search

JohannesBuchner / UltraNest

Fit and compare complex models reliably and rapidly. Advanced nested sampling.
https://johannesbuchner.github.io/UltraNest/
Other
147 stars 30 forks source link

Reading ultranest results without likelihood / cube transform #149

Closed Tenivar closed 1 month ago

Tenivar commented 1 month ago

Description

Hi, (I hope this question has not been answered before, but I did not find it ! ) I'm looking for a way to load Ultranest results from the log directory (e.g., to plot some diagnostics, corner plots, trace …) without having to provide a cube transform and a likelihood function, and “run” the nested sampler again (even if it actually just read the results). Is there a way to do that? Thank you for the great tool and work :-)

JohannesBuchner commented 1 month ago

You are looking for this function: https://johannesbuchner.github.io/UltraNest/ultranest.html#ultranest.integrator.read_file

Maybe you can think about where you looked so that we can integrate it better in the documentation?

JohannesBuchner commented 1 month ago

Alternatively, you can also load the posterior samples as csv files.

Tenivar commented 1 month ago

Thanks a lot! I was actually looking for a function like ReactiveNestedSampler.load_results (in the same page you sent, my bad, I have not looked closely enough).