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
I believe the BIDSLayout initialization might take less time on large datasets if the ignore patterns included a regular expression to select any participant folders except the ones requested by the user.
ignore_patterns= [
'code',
'stimuli',
'sourcedata',
'models',
re.compile(r'^\.'),
re.compile(
r'sub-[a-zA-Z0-9]+(/ses-[a-zA-Z0-9]+)?/(beh|dwi|eeg|ieeg|meg|perf)'
),
]
ifcls.participant_label:
# Ignore any subjects who aren't the requested ones.ignore_patterns.append(re.compile(r'sub-(?!'+'|'.join(cls.participant_label) +r')\w+'))
_indexer=BIDSLayoutIndexer(
validate=False,
ignore=ignore_patterns
)
Do you have any interest in helping implement the feature?
Closes#3235.
## Changes proposed in this pull request
- Incorporate a regular expression in the `ignore` list of the
`BIDSLayoutIndexer` that flags subjects who are _not_ in
`Config.participant_label` (when provided).
---------
Co-authored-by: Chris Markiewicz <effigies@gmail.com>
What would you like to see added in fMRIPrep?
I believe the BIDSLayout initialization might take less time on large datasets if the
ignore
patterns included a regular expression to select any participant folders except the ones requested by the user.Do you have any interest in helping implement the feature?
Yes
Additional information / screenshots
This stems from a similar request in QSIPrep (PennLINC/qsiprep#513).
The text was updated successfully, but these errors were encountered: