You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Might be a bug, or might just be the documentation that could benefit from some extra clarification (if not a bug), or both. In terms of documentation, I think what would at least be useful to specify a bit more is:
How is split_data output ordered? Is there any sorting that can be relied upon? (doesn't seem like it) Is there any way to obtain some sorting nonetheless? (sounds like a sensible use case)
What is "largest" exactly referring to in the documentation of -number?
In terms of improving functionality (even if it's currently not a bug), I'd argue that -weighted does have a potentially meaningful way of applying to split_data, which would effectively be to sort the output. But if that would be offered, I'd recommend to document it very well so users do discover it when they require it. Out of the box, they might still all too easily assume the output of split_data would be sorted by the splitted data's magnitude itself. That itself is something you could discuss or consider even (e.g. as default behaviour?).
Well, all up to you; does look like there's some useful things to consider either way.
The text was updated successfully, but these errors were encountered:
All relevant info and musings here: https://community.mrtrix.org/t/fibre-orientation-wrt-b0/2850/18
Might be a bug, or might just be the documentation that could benefit from some extra clarification (if not a bug), or both. In terms of documentation, I think what would at least be useful to specify a bit more is:
How is
split_data
output ordered? Is there any sorting that can be relied upon? (doesn't seem like it) Is there any way to obtain some sorting nonetheless? (sounds like a sensible use case)What is "largest" exactly referring to in the documentation of
-number
?In terms of improving functionality (even if it's currently not a bug), I'd argue that
-weighted
does have a potentially meaningful way of applying tosplit_data
, which would effectively be to sort the output. But if that would be offered, I'd recommend to document it very well so users do discover it when they require it. Out of the box, they might still all too easily assume the output ofsplit_data
would be sorted by the splitted data's magnitude itself. That itself is something you could discuss or consider even (e.g. as default behaviour?).Well, all up to you; does look like there's some useful things to consider either way.
The text was updated successfully, but these errors were encountered: