bmorris3 / aesop Goto Github PK

Hey @tzdwi,

I'm working on two papers right now that could both cite aesop, so why not get that review process started so we can cite it?

I'm starting to draft the JOSS "paper" in the wiki page, which I'm pretty sure you can edit too. Would you give it a look over and consider adding/editing? We can have discussion here in the issue.

continuum_normalize_lstsq has standard_spectrum listed as a parameter, which seems to be a copy&paster error from other normalization.

https://github.com/bmorris3/aesop/blob/master/aesop/spectra.py#L491

Opening this issue to remind myself to do it when I have some spare time.

Currently continuum normalization is via a standard or a least squares polynomial fit. If you don't have a standard star, the least squares polynomial can run into some issues with molecular bands, or particularly broad lines. These issues become evident when trying to stitch orders together. I'd like to implement something like IRAF's continuum task, and fit a spline with iterative outlier rejection.

Thoughts and input welcome.

License on repo is MIT, review paper for JOSS has CC-YA (probably the template default?)

The README needs to be filled out with some basic info. Could be largely the same as in RTD or review summary but needs something.

On cell #8 the five plots are produced by the last line throws an error:

Potentially confusing. Titles are missing on all plots but might not be as important.

Hi, I receive a TypeError: The wcs argument must implement either the high or low level WCS API when using aesop.EchelleSpectrum. How can I solve this issue? I join below the errors I obtain. Thank you in advance!

/usr/local/lib/python3.8/dist-packages/aesop-0.0.dev85-py3.8.egg/aesop/spectra.py in from_fits(cls, path)
329 """
330 spectrum_list = [Spectrum1D.from_specutils(s)
--> 331 for s in read_fits_spectrum1d(path)]
332 header = fits.getheader(path)
333

/usr/local/lib/python3.8/dist-packages/aesop-0.0.dev85-py3.8.egg/aesop/legacy_specutils/readspec.py in read_fits_spectrum1d(filename, dispersion_unit, flux_unit)
553 for spectrum_data, spectrum_wcs in zip(data, multi_wcs.values()):
554 multispec.append(
--> 555 Spectrum1D(spectrum_data, wcs=spectrum_wcs))
556 return multispec
557

/usr/local/lib/python3.8/dist-packages/aesop-0.0.dev85-py3.8.egg/aesop/legacy_specutils/spectrum1d.py in init(self, flux, wcs, unit, uncertainty, mask, meta, indexer, *args, **kwargs)
189 meta=None, indexer=None, *args, **kwargs):
190
--> 191 super(Spectrum1D, self).init(data=flux, unit=unit, wcs=wcs, uncertainty=uncertainty,
192 mask=mask, meta=meta, *args, **kwargs)
193

~/.local/lib/python3.8/site-packages/astropy/nddata/nddata.py in init(self, data, uncertainty, mask, wcs, meta, unit, copy)
229 if wcs is not None:
230 # Validate the wcs
--> 231 self.wcs = wcs
232 self.meta = meta # TODO: Make this call the setter sometime
233 self._unit = unit

/usr/local/lib/python3.8/dist-packages/aesop-0.0.dev85-py3.8.egg/aesop/legacy_specutils/spectrum1d.py in setattr(self, name, value)
246 self._wcs_attributes[name[:-5]]['unit'] = u.Unit(value)
247 else:
--> 248 super(Spectrum1D, self).setattr(name, value)
249
250 def dir(self):

~/.local/lib/python3.8/site-packages/astropy/nddata/nddata.py in wcs(self, wcs)
293 self._wcs = HighLevelWCSWrapper(wcs)
294 else:
--> 295 raise TypeError("The wcs argument must implement either the high or"
296 " low level WCS API.")
297

TypeError: The wcs argument must implement either the high or low level WCS API.

I don't think you necessarily need this for the JOSS review, but I noticed while looking through it that none of the defaults are noted. I'm not sure if that is the particular docstring style you are using but seemed a bit odd to me. I noticed it in the context of some of the boolean options, but it also seems like thing such as the fscale_mad_factor=0.2 should be noted in the description.

Just a comment, feel free to change or not :)

Hi, I'm receiving a TypeError: The wcs argument must implement either the high or low level WCS API when I attempt to initialize an aesop.EchelleSpectrum object for my target spectrum. How do I go about resolving this issue?

https://blog.github.com/2012-09-17-contributing-guidelines/

https://joss.readthedocs.io/en/latest/review_criteria.html#community-guidelines

This code seems to do very well at visualizing and gaining information from 1D spectrum from Echelle spectrographs, after the 2D Echelle spectra have been extracted into a stack of 1D spectra.

How do we take the 2D Echelle spectra (photons vs pixels) with the geometric distortions and needed wavelength calibrations (from comparison spectra) to generate these 1D spectra here.

Is there are section of this package that extracts the 1D spectra from a 2D image?

{silently screaming:
please don't say IRAF...
please don't say IRAF...
please don't say IRAF...
please don't say IRAF...}