def __init__(self, **kwargs):
super(TMatrixPSD, self).__init__(**kwargs)
if not kwargs.get("suppress_warning", False):
warnings.simplefilter("always")
warnings.warn("TMatrixPSD is deprecated and may be removed in " +
"a future version. Use the PSDIntegrator class and the " + \
"psd_integrator property of the Scatterer class instead.",
DeprecationWarning)
warnings.filters.pop(0)
self.num_points = 500
self.m_func = None
self.axis_ratio_func = None
self.D_max = None
self.geometries = (tmatrix_aux.geom_horiz_back,)
self.psd_integrator = PSDIntegrator()
for attr in self._aliases:
if attr in kwargs:
self.__dict__[self._aliases[attr]] = kwargs[attr]
for attr in self.__class__._psd_attr_list:
if attr in kwargs:
self.__dict__[attr] = kwargs[attr]
self.set_geometry(self.geometries[0])
class TMatrixPSD(Scatterer):
"""T-matrix class to perform computations over PSDs.
This class derives from TMatrix to perform computations on particle
ensembles given by a particle size distribution (PSD).
Most class attributes can be set as in the TMatrix class, and there
are several additional attributes. However, note that unlike with TMatrix,
the results are based on lookup tables that are initialized with
init_scatter_table. This means that after initializing
(init_scatter_table) or loading (load_scatter_table) the lookup tables,
only changes to the PSD will have an effect on the results. To change
other properties of the scattering problem (e.g. the refractive index m),
you must rerun init_scatter_table.
Initialize the class and set the attributes as described below. Call
init_scatter_table to compute the lookup table for scattering values at
different scatterer geometries.
The scattering properties can now be retrieved multiple times without
re-initializing. Use set_geometry to select the geometry you need.
After this, you can call the same functions as with the TMatrix class to
retrieve the scattering/radar parameters.
Attributes:
psd: set to a callable object giving the PSD value for a given
diameter (for example a GammaPSD instance);
default GammaPSD(D0=1.0)
n_psd: the number of points for which to sample the PSD and scattering
properties for; default n_psd=500 should be good for most purposes
psd_m_func: set to a callable object giving the refractive index as a
function of diameter, or None to use the "m" attribute for all
sizes; default None
psd_eps_func: set to a callable object giving the aspect ratio
(horizontal to rotational) as a function of diameter, or None to
use the "eps" attribute for all sizes; default None
D_max: set to the maximum single scatterer size that is desired to be
used (usually the D_max corresponding to the largest PSD you
intend to use)
geoms: tuple containing the scattering geometry tuples that are
initialized (thet0, thet, phi0, phi, alpha, beta);
default horizontal backscatter
"""
_psd_attr_list = set(["num_points", "m_func", "axis_ratio_func", "D_max",
"geometries"])
_aliases = {"n_psd": "num_points",
"psd_eps_func": "axis_ratio_func",
"psd_m_func": "m_func"
}
def __init__(self, **kwargs):
super(TMatrixPSD, self).__init__(**kwargs)
if not kwargs.get("suppress_warning", False):
warnings.simplefilter("always")
warnings.warn("TMatrixPSD is deprecated and may be removed in " +
"a future version. Use the PSDIntegrator class and the " + \
"psd_integrator property of the Scatterer class instead.",
DeprecationWarning)
warnings.filters.pop(0)
self.num_points = 500
self.m_func = None
self.axis_ratio_func = None
self.D_max = None
self.geometries = (tmatrix_aux.geom_horiz_back,)
self.psd_integrator = PSDIntegrator()
for attr in self._aliases:
if attr in kwargs:
self.__dict__[self._aliases[attr]] = kwargs[attr]
for attr in self.__class__._psd_attr_list:
if attr in kwargs:
self.__dict__[attr] = kwargs[attr]
self.set_geometry(self.geometries[0])
def __setattr__(self, name, value):
name = self._aliases.get(name, name)
super(TMatrixPSD, self).__setattr__(name, value)
def __getattr__(self, name):
if name == "_aliases":
raise AttributeError
name = self._aliases.get(name, name)
return super(TMatrixPSD, self).__getattr__(name)
def _copy_attrs(self):
for attr in self._psd_attr_list:
if attr in self.__dict__:
self.psd_integrator.__dict__[attr] = self.__dict__[attr]
def init_scatter_table(self):
self._copy_attrs()
self.psd_integrator.init_scatter_table(self)
def save_scatter_table(self, fn, description=""):
self._copy_attrs()
self.psd_integrator.save_scatter_table(fn, description=description)
def load_scatter_table(self, fn):
self.psd_integrator.load_scatter_table(fn)
