Quoting Dirk den Ouden <[log in to unmask]>:
> Dear all,
>
> I'm setting up preprocessing and modeling batches in SPM8b and working with
> the dependencies is great (!!), but I do have a few questions:
>
> 1) Is there a way to insert a module within an already existing batch
> (series of modules), or moving modules up and down, and if this is not
> currently possible, can that be included in the GUI? For large batches, it's
> 'a bit of a bummer' if you forget to include a module and the only way you
> can fix it is with a lot of replicating and deleting.
Doing this properly within the GUI would require quite a lot of
additional code. One reason why it is only allowed to move modules
down the list and not up again is to avoid circular dependencies.
Another reason is more technical - each item that appears somewhere in
the GUI has its place in an object tree. Moving parts of the tree to
another place would require to copy/update all affected objects (The
objects to be moved and all that follow after the place of insertion).
This would slow down things a lot.
There is a workaround however: you can split your batch into pieces,
save them separately and load them at once in the desired order.
> 2) Is it possible to make the 'split file' module refer back to earlier
> separations, for example by the original run directories? From what I
> understand, it is now only possible to split the files based on their
> number, or did I overlook something?
You can use the "named file selector" module to select the files and
the index vectors for you. Each of your sessions/run directories would
go into a separate file set. Note that "named file selector" does not
filter for specific file types (e.g. 'image', 'mat'). You have to take
care to select the desired files by entering an appropriate regexp
filter. For each file set, there will be an matching selection index
vector which you can select as a dependency in "split file".
Hope this helps,
Volkmar
> Thanks a lot,
> Dirk
>
|