def prepSingle(filename, save=True):
'''
preprocesses a single file by name - used for testing
'''
sub = fs.Subject(filename=filename)
X, y = preprocess(sub)
if save:
fs.write_csv(filename, X, y)
return X, y
def prepAll(force=False):
'''
preprocesses all files stored on properly on the drive. either mesa or shhs.
the amount of time and errors are logged for testing purposes. optionally
re-processing already completed files is possible.
'''
log, clock = get_log('Preprocessing', echo=True), stopwatch()
filenames = fs.getAllSubjectFilenames(preprocessed=False)
# determines already completed files
oldFiles = fs.getAllSubjectFilenames(preprocessed=True)
if not force:
filenames = [fn for fn in filenames if fn not in oldFiles]
log.print('Files already completed: {0}'.format(len(oldFiles)))
log.print('Files remaining: {0}'.format(len(filenames)))
if (len(oldFiles) > 0):
log.printHL()
for fn in oldFiles:
log.print('{0} already completed'.format(fn))
else:
log.print('Files re-preprocessing: {0}'.format(len(oldFiles)))
log.print('Files remaining: {0}'.format(len(filenames)))
log.printHL()
# processes each file with try/catch loop incase of errors in single files.
clock.round()
for i, filename in enumerate(filenames):
try:
subject = fs.Subject(filename=filename)
X, y = preprocess(subject)
fs.write_csv(filename, X, y)
log.print('{0} preprocessed in {1}s'.format(
filename, clock.round()))
except Exception as e:
log.print('{0} Exception: {1}'.format(filename, str(e)))
clock.round()
clock.stop()
def prep_X(edf, anno):
'''
converts two edf and anno filepaths into X
'''
sub = fs.Subject(edfPath=edf, annoPath=anno, ArousalAnno=False)
return preprocess(sub, arousals=False)