def __init__(self):
""" Initialize ExtractReferenceB0Image class
"""
# Inheritance
super(ExtractReferenceB0Image, self).__init__()
# Inputs
self.add_trait(
"dw_image",
File(_Undefined(),
optional=False,
output=False,
exists=True,
desc="an existing diffusion weighted image"))
self.add_trait(
"bvals",
File(_Undefined(),
optional=False,
output=False,
exists=True,
desc="the the diffusion b-values"))
self.add_trait(
"specified_index_of_ref_image",
Int(_Undefined(),
optional=True,
output=False,
desc="index of the reference b=0 volume"))
# Outputs
self.add_trait(
"index_of_ref_image",
Int(_Undefined(),
output=True,
desc="index of the reference b=0 volume"))
def __init__(self):
""" Initialize ExtractReferenceB0Image class
"""
# Inheritance
super(ExtractReferenceB0Image, self).__init__()
# Inputs
self.add_trait("dw_image", File(
_Undefined(),
optional=False,
output=False,
exists=True,
desc="an existing diffusion weighted image"))
self.add_trait("bvals", File(
_Undefined(),
optional=False,
output=False,
exists=True,
desc="the the diffusion b-values"))
self.add_trait("specified_index_of_ref_image", Int(
_Undefined(),
optional=True,
output=False,
desc="index of the reference b=0 volume"))
# Outputs
self.add_trait("index_of_ref_image", Int(
_Undefined(),
output=True,
desc="index of the reference b=0 volume"))
def pipeline_definition(self):
""" Diffusion BET pipeline definition
"""
# Create processes
self.add_process("extractor",
"caps.diffusion_preproc.bet.ExtractReferenceB0Image")
self.add_process("spliter",
"nipype.interfaces.fsl.Split",
make_optional=["terminal_output", "dimension"])
self.add_process("picker",
"caps.utils.misc.Select")
self.add_process("bet",
"caps.utils.bet.BET")
# Export Inputs (automatic export of BET parameters)
self.export_parameter("extractor", "dw_image")
self.export_parameter("extractor", "bvals")
self.export_parameter("extractor", "specified_index_of_ref_image")
# Link inputs
self.add_link("dw_image->spliter.in_file")
# Link b0_extractor
self.add_link("extractor.index_of_ref_image->picker.index")
# Link spliter
self.add_link("spliter._out_files->picker.inlist")
# Link picker
self.add_link("picker.outelement->bet.input_file")
# Export outputs
self.export_parameter("extractor", "index_of_ref_image")
self.export_parameter("spliter", "_out_files",
pipeline_parameter="splited_images")
self.export_parameter("picker", "outelement",
pipeline_parameter="ref_image")
# dBET algorithm parameters
self.dw_image = _Undefined()
self.bvals = _Undefined()
self.nodes["spliter"].process.dimension = "t"
def pipeline_definition(self):
""" Diffusion BET pipeline definition
"""
# Create processes
self.add_process("extractor",
"caps.diffusion_preproc.bet.ExtractReferenceB0Image")
self.add_process("spliter",
"nipype.interfaces.fsl.Split",
make_optional=["terminal_output", "dimension"])
self.add_process("picker", "caps.utils.misc.Select")
self.add_process("bet", "caps.utils.bet.BET")
# Export Inputs (automatic export of BET parameters)
self.export_parameter("extractor", "dw_image")
self.export_parameter("extractor", "bvals")
self.export_parameter("extractor", "specified_index_of_ref_image")
# Link inputs
self.add_link("dw_image->spliter.in_file")
# Link b0_extractor
self.add_link("extractor.index_of_ref_image->picker.index")
# Link spliter
self.add_link("spliter._out_files->picker.inlist")
# Link picker
self.add_link("picker.outelement->bet.input_file")
# Export outputs
self.export_parameter("extractor", "index_of_ref_image")
self.export_parameter("spliter",
"_out_files",
pipeline_parameter="splited_images")
self.export_parameter("picker",
"outelement",
pipeline_parameter="ref_image")
# dBET algorithm parameters
self.dw_image = _Undefined()
self.bvals = _Undefined()
self.nodes["spliter"].process.dimension = "t"
def falsetoundefined(arr):
"""
replace none with undefined
"""
from traits.trait_base import _Undefined
ret = []
for val in arr:
if not val:
ret.append(_Undefined())
else:
ret.append(val)
return ret
def firsttrait(obj):
"""
get first element from iterable, or undefined
:param obj: input list
"""
try:
return next(iter(obj))
except TypeError:
pass
except StopIteration:
pass
from traits.trait_base import _Undefined
return _Undefined()
def set_parameter(self, name, value):
""" Method to set a process instance trait value.
For File and Directory traits the None value is replaced by the
special _Undefined trait value.
Parameters
----------
name: str (mandatory)
the trait name we want to modify
value: object (mandatory)
the trait value we want to set
"""
# The None trait value is _Undefined, do the replacement
if value is None:
value = _Undefined()
# Set the new trait value
setattr(self, name, value)
def set_parameter(self, name, value):
""" Method to set a process instance trait value.
For File and Directory traits the None value is replaced by the
special _Undefined trait value.
Parameters
----------
name: str (mandatory)
the trait name we want to modify
value: object (mandatory)
the trait value we want to set
"""
# The None trait value is _Undefined, do the replacement
if value is None:
value = _Undefined()
# Set the new trait value
setattr(self, name, value)
TraitError: The 'vertical_gradient' trait of a BetInputSpec instance must be a float, but a value of <undefined> <class 'nipype.interfaces.traits._Undefined'> was specified.
So... in order to keep the same type but add the missing method, I
monkey patched.
"""
def length(self):
return 0
##########################################################################
# Apply monkeypatch here
_Undefined.__len__ = length
##########################################################################
Undefined = _Undefined()
def isdefined(object):
return not isinstance(object, _Undefined)
def has_metadata(trait, metadata, value=None, recursive=True):
'''
Checks if a given trait has a metadata (and optionally if it is set to particular value)
'''
count = 0
if hasattr(trait, "_metadata") and metadata in list(trait._metadata.keys()) and (trait._metadata[metadata] == value or value is None):
count += 1
if recursive:
if hasattr(trait, 'inner_traits'):
So... in order to keep the same type but add the missing method, I
monkey patched.
"""
def length(self):
return 0
##########################################################################
# Apply monkeypatch here
_Undefined.__len__ = length
##########################################################################
Undefined = _Undefined()
def isdefined(object):
return not isinstance(object, _Undefined)
def has_metadata(trait, metadata, value=None, recursive=True):
'''
Checks if a given trait has a metadata (and optionally if it is set to particular value)
'''
count = 0
if hasattr(trait, "_metadata") and metadata in trait._metadata.keys() and (
trait._metadata[metadata] == value or value == None):
count += 1
if recursive: