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

Add support for inessai sampler #4694

Draft
wants to merge 4 commits into
base: master
Choose a base branch
from

Conversation

mj-will
Copy link
Contributor

@mj-will mj-will commented Apr 12, 2024

Standard information about the request

This is a: bug fix, new feature, efficiency update, other (please describe)

This change affects: inference

This change changes: the nessai sampler to support the importance nested sampler version

This change: is tested via an example, follows style guidelines (See e.g. PEP8), has been proposed using the contribution guidelines

This change will: does not break any existing code

Motivation

With the release of nessai 0.12.0, a method for mapping from the prior to the unit hypercube is no longer required for the importance nested sampler (ins). This means we can support the ins (inessai) with a few minor changes to the existing implementation.

Contents

Updates to the existing class so that it supports the importance_nested_sampler option. This is False by default.

I also added a second class INessaiSampler that has this default switch to True and has the name set to inessai.

This means there are two ways of calling the importance nested sampler.

Testing performed

Run the example included in examples/inference/samplers/

Additional notes

To-do

  • Test checkpoint after this change, since it reworks the samples property.

  • Test on BBH

  • The author of this pull request confirms they will adhere to the code of conduct

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.

None yet

1 participant