multiwavelength / gleam Goto Github PK

Change the way that metadata files are searched on disk, ie using the name 'meta.' instead of master.

The zoom-in plots are sometimes too narrow and lots of space is left on the right side. This is because the number of zoom-in plots is calculated as a function of the number of connected line components. However, some of these spectral lines might not be covered by the spectrum (i.e. too low or too high wavelength), so a fit is never attempted for them. The number of plots should thus take into account only the components that are covered by the spectrum.

While the tolerance parameter offers a very succinct way to uniquely define the groups of lines to be fitted jointly, it requires some validation on the part of the user to make sure there are no surprising groupings.

Here are two things we could consider to avoid any confusion:

1. A sub-command (gleam list groups) to list the groups (for each override scenario) so the user can easily inspect them.
2. Explicitly define the groups in the config file, using line names:

   groups:
     - 
       - Halpha
       - NII1
       - NII2
     -
       - SII1
       - SII2

   All other lines would be fit individually.

At the moment, gleam works really well for narrow emission lines (such as those associated with narrow-line regions in AGN and star-forming regions). This is because the starting solution for the Gaussian fits is set to the resolution of the instruments. This starting solution also works for broad lines (>500 km/s), as long as the line is well separated from other lines and the fit is simple. However, the fits do not converge to the correct solution when one has both broad and narrow components of the same emission line. To fix this, we should give users the option to label lines in the spectral lines list as broad and provide a different starting solution for the Gaussian fits, in line with expectations for broad lines (i.e. a FWHM starting solution equivalent to 500-1000 km/s)

In the example (A773), the meta file does not start with "meta." so that user will get the following message:

Error! Cannot find any metadata files.

Renameing the file to "meta.A773.WHT_R316R.Q1_EXT1.dat" solves the issue.

At the moment, gleam outputs the fit results and plots next to the corresponding input spectra.

It would be great to have the options to output the results to a different folder. This can help:

• speed up the execution when the input data is hosted on a slower drive (like a network drive).
• back up the results without having to pick them out from among the data.
• produce comparable & archivable snapshots of different runs with different input configurations. Maybe the input configuration could be copied or somehow linked to the output folder in order to help with keeping track of the results.

GLEAM did not scale the flux density (lambda) when restframing the spectrum. This resulted in fluxes, luminosities and Gaussian amplitudes that were a factor of (1+z) too small. Equivalent widths are not affected. The effect is small at small redshifts, but very important for high redshifts.

Mariusz Tarnopolski noticed the following issue in GLEAM:

"When an absorption line is fitted, the corresponding amplitude is negative, which results in a negative EW_rest (which can be positive or negative depending on convention), but then - as per line 80 of gaussian_fitting.py - this negative value it carried on to the error, which should always be positive. So I just changed v to abs(v) in that line of the code."

This should be an easy fix as Mariusz suggests.

How can I enable plotting in gleam?

Error when processing multi files, but process single file works

(py38_gleam) zijianzhang@fushuqis-iMac formated_Data % gleam --path gleam_Data --spectra .//spec1d.MMT.Hectlow.a.24.fits --config gleamconfig.yaml --plot
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
Usage: gleam [OPTIONS]
Try 'gleam --help' for help.

Error: Got unexpected extra argument (./gleam_Data/spec1d.MMT.Hectlow.a.243.fits)
(py38_gleam) zijianzhang@fushuqis-iMac formated_Data % gleam --path gleam_Data --spectra ./**/spec1d.MMT.Hectlow.a.243.fits --config gleamconfig.yaml --plot
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
Now working in ./gleam_Data on MMT in Hectlow + a on source 243 at z=0.322.

With different installations of GLEAM, I am getting different behaviors on the empty cell in the output FITS file that contains the emission and absorption line fits.

If I run GLEAM from a local GLEAM sandbox (clone of git repository), the cell is empty, as intended.
If I run GLEAM from my PyPI installation, the cell is filled with 1.0E20, not as intended.

Both GLEAM installations use Astropy 4.2.

This could be an issue solved by using more recent versions of dependencies. At the moment we cannot upgrade dependencies until we figure out how to deal with #26.

The defaults for the fitting parameters are hidden in the code. It would be helpful if we had an option to print them for the user to see.

I am getting two warnings when importing gleam:

>>> import gleam
WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]
/home/victor/Code/GitHub/gleamenv/lib/python3.8/site-packages/gleam/matplotlibparams.py:12: MatplotlibDeprecationWarning: Support for setting the 'text.latex.preamble' or 'pgf.preamble' rcParam to a list of strings is deprecated since 3.3 and will be removed two minor releases later; set it to a single string instead.
  plt.rcParams["text.latex.preamble"] = [

When importing/running GLEAM this issue arises:

WARNING: AstropyDeprecationWarning: The truth value of a Quantity is ambiguous. In the future this will raise a ValueError. [astropy.units.quantity]

This was noticed by @Victor-Savu in #9.
Preliminary investigations indicate the issue is caused by how Pydantic dataclasses interact with Astropy Quantities.

Since Matplotlib 3.4, closing a figure in interactive mode hangs the process. We fixed the required Matplotlib version to 3.3.4 as a temporary fix.