Weird Plotting Glitch

[ericasaw]

Was working on plotting some IGRINS spectra order by order this morning and I found this weird glitch. When you execute a line like specs[0][order].normalize().plot() (where specs is a list of IGRINSSpectrumLists) the plotted spectrum for a particular order is nicely normalized, but if you initialize a matplotlib figure and subplot and instead execute specs[0][order].normalize().plot(ax = ax) the plotted order is not normalized. If instead you set your new axis equal to the plot for the normalized order (ax = specs[0][order].normalize().plot()) the order is normalized and you can change things about the plot like the title, range, etc. Thought it was kind of weird, is anyone else running into this?

[gully]

My guess is this is a copy issue--- whether the .plot operates on a copy or a view of the original. As a workaround, I recommend making an intermediate spectrum:

temp_spec = specs[0][order].normalize()
temp_spec.plot(ax = ax)

or maybe even:

temp_spec = specs[0][order]
temp_spec = temp_spec.normalize()
temp_spec.plot(ax = ax)

Merely a workaround until a scrupulous look at how/whether intermediate copies are made. The copying behavior is a legacy of how specutils stores SpectrumCollections. Someday I'd like to get a specutils contributor to give a critical look at muler-- they'd probably have some good guidance.

[gully]

Hi @ericasaw is this plotting glitch still an issue? If so, can you provide a minimal reproducible example, and which version of muler you are using?