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.

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

Adds Annealed Importance Sampling #550

Draft
wants to merge 6 commits into
base: develop
Choose a base branch
from

Conversation

BradyPlanden
Copy link
Member

Description

Adds an annealed importance sampling class for marginal likelihood computation.

Issue reference

Fixes #549

Review

Before you mark your PR as ready for review, please ensure that you've considered the following:

  • Updated the CHANGELOG.md in reverse chronological order (newest at the top) with a concise description of the changes, including the PR number.
  • Noted any breaking changes, including details on how it might impact existing functionality.

Type of change

  • New Feature: A non-breaking change that adds new functionality.
  • Optimization: A code change that improves performance.
  • Examples: A change to existing or additional examples.
  • Bug Fix: A non-breaking change that addresses an issue.
  • Documentation: Updates to documentation or new documentation for new features.
  • Refactoring: Non-functional changes that improve the codebase.
  • Style: Non-functional changes related to code style (formatting, naming, etc).
  • Testing: Additional tests to improve coverage or confirm functionality.
  • Other: (Insert description of change)

Key checklist:

  • No style issues: $ pre-commit run (or $ nox -s pre-commit) (see CONTRIBUTING.md for how to set this up to run automatically when committing locally, in just two lines of code)
  • All unit tests pass: $ nox -s tests
  • The documentation builds: $ nox -s doctest

You can run integration tests, unit tests, and doctests together at once, using $ nox -s quick.

Further checks:

  • Code is well-commented, especially in complex or unclear areas.
  • Added tests that prove my fix is effective or that my feature works.
  • Checked that coverage remains or improves, and added tests if necessary to maintain or increase coverage.

Thank you for contributing to our project! Your efforts help us to deliver great software.

Copy link

codecov bot commented Nov 8, 2024

Codecov Report

Attention: Patch coverage is 71.83099% with 20 lines in your changes missing coverage. Please review.

Project coverage is 98.68%. Comparing base (c8b00e6) to head (c5eef05).
Report is 13 commits behind head on develop.

Files with missing lines Patch % Lines
pybop/parameters/priors.py 9.09% 10 Missing ⚠️
pybop/samplers/annealed_importance.py 83.05% 10 Missing ⚠️
Additional details and impacted files
@@             Coverage Diff             @@
##           develop     #550      +/-   ##
===========================================
- Coverage    99.08%   98.68%   -0.41%     
===========================================
  Files           52       56       +4     
  Lines         3605     3941     +336     
===========================================
+ Hits          3572     3889     +317     
- Misses          33       52      +19     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Copy link
Contributor

@martinjrobins martinjrobins left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the algorithm doesn't seem correct to me, but I could be miss-reading it. In addition, I would recommend writing a test for the algorithm, using a known distribution. I'd use the same unimodel distribution used in the paper, so you can compare with its results.

pybop/samplers/annealed_importance.py Outdated Show resolved Hide resolved
pybop/samplers/annealed_importance.py Outdated Show resolved Hide resolved
# Main sampling loop
for j in range(1, self._num_beta):
# Compute jth transition with current sample
log_density_current[j] = self.transition_distribution(current, j)
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

you are doing f_j(x_j) / f_{j-1}(x_j), should be f_j(x_j) / f_{j+1}(x_j) from eqn 11

log_density_current = np.zeros(self._num_beta)
log_density_current[0] = current_f
log_density_previous = np.zeros(self._num_beta)
log_density_previous[0] = current_f
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

same here, log_density_previous[0] should be f_{1}(x_0) from eqn 11

proposed_f = self.transition_distribution(proposed, j)

# Metropolis acceptance
acceptance_log_prob = proposed_f - current_f
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

for the gaussian test case they use in the paper, a much more complicated transition T_j is used, a sequence of 3 metropolis tests repeated 5-10 times. I'm not sure if all that is neccessary however

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Add marginal log likelihood sampling
2 participants