gm.generate_ldd_model_csv(
'subjects_ldd_cut.lst',
work_prefix='tmp_avg_SSD_ldd',
options={
'protocol': [
{'iter':4, 'level':8, 'blur_int': None, 'blur_vel': None },
{'iter':4, 'level':4, 'blur_int': None, 'blur_vel': None },
#{'iter':16, 'level':2, 'blur_int': None, 'blur_vel': None },
#{'iter':4, 'level':2, 'blur_int': None, 'blur_vel': 2 },
#{'iter':4, 'level':2, 'blur_int': None, 'blur_vel': 1 },
],
'parameters': {'smooth_update':2,
'smooth_field':2,
'conf': { 8:200, 4:200, 2:40 },
'LCC':False },
'start_level': 8,
'refine': True,
'cleanup':False,
'debug': True,
'debias': True,
'qc': True,
'incremental': True
},
#regress_model=['data/object_0_4.mnc'],
model='data_ldd/object_0_0.mnc',
mask='data_ldd/mask_0_0.mnc',
)
from scoop import futures, shared
import iplScoopGenerateModel as gm
if __name__ == '__main__':
# setup data for parallel processing
os.environ['ITK_GLOBAL_DEFAULT_NUMBER_OF_THREADS']='1'
gm.generate_ldd_model_csv('subjects.lst',
work_prefix='tmp_ldd_nr',
options={'symmetric':False,
'refine':False,
'protocol': [{'iter':4,'level':4},
{'iter':4,'level':2},
],
'parameters': {'smooth_update':2,
'smooth_field':1,
'conf': { 32:40,16:40,8:40,4:40,2:40 } }
},
model='ref.mnc',
mask='mask.mnc'
)
gm.generate_ldd_model_csv('subjects.lst',
work_prefix='tmp_ldd_sym',
options={
'symmetric':
True,
'refine':
True,
'protocol': [
{
'iter': 4,
'level': 16
},
{
'iter': 4,
'level': 8
},
{
'iter': 4,
'level': 4
},
],
'parameters': {
'smooth_update': 2,
'smooth_field': 2,
'conf': {
32: 20,
16: 20,
8: 20,
4: 20,
2: 20
}
}
},
model='test_data/ellipse_1.mnc',
mask='test_data/mask.mnc')
# setup data for parallel processing
gm.generate_ldd_model_csv('subjects_cut.lst',
work_prefix='tmp_ldd',
options={
'symmetric':
False,
'refine':
True,
'protocol': [
{
'iter': 4,
'level': 8
},
{
'iter': 4,
'level': 4
},
],
'parameters': {
'smooth_update': 2,
'smooth_field': 2,
'conf': {
32: 20,
16: 20,
8: 20,
4: 20,
2: 20,
1: 20
}
}
})