eyetrack should not count as picks=data

[behinger]

Reference issue (if any)

Fixes #13657

What does this implement/fix?

Set picks='data' for eyegaze to False, as already in the documentation/glossary

Additional information

I had to fix one test, running plot_compare_evokeds on pupil data, because picks='data'was the default. with the change, that ended in no picks remaining, thus erroring. This fixes the test, by adding picks="eyetrack"

[larsoner]

Probably needs a doc/changes/dev/13723.bugfix.rst?

@scott-huberty probably a good one for you to review and merge once you're happy

[behinger]

• I'll add the rst asap
• I'll also setup circleci

[scott-huberty]

Thanks @behinger !

A few years ago I wrote a test to make sure we do treat eyetracking data as data channels, but I guess it was not comprehensive enough:

mne-python/mne/io/eyelink/tests/test_eyelink.py

Line 40 in 807d51b

def test_eyetrack_not_data_ch():

[behinger]

I didnt see that - I added a test just in case :) - btw. your post on github says it the wrong way round, eyetrack is NOT data channel (the code / docs etc. all correct)

[scott-huberty]

didnt see that - I added a test just in case :) - btw. your post on github says it the wrong way round, eyetrack is NOT data channel (the code / docs etc. all correct)

Arg sorry @behinger, I wrote my message too hastily. Yes I meant to say that eyetrack channel types are NOT data channels! Thank you for updating the test! Marking for merge when green.

[welcome]

🎉 Congrats on merging your first pull request! 🥳 Looking forward to seeing more from you in the future! 💪