[Fix] SCR_Peaks_Amplitude_Mean should be only computed on peaks

[mperreir]

Description

For the moment eda_intervalrelated() computes the mean peaks amplitude based on the whole signal (where most of the amplitude values are zero). This leads to very low (and wrong) SCR_Peaks_Amplitude_Mean values.

Proposed Changes

Compute the mean peaks amplitude only on the peaks (and not on the whole signal). For that we first select rows where SCR_Peaks has a value of 1.

Checklist

Here are some things to check before creating the PR. If you encounter any issues, do let us know :)

• I have read the CONTRIBUTING file.
• My PR is targeted at the dev branch (and not towards the master branch).
• I ran the CODE CHECKS on the files I added or modified and fixed the errors.
• I have added the newly added features to News.rst (if applicable)

[codecov-commenter]

Codecov Report

Attention: 1 lines in your changes are missing coverage. Please review.

Comparison is base (44959c6) 55.00% compared to head (984194a) 55.01%.

Additional details and impacted files

@@           Coverage Diff           @@
##              dev     #917   +/-   ##
=======================================
  Coverage   55.00%   55.01%           
=======================================
  Files         303      303           
  Lines       14210    14213    +3     
=======================================
+ Hits         7816     7819    +3     
  Misses       6394     6394           

Files	Coverage Δ
neurokit2/eda/eda_intervalrelated.py	78.26% <87.50%> (+1.51%)	⬆️

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[DominiqueMakowski]

The docstring example now fails I think

https://github.com/neuropsychology/NeuroKit/actions/runs/6382716523/job/17321902504?pr=917#step:5:99

[mperreir]

Yes, it seems. I was not expecting that. It might be fixed with something like this:

only_peaks = data["SCR_Peaks"] == 1
if only_peaks.sum() > 0:
    output["SCR_Peaks_Amplitude_Mean"] = np.nanmean(data[data["SCR_Peaks"] == 1]["SCR_Amplitude"].values)
else:
    output["SCR_Peaks_Amplitude_Mean"] = np.nan

I'll try to push this tomorrow.