fitbenchmarking.parsing.fitting_problem module¶

Implements the FittingProblem class, this will be the object that inputs are parsed into before being passed to the controllers

class fitbenchmarking.parsing.fitting_problem.FittingProblem(options)¶

Bases: object

Definition of a fitting problem, which will be populated by a parser from a problem definition file.

This defines a fitting problem where, given a set of $n$ data points $(x_i,y_i)$ , associated errors $e_i$ , and a model function $f(x,p)$ , we find the optimal parameters in the least-squares sense by solving:

$\min_p \sum_{i=1}^n \left( \frac{y_i - f(x_i, p)}{e_i} \right)^2$

where $p$ is a vector of length $m$ , and we start from a given intial guess for the optimal parameters.

additional_info = None¶: dict Container for software specific information. This should be avoided if possible.

correct_data()¶: Strip data that overruns the start and end x_range, and approximate errors if not given. Modifications happen on member variables.

data_e = None¶: numpy array The errors

data_x = None¶: numpy array The x-data

data_y = None¶: numpy array The y-data

end_x = None¶: float The end of the range to fit model data over (if different from entire range) (/float/)

equation = None¶: string Equation (function or model) to fit against data

eval_f(params, x=None)¶

Function evaluation method

Parameters:	params (list) – parameter value(s) x (numpy array) – x data values or None, if None this uses self.data_x
Returns:	y data values evaluated from the function of the problem
Return type:	numpy array

eval_r(params, x=None, y=None, e=None)¶

Calculate residuals and weight them if using errors

Parameters:	params (list) – The parameters to calculate residuals for x (numpy array, optional) – x data points, defaults to self.data_x y (numpy array, optional) – y data points, defaults to self.data_y e (numpy array, optional) – error at each data point, defaults to self.data_e
Returns:	The residuals for the datapoints at the given parameters
Return type:	numpy array

eval_r_norm(params, x=None, y=None, e=None)¶

Evaluate the square of the L2 norm of the residuals

Parameters:	params (list) – The parameters to calculate residuals for x (numpy array, optional) – x data points, defaults to self.data_x y (numpy array, optional) – y data points, defaults to self.data_y e (numpy array, optional) – error at each data point, defaults to self.data_e
Returns:	The sum of squares of residuals for the datapoints at the given parameters
Return type:	numpy array

eval_starting_params(param_set)¶

Evaluate the function using the starting parameters.

Parameters:	param_set (int) – The index of the parameter set in starting_params
Returns:	Results from evaluation
Return type:	numpy array

function = None¶: Callable function

get_function_params(params)¶

Return the function definition in a string format for output

Parameters:	params (list) – The parameters to use in the function string
Returns:	Representation of the function example format: ‘b1 * (b2+x) \| b1=-2.0, b2=50.0’
Return type:	string

multifit = None¶: bool Used to check if a problem is using multifit.

name = None¶: string Name (title) of the fitting problem

param_names¶

Utility function to get the parameter names

Returns:	the names of the parameters
Return type:	list of str

sorted_index = None¶: numpy array The index for sorting the data (used in plotting)

start_x = None¶: float The start of the range to fit model data over (if different from entire range)

starting_values = None¶

list of dict Starting values of the fitting parameters

e.g. [{p1_name: p1_val1, p2_name: p2_val1, ...}, {p1_name: p1_val2, ...}, ...]

value_ranges = None¶

dict Smallest and largest values of interest in the data

e.g. {p1_name: [p1_min, p1_max], ...}

verify()¶

Basic check that minimal set of attributes have been set.

Raise FittingProblemError if object is not properly initialised.

fitbenchmarking.parsing.fitting_problem.correct_data(x, y, e, startx, endx, use_errors)¶

Strip data that overruns the start and end x_range, and approximate errors if not given.

Parameters:	x (np.array) – x data y (np.array) – y data e (np.array) – error data startx (float) – minimum x value endx (float) – maximum x value use_errors (bool) – whether errors should be added if not present