Source code for gpi.types.numpy_GPITYPE
#!/usr/bin/env python
# Copyright (C) 2014 Dignity Health
#
# This program is free software: you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as published by
# the Free Software Foundation, either version 3 of the License, or
# (at your option) any later version.
#
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU Lesser General Public License for more details.
#
# You should have received a copy of the GNU Lesser General Public License
# along with this program. If not, see <http://www.gnu.org/licenses/>.
#
# NO CLINICAL USE. THE SOFTWARE IS NOT INTENDED FOR COMMERCIAL PURPOSES
# AND SHOULD BE USED ONLY FOR NON-COMMERCIAL RESEARCH PURPOSES. THE
# SOFTWARE MAY NOT IN ANY EVENT BE USED FOR ANY CLINICAL OR DIAGNOSTIC
# PURPOSES. YOU ACKNOWLEDGE AND AGREE THAT THE SOFTWARE IS NOT INTENDED FOR
# USE IN ANY HIGH RISK OR STRICT LIABILITY ACTIVITY, INCLUDING BUT NOT
# LIMITED TO LIFE SUPPORT OR EMERGENCY MEDICAL OPERATIONS OR USES. LICENSOR
# MAKES NO WARRANTY AND HAS NO LIABILITY ARISING FROM ANY USE OF THE
# SOFTWARE IN ANY HIGH RISK OR STRICT LIABILITY ACTIVITIES.
# Brief: A GPI type extension for numpy arrays.
import inspect
import numpy as np
# gpi
from gpi import GPIDefaultType, osuper
[docs]class NPYarray(GPIDefaultType):
"""The NPYarray type provides port enforcement for numpy
multidimensional arrays (ndarray). The enforcement parms
are type (ndarray), dtype, ndim, dimension range (drange),
shape, and vec (the len of the last dim). Enforcement
priority for aliased parms goes ndim->drange->shape. Although
shape and vec can overlap, they are enforced independently.
"""
# Have the user pass the defining port traits to be checked.
def __init__(self):
super(NPYarray, self).__init__()
self._type = [np.ndarray, np.memmap] # the class is implicitly this type
self._dtype = [] # a list of types
self._ndim = None
self._drange = None
self._shape = None
self._vec = None # the last dim-length
[docs] def edgeTip(self, data):
if type(data) in self._type:
msg = str(data.shape)
return msg
return osuper(NPYarray, self).edgeTip(data)
[docs] def setDataAttr(self, data):
# set NPY array to read-only
# so downstream mods don't accidentally
# edit it.
if type(data) in self._type:
data.flags.writeable = False
if not data.flags['C_CONTIGUOUS']:
cf = inspect.currentframe().f_back.f_back
try:
pname = '\''+str(cf.f_locals['pnumORtitle'])+'\''
except:
pname = '(can\'t resolve port name)'
try:
fname = '\''+str(cf.f_back.f_locals['self'].__module__)+'\''
except:
fname = '(can\'t reslove file name)'
self.log.warn(__name__+': Output array is not contiguous, forcing contiguity:\n\tFILE: '+fname+', PORT: '+pname)
data = np.ascontiguousarray(data)
return data
else:
return osuper(NPYarray, self).setDataAttr(data)
[docs] def matchesType(self, type_cls):
if self.isFreeType(type_cls):
self.log.info(str(self.__class__)+ "matchesType(): upstream port is free.")
return True
# if this isn't true then these classes
# cannot be compared any further
# the input is the upstream port
if type(type_cls) != type(self):
self.log.info(str(self.__class__)+ "matchesType(): port class cannot be compared.")
return False
if len(self._dtype) and len(type_cls._dtype):
# if at least one type in the list matches, then pass it
typ_passed = False
for typ in type_cls._dtype:
if typ in self._dtype:
typ_passed = True
if not typ_passed:
self.log.info(str(self.__class__)+ "matchesType(): dtype doesn't match.")
return False
if self._ndim is not None:
if type_cls._ndim is not None:
if self._ndim != type_cls._ndim:
self.log.info(str(self.__class__)+ "matchesType(): ndim doesn't match.")
return False
elif self._drange is not None:
if type_cls._drange is not None:
if self._drange != type_cls._drange:
self.log.info(str(self.__class__)+ "matchesType(): drange doesn't match.")
return False
elif self._shape is not None:
if type_cls._shape is not None:
if self._shape != type_cls._shape:
self.log.info(str(self.__class__)+ "matchesType(): shape doesn't match.")
return False
if self._vec is not None:
if type_cls._vec is not None:
if self._vec != type_cls._vec:
self.log.info(str(self.__class__)+ "matchesType(): vec doesn't match.")
return False
return True
[docs] def matchesData(self, data):
if type(data) not in self._type:
return False
if len(self._dtype):
if not data.dtype in self._dtype:
return False
if self._ndim:
if self._ndim != data.ndim:
return False
elif self._drange:
if (data.ndim < self._drange[0]) or (data.ndim > self._drange[1]):
return False
elif self._shape:
if self._shape != data.shape:
return False
if self._vec:
if self._vec != data.shape[-1]:
return False
return True
# setters
[docs] def set_ndim(self, val):
"""int | Set enforcement for the number of dimensions. Requires (int)."""
if type(val) != int:
raise Exception("ERROR: \'ndim\' requires an \'int\'!")
self._ndim = val
[docs] def set_drange(self, val):
"""tuple(int,int) | Requires a dimension range tuple (min, max)."""
if type(val) != tuple:
raise Exception("ERROR: \'drange\' requires an \'tuple\'!")
if len(val) != 2:
raise Exception(
"ERROR: \'drange\' requires an \'tuple\' of len = 2!")
for dim in val:
if type(dim) != int:
raise Exception(
"ERROR: \'drange\' requires a \'tuple\' of \'int\' types!")
self._drange = val
[docs] def set_dtype(self, val):
"""numpy dtype | Requires an NPY type (i.e. float32, complex64, etc...)"""
if type(val) is list:
for t in val:
if type(t) != type:
raise Exception("ERROR: \'dtype\' requires a \'type\'!")
self._dtype = val
return
if type(val) != type:
raise Exception("ERROR: \'dtype\' requires a \'type\'!")
self._dtype.append(val)
[docs] def set_shape(self, val):
"""tuple(int,int,...) | Requires the dimension lengths in an integer tuple."""
if type(val) != tuple:
raise Exception("ERROR: \'shape\' requires a \'tuple\'!")
for dim in val:
if type(dim) != int:
raise Exception(
"ERROR: \'shape\' requires a \'tuple\' of \'int\' types!")
self._shape = val
[docs] def set_vec(self, val):
"""int | Requires an integer representing the length of the last
(most varying) dimension (or shape[-1]).
"""
if type(val) != int:
raise Exception("ERROR: \'vec\' requires an \'int\'!")
if val <= 0:
raise Exception("ERROR: \'vec\' requires an \'int\' > 0!")
self._vec = val