Closed choldgraf closed 1 year ago
raw => decoding
would be my ideal pipeline ;)
That could be a big part of it :) I'm just thinking of how much people like that sklearn "what kind of analysis should I do" graph...this could be something like it
this graph is generated by :
https://github.com/mne-tools/mne-python/blob/master/doc/sphinxext/flow_diagram.py
improvements and suggestions welcome
The coolest thing we could do is make some interactive webpage with a dynamic graph with collapsible entries. Maybe D3 has something we could use. Then we could probably put everything in one chart. Otherwise if we try to put decoding, encoding (eventually), time frequency, ERP, statistics, connectivity, etc. all in one static graph, it will probably be too overwhelming. But we could try expanding the existing one a bit to see if that is indeed the case.
We could actually switch to a nice client-side rendered flowchart, which would be nice:
http://marvl.infotech.monash.edu/webcola/examples/gridifiedSmallGroups.html http://flowchart.js.org/ http://live.yworks.com/yfiles-for-html/1.3/demos/databinding/demo.yfiles.binding.interactivegraphsource/index.html http://live.yworks.com/demobrowser/index.html#Incremental-Hierarchic-Layout
Nice for a Sunday hack or documentation session
I wonder if the D3 example can be made interactive? If so that'd be pretty cool.
Nelle and I were thinking of organizing a little documentation hackathon / code sprint sometime in the coming months (not necessarily package specific). That could be a fun project.
In this spirit, it would be great to make a workflow for the coregistration: when you don't have an elekta, it is pretty difficult to get around the headshape/marker/hpi/fiducial files correctly, what they do and which coordinate system they're expected to be.
Is there a "todo" list for the documentation somewhere?
On Fri, Oct 21, 2016 at 6:00 PM Jean-Rémi KING notifications@github.com wrote:
In this spirit, it would be great to make a workflow for the coregistration: when you don't have an elekta, it is pretty difficult to get around the headshape/marker/hpi/fiducial files correctly, what they do and which coordinate system they're expected to be.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/mne-tools/mne-python/issues/3669#issuecomment-255485031, or mute the thread https://github.com/notifications/unsubscribe-auth/ABwSHRqGMjiTYEud0PL5PNVTjzuYO1zfks5q2UQYgaJpZM4KZMZY .
Closing in favor of #6778
I've found this image to be useful in explaining the general objects etc that exist in MNE:
I wonder if something like this exists for the kinds of analyses one can create in MNE as well? E.g., one for AverageTFR or EpochsTFR, one for time decoding, or maybe one for a function I don't even know about. Does anybody else think it'd be useful to either update the above diagram, or create complementary diagrams for modules etc?