Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Removed unncessary sigma in function elbo of gplvm.py (and save N^2+1 operations) #1685

Open
wants to merge 2 commits into
base: develop
Choose a base branch
from
Open

Removed unncessary sigma in function elbo of gplvm.py (and save N^2+1 operations) #1685

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
531af34
Removed unncessary calculation/use of sigma
joacorapela May 16, 2021
9f110b6
Merge remote-tracking branch 'upstream/develop' into improve_gplvm_elbo
joacorapela May 26, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 3 additions & 3 deletions gpflow/models/gplvm.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -149,6 +149,7 @@ def maximum_log_likelihood_objective(self) -> tf.Tensor:
return self.elbo()

def elbo(self) -> tf.Tensor:

"""
Construct a tensorflow function to compute the bound on the marginal
likelihood.
Expand All @@ -169,16 +170,15 @@ def elbo(self) -> tf.Tensor:
cov_uu = covariances.Kuu(self.inducing_variable, self.kernel, jitter=default_jitter())
L = tf.linalg.cholesky(cov_uu)
sigma2 = self.likelihood.variance
sigma = tf.sqrt(sigma2)

# Compute intermediate matrices
A = tf.linalg.triangular_solve(L, tf.transpose(psi1), lower=True) / sigma
A = tf.linalg.triangular_solve(L, tf.transpose(psi1), lower=True)
tmp = tf.linalg.triangular_solve(L, psi2, lower=True)
AAT = tf.linalg.triangular_solve(L, tf.transpose(tmp), lower=True) / sigma2
B = AAT + tf.eye(num_inducing, dtype=default_float())
LB = tf.linalg.cholesky(B)
log_det_B = 2.0 * tf.reduce_sum(tf.math.log(tf.linalg.diag_part(LB)))
c = tf.linalg.triangular_solve(LB, tf.linalg.matmul(A, Y_data), lower=True) / sigma
c = tf.linalg.triangular_solve(LB, tf.linalg.matmul(A, Y_data), lower=True) / sigma2

# KL[q(x) || p(x)]
dX_data_var = (
Expand Down