def extract_rois_signals(preprocessing_folder ='pipeline_2', prefix= 'resampled_wr'):
dataset = load_dynacomp(preprocessing_folder = preprocessing_folder,prefix = prefix)
for idx, func in enumerate([dataset.func1, dataset.func2]):
for i in range(len(dataset.subjects)):
tic = time.clock()
print func[i]
output_path, _ = os.path.split(func[i])
print dataset.subjects[i]
maps_img = dict_to_list(dataset.rois[i])
#add mask, smoothing, filter and detrending
print 'Nifti'
masker = NiftiMapsMasker(maps_img=maps_img,
mask_img = dataset.mask,
low_pass = .1,
high_pass = .01,
smoothing_fwhm =6.,
t_r = 1.05,
detrend = True,
standardize = False,
resampling_target ='data',
memory_level = 0,
verbose=5)
#extract signal to x
print 'masker'
x = masker.fit_transform(func[i])
print x
np.save(os.path.join(PATH_TO_SAVE_DATA,'output' + str(i+1) +'_rois_filter'),x)
print time.clock() - tic
return x
def extract_from_masker(dataset_like,dataset_mask, funci, timer =True):
tic = time.clock()
maps_img = dict_to_list(dataset_like)
#add mask, smoothing, filter and detrending
print 'Nifti'
masker = NiftiMapsMasker(maps_img=maps_img,
mask_img = dataset_mask,
low_pass = .1,
high_pass = .01,
smoothing_fwhm =6.,
t_r = 1.05,
detrend = True,
standardize = False,
resampling_target ='data',
memory_level = 0,
verbose=5)
#extract signal to x
print 'masker'
x = masker.fit_transform(funci)
if timer:
print time.clock() - tic
return x
def extract_one_signal(dataset):
for idx, func in enumerate([dataset.func1, dataset.func2]):
for i in range(len(dataset.subjects)):
tic = time.clock()
#maps_img = dict_to_list(func)
#add mask, smoothing, filter and detrending
maps_img = dict_to_list(dataset.rois[i])
masker = NiftiMapsMasker(maps_img=maps_img,
mask_img = dataset.mask,
low_pass = .1,
high_pass = .01,
smoothing_fwhm =6.,
t_r = 1.05,
detrend = True,
standardize = False,
resampling_target ='data',
memory_level = 0,
verbose=5)
#extract signal to x
x = masker.fit_transform(func[i])
print "loading time : "+ str(time.clock() - tic)
return x,maps_img
