def _create_hpgl_nonparam_cdf(cdf_data): cd2 = cdf_data assert isinstance(cdf_data, CdfData) return __checked_create( hpgl_non_parametric_cdf_t, values = cd2.values.ctypes.data_as(C.POINTER(C.c_float)), probs = cd2.probs.ctypes.data_as(C.POINTER(C.c_float)), size = cd2.values.size)
def _create_hpgl_nonparam_cdf(cdf_data):
cd2 = cdf_data
assert isinstance(cdf_data, CdfData)
return __checked_create(
hpgl_non_parametric_cdf_t,
values=cd2.values.ctypes.data_as(C.POINTER(C.c_float)),
probs=cd2.probs.ctypes.data_as(C.POINTER(C.c_float)),
size=cd2.values.size)
def __create_hpgl_ik_params(data, indicator_count, is_lvm, marginal_probs):
ikps = []
assert len(data) == indicator_count
for i in range(indicator_count):
ikd = data[i]
ikp = __checked_create(
_HPGL_IK_PARAMS,
covariance_type=ikd["cov_model"].type,
ranges=(C.c_double * 3)(*ikd["cov_model"].ranges),
angles=(C.c_double * 3)(*ikd["cov_model"].angles),
sill=ikd["cov_model"].sill,
nugget=ikd["cov_model"].nugget,
radiuses=(C.c_int * 3)(*ikd["radiuses"]),
max_neighbours=ikd["max_neighbours"],
marginal_prob=0 if is_lvm else marginal_probs[i])
ikps.append(ikp)
return _c_array(_HPGL_IK_PARAMS, indicator_count, ikps)
def __create_hpgl_ik_params(data, indicator_count, is_lvm, marginal_probs): ikps = [] assert len(data) == indicator_count for i in range(indicator_count): ikd = data[i] ikp = __checked_create( _HPGL_IK_PARAMS, covariance_type = ikd["cov_model"].type, ranges = (C.c_double * 3)(*ikd["cov_model"].ranges), angles = (C.c_double * 3)(*ikd["cov_model"].angles), sill = ikd["cov_model"].sill, nugget = ikd["cov_model"].nugget, radiuses = (C.c_int * 3)(*ikd["radiuses"]), max_neighbours = ikd["max_neighbours"], marginal_prob = 0 if is_lvm else marginal_probs[i]) ikps.append(ikp) return _c_array(_HPGL_IK_PARAMS, indicator_count, ikps)
