MRG, ENH: Improve Epochs._name and Evoked.comment#9027
MRG, ENH: Improve Epochs._name and Evoked.comment#9027drammock merged 15 commits intomne-tools:mainfrom
Conversation
I'm surprised that it allows × but not − ... I'm guessing the FIF format encodes text in something bigger than ASCII but not as big as UTF-8? BTW, this is still marked as draft. LMK when it's ready for review. |
agramfort
left a comment
There was a problem hiding this comment.
very useful @hoechenberger ! you have my full support !
thanks for tackling this
b9f64a7 to
e9dc642
Compare
|
@hoechenberger is it ready for review now? Several more commits were just pushed while I was mid-review. |
|
e9dc642 fixes a bug that is present in
with e9dc642
|
Oh noes, I'm sorry about this! I rebased and added a few commits … It is ready for review now. Ouch … I owe you a review, I guess 😱 |
Don't sweat it :) But it's the end of my day now, so I'll have to wait until tomorrow morning |
Co-authored-by: Daniel McCloy <dan@mccloy.info>
Co-authored-by: Daniel McCloy <dan@mccloy.info>
…python into epochs-evoked-comment
larsoner
left a comment
There was a problem hiding this comment.
Probably needs latest.inc in the BUG section since this is technically a backward-incompatible change (some names in files / instances and accessibility via read_evokeds will change, but for the better / as a bugfix)
|
@larsoner I've added a changelog entry to the |
|
The choice of |
Co-authored-by: Daniel McCloy <dan@mccloy.info>
I do see what you mean, and I agree that Can we not introduce a new section, |
I don't think of it as historical but rather as functional based on our current working definitions for each section: "Bugs" is where people should currently look for backward-incompatible changes that might break their code without any warning.
My response is the same as last time renaming changelog sections came up in a PR -- let's discuss in a dedicated issue/PR dedicated to that... |
Believe it or not, but I had forgotten about the fact that we'd been there before … ouch. Alright then :) |
|
This should be good to merge now. |
|
thx @hoechenberger ! |


Rework
Epochs._nameandEvoked.comment:*replaced with×inEpochs._name0.50 * auditory/left + 0.50 * visual/left0.50 × auditory/left + 0.50 × visual/leftEvoked.commentin evoked differences created viamne.combine_evoked()now groups conditions via parens:1.500 × 0.50 * auditory/left + 0.50 * visual/left + -1.500 × × 0.51 * auditory/right + 0.49 * visual/right1.500 × (0.50 × auditory/left + 0.50 × visual/left) - 1.500 × (0.51 × auditory/right + 0.49 × visual/right)1and-1:-1.000 × 0.50 * auditory/left + 0.50 * visual/left + 1.000 × 0.51 * auditory/right + 0.49 * visual/right- (0.50 × auditory/left + 0.50 × visual/left) + (0.51 × auditory/right + 0.49 × visual/right)Example using
mne.viz.plot_compare_evokeds():old:

new:

What do you think?
(btw unfortunately
Evoked.commentcannot contain a proper minus character, as saving will fail then)