Add plot display range

[ericpre]

Description of the change

Improve setting x limits when plotting EDS signal, which can happen in cases, as the one described in #2899. Since this is not the first time that this question is asked and this is simple issue, it would be good to make it more convenient.

Progress of the PR

• Display EDS spectrum from 0.1 keV,
• update docstring (if appropriate),
• [n/a] update user guide (if appropriate),
• add an changelog entry in the upcoming_changes folder (see upcoming_changes/README.rst),
• Check formatting changelog entry in the readthedocs doc build of this PR (link in github checks)
• add tests,
• ready for review.

Minimal example of the bug fix or the new feature

From the example given in #2899

import hyperspy.api as hs

# For Signal1D
s = hs.load('2030 8.00 Mx 0001.emd')[-1]
s.sum().plot()

# For Signal2D
ima = hs.datasets.example_signals.object_hologram()
ima.plot(display_range=(150, 400, 200, 450))

Before this PR:

With this PR:

[codecov]

Codecov Report

Merging #2958 (5790878) into RELEASE_next_patch (8d37cc1) will increase coverage by 0.01%.
The diff coverage is 95.34%.

@@                  Coverage Diff                   @@
##           RELEASE_next_patch    #2958      +/-   ##
======================================================
+ Coverage               80.79%   80.80%   +0.01%     
======================================================
  Files                     209      209              
  Lines                   32696    32726      +30     
  Branches                 7321     7336      +15     
======================================================
+ Hits                    26416    26444      +28     
- Misses                   4516     4517       +1     
- Partials                 1764     1765       +1     

Impacted Files	Coverage Δ
hyperspy/docstrings/plot.py	100.00% <ø> (ø)
hyperspy/drawing/signal1d.py	80.61% <93.75%> (+0.55%)	⬆️
hyperspy/_signals/eds.py	85.71% <100.00%> (+0.15%)	⬆️
hyperspy/drawing/image.py	74.30% <100.00%> (+0.36%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 8d37cc1...5790878. Read the comment docs.

[francisco-dlp]

I am not sure if it is a good idea to adjust the x-limit to 0.1 keV automatically—this hides a feature of the data from the user, what can cause other issues. Wouldn't it be better to simply document this particularity of EDX datasets and suggest to crop the data using s.isig[0.1:] as a cure?

[francisco-dlp]

Also, this looks more like an "enhancement" than a bug-fix to me. Do you agree @ericpre?

[francisco-dlp]

This assumes that the units are keV.

By the way, is this documented?

[ericpre]

Yes, you are right!

[ericpre]

I am not sure if it is a good idea to adjust the x-limit to 0.1 keV automatically—this hides a feature of the data from the user, what can cause other issues. Wouldn't it be better to simply document this particularity of EDX datasets and suggest to crop the data using s.isig[0.1:] as a cure?

I was torn on this for a while and was thinking the same, however, I end up convincing myself that there is a use case to have this:

• the issue with high intensity of zero energy peak in EDS spectrum is mentioned once in a while by users. As you said, this can be work around by using isig, but it come at the cost of making a copy of the data, which is not always an option. Assuming that 99% of the user aren't interested in the zero energy peak, it is useful to have it as default. The other 1% would be advanced EDS user and spend effort looking at the data can afford changing the default setting to their preference.
• for Signall2D, I find it useful to be able to specify the range of displayed data without having to make a copy. An alternative to this would be use matplotlib differently, but there is currently a bug with the scalebar and it is more verbose.

Overall, because it is generic and the logic is simple (I can't think of any issue that it could cause and I don't expect any because it is simple logic, EDIT: clearly this statement didn't hold very long!! :D See #2958 (comment)), I came to the conclusion that this is a good compromise. The EDS plotting uses specific default values, which is not documented correctly.

Also, this looks more like an "enhancement" than a bug-fix to me. Do you agree @ericpre?

Yes, it is mostly an enhancement, on the edge being a bug! ;)

[francisco-dlp]

isig doesn't make a copy of the data, just a view, so it shouldn't be a problem to use it/

The issue I see is that the offending channel is also an issue for decomposition and curve fitting. By taking it out of the view, we hide the issue, making it harder for the users to understand what's wrong in their analysis.