self, self.Z+np.random.randn(*self.Z.shape)*0.0001)
mu, var, upper, lower = sparse_GP_regression.predict(self, self.Z+np.random.randn(*self.Z.shape)*0.0001)
pb.plot(mu[:, 0] , mu[:, 1], 'ko')