search

adivijaykumar / bilby

MIT License
0 stars 2 forks source link

Fixes to the source model for tests to pass #32

Closed adivijaykumar closed 3 years ago

adivijaykumar commented 3 years ago

https://git.ligo.org/aditya.vijaykumar/bilby_github/-/pipelines/192851 -- tests pass. @Kruthi24 could you test this branch with the likelihood notebooks like we had discussed?

Please also feel free to review the changes in this branch.

Kruthi24 commented 3 years ago

@Kruthi24 could you test this branch with the likelihood notebooks like we had discussed?

I checked if the likelihoods matched for fiducial parameters and if the strain calculated with get_detector_response_relative_binning and get_detector_response matched at the bin edges. Anything else?

Kruthi24 commented 3 years ago

I also ran the sampler.

adivijaykumar commented 3 years ago

Thanks! Could you also calculate the likelihood for parameters different from the fiducial waveform and see if they match?

Kruthi24 commented 3 years ago

Could you also calculate the likelihood for parameters different from the fiducial waveform and see if they match?

I checked this for a couple of values. They were in the same ballpark always

adivijaykumar commented 3 years ago

Thanks! Shall I merge then?

Kruthi24 commented 3 years ago

Yup!