DOC - document the units of the inverse operation for eLORETA

[robertoostenveld]

• added check that the user did not specify leadfield options in case the input already contains the leadfield
• added check that the user did not specify a regularization parameter in case the input already contains the filter

[robertoostenveld]

This follows from a conversation with @mcpiastra. The page https://www.fieldtriptoolbox.org/faq/units/ is related, as are the occasional questions on the mailing list such as https://mailman.science.ru.nl/pipermail/fieldtrip/2021-June/040932.html

It would be nice to have the units in general better documented. Documenting them at the lowest accessible level (i.e. ft_inverse_xxx) seems feasible. Documenting them at a higher level is more difficult, since there can be so many choices in the analysis pipeline that affect this. What might be good to document generally though is that inverse computations are not arbitrary if all data, geometrical objects and models (like the leadfield) are specified in SI units.

[robertoostenveld]

@schoffelen can I ask you to review this content-wise? I will make sure to regression-test the checks for the leadfield and filter options being empty before it gets merged.

[mcpiastra]

hi @robertoostenveld and @schoffelen ,
there are 10 left ft_inverse_something where I would now add checks for the leadfield. is one PR for all functions a good strategy?
thanks,
MC

[robertoostenveld]

is one PR for all functions a good strategy?

Yes, that would be fine. Probably the changes will be very similar to all functions, so that facilitates review.