Skip to content

Fix: MultivariateQExponential.log_prob ignores power when fast log-prob is off #1

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

Open
richardcsuwandi wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix: MultivariateQExponential.log_prob ignores power when fast log-prob is off #1

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
7 changes: 6 additions & 1 deletion qpytorch/distributions/multivariate_qexponential.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -241,7 +241,12 @@ def log_prob(self, value: Tensor) -> Tensor:
See :py:meth:`torch.distributions.Distribution.log_prob
<torch.distributions.distribution.Distribution.log_prob>`.
"""
if settings.fast_computations.log_prob.off():
# The torch.distributions.MultivariateNormal fallback is Gaussian-only.
# Only use it when power == 2. When fast_computations.log_prob is off
# (e.g. BoTorch disables it on import; QePyTorch shares gpytorch.settings),
# falling back for power != 2 silently ignores `power` and returns the
# Gaussian log-density.
if settings.fast_computations.log_prob.off() and bool(torch.as_tensor(self.power == 2).all()):
return super().log_prob(value)

if self._validate_args:
Expand Down