Repository navigation
Conversation
… compliant files, not just specific to NuSTAR.
… class then everything is done for NuSTAR at least
…ght with pre-commit to stop it changing livetime to lifetime and lost (will return to this).
|
Pre-commit fails because of my |
| "sunkit_spex>=0.5.0", | ||
| "ndcube>=2.4.1", |
There was a problem hiding this comment.
I'm not sure how this would work if sunpy_spex also depends on sunkit_inst maybe it's a non-issue
There was a problem hiding this comment.
I'm not sure either but I was thinking, will sunkit_spex depend on sunkit_inst?
|
|
||
| __all__ = ["regroup_any_array", "rebin_rmf"] | ||
|
|
||
| def regroup_any_array(data:np.ndarray|u.Quantity, old_bins:np.ndarray|u.Quantity, new_bins:np.ndarray|u.Quantity, combine_by:str|None=None): |
There was a problem hiding this comment.
So I tink this should live in sunkit-spex on an ARM / RMF object and use/extend NDCube API to do the rebin
There was a problem hiding this comment.
I'm not sure where my thoughts are on that.
My main thought is as long as it's easily gettable and I could see why a user would want to rebin the data prior to fitting, or maybe after for plotting, but I'm not sure of a use case that the rebinning should be done during fitting.
I feel a user should get the data in whatever stage they want and then fit that. In that case, I feel that the object handling the spectral data should also be able to handle the rebinning. I would worry that an instrument would need rebinning for either the ARF/RMF to be handled in a weird way meaning that an ARF/RMF object's rebin method would need to be edited on a instrument basis anyway.
Ultimately, it can go wherever but I just wanted to put some thoughts out there.
|
@samaloney , I think this is starting to take its final shape. The only outstanding thing for me to change is _exactly_the form of the spectrum object. Other than that, I think it a case of reviewing, tearing this apart, and getting it merged. I'd imagine that the NuSTAR files I'm hosting to download for the example would be better hosted elsewhere and I'm really happy to take guidance on that. Since this is actually quite nice to use to look at NuSTAR data outside of spectral fitting, I'll also get other NuSTAR folk to have a look at some point. |
2 participants
I'm opening as a work-in-progress draft but more than happy for comments.
There is a bunch of instrument specific code in the
sunkit-spexX-ray spectral fitting package and on local machines to load in spectra from those instruments and fit it. This code does not fall into the scope forsunkit-spexand does not have a home anywhere else.Therefore, this is the first PR at moving, and updating, the instrument specific code to
sunkit-instruments. The main focus here is for the NuSTAR observatory, but other PRs for other instrument code that doesn't have a home will likely come soon.To do: