inverse
"""
if self.is_orthogonal:
scale = self.range.partition.cell_volume
return scale * self.inverse
else:
# TODO: put adjoint here
return super().adjoint
@property
def inverse(self):
"""Inverse of this operator.
Returns
-------
inverse : `WaveletTransform`
See Also
--------
adjoint
"""
return WaveletTransform(
domain=self.range, nlevels=self.nlevels, wavelet=self.pywt_wavelet,
pad_mode=self.pad_mode, pad_const=self.pad_const, impl=self.impl)
if __name__ == '__main__':
# pylint: disable=wrong-import-position
from odl.util.testutils import run_doctests
run_doctests(skip_if=not PYWT_AVAILABLE)
# at a long edge
x0 = (cen + 2 * dx)[1]
x1 = (cen + 3 * dx)[1]
y1 = cen[1]
z = (cen - dx)[2]
phan[x0:x1, y:y1, z:-z] = 1
return discr_space.element(phan)
if __name__ == '__main__':
# Show the phantoms
import odl
# 1D
discr = odl.uniform_discr(-1, 1, 300)
cuboid(discr).show('cuboid 1d')
# 2D
discr = odl.uniform_discr([-1, -1], [1, 1], [300, 300])
cuboid(discr).show('cuboid 2d')
# 3D
discr = odl.uniform_discr([-1, -1, -1], [1, 1, 1], [300, 300, 300])
cuboid(discr).show('cuboid 3d')
indicate_proj_axis(discr).show('indicate_proj_axis 3d')
# Run also the doctests
# pylint: disable=wrong-import-position
from odl.util.testutils import run_doctests
run_doctests()
raise_from(IndexError('axis index {} out of range for array '
'with {} axes'
''.format(axes[-1], arr_out.ndim)),
err)
if arr_in.shape != tuple(in_shape):
raise ValueError('expected input shape {}, got {}'
''.format(tuple(in_shape), arr_in.shape))
if is_real_dtype(arr_out.dtype):
in_dtype = complex_dtype(arr_out.dtype)
elif halfcomplex:
raise ValueError('cannot combine halfcomplex backward transform '
'with complex output')
else:
in_dtype = arr_out.dtype
if arr_in.dtype != in_dtype:
raise ValueError('expected input dtype {}, got {}'
''.format(dtype_repr(in_dtype),
dtype_repr(arr_in.dtype)))
else: # Shouldn't happen
raise RuntimeError
if __name__ == '__main__':
# pylint: disable=wrong-import-position
from odl.util.testutils import run_doctests
run_doctests(skip_if=not PYFFTW_AVAILABLE)
--------
inverse
"""
if self.is_orthogonal:
scale = self.range.partition.cell_volume
return scale * self.inverse
else:
# TODO: put adjoint here
return super(WaveletTransformInverse, self).adjoint
@property
def inverse(self):
"""Inverse of this operator.
Returns
-------
inverse : `WaveletTransform`
See Also
--------
adjoint
"""
return WaveletTransform(
domain=self.range, wavelet=self.pywt_wavelet, nlevels=self.nlevels,
pad_mode=self.pad_mode, pad_const=self.pad_const, impl=self.impl)
if __name__ == '__main__':
from odl.util.testutils import run_doctests
run_doctests(skip_if=not PYWT_AVAILABLE)
to:
.. math::
\sigma = \frac{0.9}{\tau \|L\|^2}
- If both are given, they are returned as-is without further validation.
"""
if tau is None and sigma is None:
L_norm = L.norm(estimate=True)
tau = sigma = np.sqrt(0.9) / L_norm
return tau, sigma
elif tau is None:
L_norm = L.norm(estimate=True)
tau = 0.9 / (sigma * L_norm**2)
return tau, float(sigma)
elif sigma is None:
L_norm = L.norm(estimate=True)
sigma = 0.9 / (tau * L_norm**2)
return float(tau), sigma
else:
return float(tau), float(sigma)
if __name__ == '__main__':
from odl.util.testutils import run_doctests
run_doctests()
in_shape[axes[-1]] = arr_out.shape[axes[-1]] // 2 + 1
except IndexError as err:
raise IndexError('axis index {} out of range for array '
'with {} axes'
''.format(axes[-1], arr_out.ndim))
if arr_in.shape != tuple(in_shape):
raise ValueError('expected input shape {}, got {}'
''.format(tuple(in_shape), arr_in.shape))
if is_real_dtype(arr_out.dtype):
in_dtype = complex_dtype(arr_out.dtype)
elif halfcomplex:
raise ValueError('cannot combine halfcomplex backward transform '
'with complex output')
else:
in_dtype = arr_out.dtype
if arr_in.dtype != in_dtype:
raise ValueError('expected input dtype {}, got {}'
''.format(dtype_repr(in_dtype),
dtype_repr(arr_in.dtype)))
else: # Shouldn't happen
raise RuntimeError
if __name__ == '__main__':
from odl.util.testutils import run_doctests
run_doctests(skip_if=not PYFFTW_AVAILABLE)
# Reshape the resulting stack to the expected output shape
stack_flat_xtra = torch.stack(results)
return stack_flat_xtra.view(extra_shape + op_out_shape)
def __repr__(self):
"""Return ``repr(self)``."""
op_name = self.op_func.operator.__class__.__name__
op_dom_shape = self.op_func.operator.domain.shape
if len(op_dom_shape) == 1:
op_dom_shape = op_dom_shape[0]
op_ran_shape = self.op_func.operator.range.shape
if len(op_ran_shape) == 1:
op_ran_shape = op_ran_shape[0]
return '{}({}) ({} -> {})'.format(self.__class__.__name__, op_name,
op_dom_shape, op_ran_shape)
if __name__ == '__main__':
from odl.util.testutils import run_doctests
import odl
from torch import autograd, nn
run_doctests(extraglobs={
'np': np,
'odl': odl,
'torch': torch,
'nn': nn,
'autograd': autograd
})