def _get_fit(self, errors=None):
'''convenience'''
errors = errors or [dict(axis='y', err_val=1.0)]
_fit = XYFit(xy_data=self._ref_xy_data,
model_function=simple_xy_model,
cost_function=XYCostFunction_Chi2(
axes_to_use='xy', errors_to_use='covariance'),
minimizer=self.MINIMIZER)
for _err in errors:
_fit.add_error(**_err)
return _fit
def _get_fit(self, model_function=None, cost_function=None, error=None):
'''convenience'''
model_function = model_function or simple_xy_model
# TODO: fix default
cost_function = cost_function or XYCostFunction_Chi2(
axes_to_use='xy', errors_to_use='covariance')
error = error or 1.0
_fit = XYFit(xy_data=self._ref_xy_data,
model_function=model_function,
cost_function=cost_function,
minimizer=self.MINIMIZER)
_fit.add_error(axis='y', err_val=error)
return _fit
def _get_fit(self, errors=None):
'''convenience'''
_fit = XYFit(xy_data=self._ref_xy_data,
model_function=simple_xy_model,
cost_function=XYCostFunction_Chi2(
axes_to_use='xy', errors_to_use='covariance'),
minimizer=self.MINIMIZER)
if errors is None:
_fit.add_matrix_error(axis='y',
err_matrix=np.eye(self._n_points),
matrix_type='cov')
else:
for _err in errors:
if 'err_matrix' in _err:
_fit.add_matrix_error(**_err)
else:
_fit.add_error(**_err)
return _fit
def _get_fit(self,
model_function=None,
cost_function=None,
errors=None,
dynamic_error_algorithm=None):
'''convenience'''
model_function = model_function or simple_xy_model
# TODO: fix default
cost_function = cost_function or XYCostFunction_Chi2(
axes_to_use='xy', errors_to_use='covariance')
errors = errors or [dict(axis='y', err_val=1.0)]
dynamic_error_algorithm = dynamic_error_algorithm or "nonlinear"
_fit = XYFit(xy_data=self._ref_xy_data,
model_function=model_function,
cost_function=cost_function,
minimizer=self.MINIMIZER,
dynamic_error_algorithm=dynamic_error_algorithm)
for _err in errors:
_fit.add_error(**_err)
return _fit
def setUp(self):
self._n_points = 16
self._default_cost_function = XYCostFunction_Chi2(
axes_to_use='xy', errors_to_use='covariance')
# "jitter" for data smearing
self._y_jitter = np.array([
0.2991, 1.558, -0.714, 0.825, -1.157, 0.462, 0.103, 1.167, 0.783,
-0.688, -1.019, 0.14, -0.11, -0.87, 1.81, 0.35
])
assert len(self._y_jitter) == self._n_points
# reference initial values
self._ref_initial_pars = np.array([1.0, 0.0])
self._ref_x = np.arange(self._n_points)
self._ref_initial_y_model = line_xy_model(self._ref_x,
*self._ref_initial_pars)
self._ref_initial_xy_model = np.array(
[self._ref_x, self._ref_initial_y_model])
# reference matrices/errors
self._ref_y_error_vector = np.ones(self._n_points) * 0.1
self._ref_y_error_matrix = np.diag(self._ref_y_error_vector**2)
self._ref_x_error_vector = np.ones(self._n_points)
self._ref_x_error_matrix = np.diag(self._ref_x_error_vector**2)
def line_xy_model_derivative(x, a=1.0, b=0.0):
return a
_fp = line_xy_model_derivative(self._ref_x, *self._ref_initial_pars)
self._ref_projected_xy_matrix = (
self._ref_y_error_matrix +
self._ref_x_error_matrix * np.outer(_fp, _fp))
self._ref_projected_xy_errors = np.diag(
np.sqrt(self._ref_projected_xy_matrix))
# fit data
self._ref_y_data = self._ref_initial_y_model + self._y_jitter
self._ref_xy_data = np.array([self._ref_x, self._ref_y_data])
# pre-fit cost value
self._ref_initial_cost = self._default_cost_function(
self._ref_y_data,
self._ref_initial_y_model,
np.linalg.inv(self._ref_projected_xy_matrix),
self._ref_initial_pars,
[],
)
# reference fit result values
self._nominal_fit_result_pars = np.array([1.02590618, -0.00967721])
self._nominal_fit_result_y_model = line_xy_model(
self._ref_x, *self._nominal_fit_result_pars)
self._nominal_fit_result_xy_model = np.array(
[self._ref_x, self._nominal_fit_result_y_model])
self._nominal_fit_result_projected_xy_matrix = 1.0624835 * np.eye(
self._n_points)
self._nominal_fit_result_projected_xy_errors = np.diag(
np.sqrt(self._nominal_fit_result_projected_xy_matrix))
self._nominal_fit_result_cost = self._default_cost_function(
self._ref_y_data,
self._nominal_fit_result_y_model,
np.linalg.inv(self._nominal_fit_result_projected_xy_matrix),
self._nominal_fit_result_pars,
[],
)
# helper dict with all reference property values
self._ref_prop_dict = dict(
parameter_values=self._ref_initial_pars,
x_data=self._ref_x,
x_model=self._ref_x,
y_data=self._ref_y_data,
y_model=self._ref_initial_y_model,
data=self._ref_xy_data,
model=self._ref_initial_xy_model,
did_fit=False,
x_data_cov_mat=self._ref_x_error_matrix,
y_data_cov_mat=self._ref_y_error_matrix,
x_data_cov_mat_inverse=np.linalg.inv(self._ref_x_error_matrix),
y_data_cov_mat_inverse=np.linalg.inv(self._ref_y_error_matrix),
x_data_cor_mat=np.eye(self._n_points),
y_data_cor_mat=np.eye(self._n_points),
x_data_error=self._ref_x_error_vector,
y_data_error=self._ref_y_error_vector,
cost_function_value=self._ref_initial_cost,
total_cov_mat=self._ref_projected_xy_matrix,
total_error=self._ref_projected_xy_errors,
)