jamespaynter / pygrb Goto Github PK

Running the example here: https://pygrb.readthedocs.io/en/latest/user/usage.html

The output results in some nans to the stdio:

22:41 bilby INFO    : Sampling time: 0:00:18.407568
22:41 bilby INFO    : Summary of results:
nsamples: 4645
log_noise_evidence:    nan
log_evidence: -3709.519 +/-  0.310
log_bayes_factor:    nan +/-  0.310

is this an error?

The data loading class appears to be able to read both TTE and pre-binned counts data. However, I cannot find the functionality to handle the binning of the TTE data.

1. How would TTE data be handled?

2. I see that you convert rates to counts by multiplying with the duration of the time bin. However, this does not account for the instrumental dead time in the time bin and it results in non-integer values for the Poisson likelihood. It is better to get the counts directly, and multiply the rate function you are estimating by the exposure = bin duration - dead time. This maintains numerical stability and is mathematically proper.

I'm not sure what info is in the BATSE rate files at the moment, but will investigate.

JOSS requires:

 Community guidelines: Are there clear guidelines for third parties wishing to 1) Contribute to the software 2) Report issues or problems with the software 3) Seek support

The only contribution information I see is at the bottom of the README.md, which looks more like a TODO list.

Can you expand this section, perhaps explaining that contributions are welcomed in the form of PRs against this repo, support is via issues, etc.

Following the usage instructions here: https://pygrb.readthedocs.io/en/latest/user/usage.html

I get output in a products directory, which contains the first plot shown in the usage docs but not the others.

In particular, I don't get this plot:
https://pygrb.readthedocs.io/en/latest/_images/trigger7475chan3.PNG

or the corner plot.

The usage instructions should walk through how those plots are created.

When doing:

from PyGRB.main.fitpulse import PulseFitter

I get the errors:

22:38 bilby WARNING : You do not have gwpy installed currently. You will  not be able to use some of the prebuilt functions.
22:38 bilby WARNING : You do not have lalsuite installed currently. You will not be able to use some of the prebuilt functions.
22:38 bilby WARNING : You do not have gwpy installed currently. You will  not be able to use some of the prebuilt functions.
22:38 bilby WARNING : You do not have lalsuite installed currently. You will not be able to use some of the prebuilt functions.
22:38 bilby WARNING : You do not have gwpy installed currently. You will  not be able to use some of the prebuilt functions.
22:38 bilby WARNING : You do not have lalsuite installed currently. You will not be able to use some of the prebuilt functions.
22:38 bilby WARNING : You do not have lalsuite installed currently. You will not be able to use some of the prebuilt functions.

These packages are not part of the requirements.txt and I don't see mention of them in the docs.

Are these okay? Perhaps the install docs should address this.

Would you be interested to add support for https://johannesbuchner.github.io/UltraNest/ ?
The interface should be very similar to pymultinest.

UltraNest is a very reliable tuning-parameter-free algorithm. I have published some examples where the UltraNest algorithm is unbiased while multinest's algorithm (implemented in pymultinest,nestle,dynesty) gives a different answer. UltraNest is a pure-python package and very easy to install with pip or conda.

UltraNest also supports resuming from disk and MPI parallelisation, if that is useful to you.

Perhaps this is already possible for you through bilby.