(very simple to solve) bug in passing evidence_tolerance (aka dlogz) to ultranest

JohannesBuchner / BXA

Bayesian X-ray analysis (nested sampling for Xspec and Sherpa)

https://johannesbuchner.github.io/BXA/

GNU General Public License v3.0

57 stars 19 forks source link

(very simple to solve) bug in passing evidence_tolerance (aka dlogz) to ultranest #40

Closed ferrigno closed 2 years ago

ferrigno commented 2 years ago

BXA version: 4.0.6
UltraNest version: 3.5.5
Python version:3.8.10
Xspec or Sherpa and version: XSPEC version: 12.12.0
Operating System: Linux

Description

When passing the parameter evidence_tolerance the run does not work

What I Did

line 194 of solver.py put run_kwargs['dlogz'] = run_kwargs.pop('dlogz', evidence_tolerance) instead of run_kwargs['evidence_tolerance'] = run_kwargs.pop('dlogz', evidence_tolerance)

It fixes easily the issue.

JohannesBuchner commented 2 years ago

Yes, this should be fixed!

ferrigno commented 2 years ago

the line number is for bxa/xspec/solver.py but the same issue is also present for the sherpa version

JohannesBuchner commented 2 years ago

If you have time to prepare a pull request, that would be great.

JohannesBuchner commented 2 years ago

Thank you for the contribution!