Result¶

class glotaran.analysis.result.Result(scheme: Scheme, data: dict[str, xr.Dataset], optimized_parameters: ParameterGroup, additional_penalty: np.ndarray | None, least_squares_result: OptimizeResult, free_parameter_labels: list[str], termination_reason: str)[source]¶

Bases: object

The result of a global analysis

Parameters

scheme (Scheme) – An analysis scheme
data (Dict[str, xr.Dataset]) – A dictionary containing all datasets with their labels as keys.
optimized_parameters (ParameterGroup) – The optimized parameters, organized in a ParameterGroup
additional_penalty (Union[np.ndarray, None]) – A vector with the value for each additional penalty, or None
least_squares_result (OptimizeResult) – See scipy.optimize.OptimizeResult() scipy.optimize.least_squares()
free_parameter_labels (List[str]) – The text labels of the free parameters that were optimized
termination_reason (str) – The reason (message when) the optimizer terminated

Attributes Summary

`additional_penalty`	The additional penalty vector.
`chi_square`	The chi-square of the optimization.
`covariance_matrix`	Covariance matrix.
`data`	The resulting data as a dictionary of xarray.Dataset.
`degrees_of_freedom`	Degrees of freedom in optimization $N - N_{vars}$ .
`free_parameter_labels`	List of labels of the free parameters used in optimization.
`initial_parameters`	The initital parameters.
`jacobian`	Modified Jacobian matrix at the solution See also: `scipy.optimize.least_squares()`
`model`	The model for analysis.
`nnls`	If True non-negative least squares optimization is used instead of the default variable projection.
`number_of_data_points`	Number of data points $N$ .
`number_of_function_evaluations`	The number of function evaluations.
`number_of_jacobian_evaluations`	The number of jacobian evaluations.
`number_of_variables`	Number of variables in optimization $N_{vars}$
`optimized_parameters`	The optimized parameters.
`reduced_chi_square`	The reduced chi-square of the optimization.
`root_mean_square_error`	The root mean square error the optimization.
`scheme`	The scheme for analysis.
`success`	Indicates if the optimization was successful.
`termination_reason`	The reason of the termination of the process.

Methods Summary

`get_dataset`	Returns the result dataset for the given dataset label.
`get_scheme`	Return a new scheme from the Result object with optimized parameters.
`markdown`	Formats the model as a markdown text.
`save`	Saves the result to given folder.

Methods Documentation

property additional_penalty¶: The additional penalty vector.

property chi_square¶

The chi-square of the optimization.

$\chi^2 = \sum_i^N [{Residual}_i]^2$ .

property covariance_matrix¶

Covariance matrix.

The rows and columns are corresponding to free_parameter_labels.

property data¶

The resulting data as a dictionary of xarray.Dataset.

Notes

The actual content of the data depends on the actual model and can be found in the documentation for the model.

property degrees_of_freedom¶: Degrees of freedom in optimization $N - N_{vars}$ .

property free_parameter_labels¶: List of labels of the free parameters used in optimization.

get_dataset(dataset_label: str) → xarray.core.dataset.Dataset[source]¶

Returns the result dataset for the given dataset label.

Parameters: dataset_label – The label of the dataset.

get_scheme() → glotaran.analysis.scheme.Scheme [source]¶

Return a new scheme from the Result object with optimized parameters.

Returns: A new scheme with the parameters set to the optimized values. For the dataset weights the (precomputed) weights from the original scheme are used.
Return type: Scheme

property initial_parameters¶: The initital parameters.

property jacobian¶

Modified Jacobian matrix at the solution See also: scipy.optimize.least_squares()

Returns: Numpy array
Return type: np.ndarray

markdown(with_model=True) → str [source]¶

Formats the model as a markdown text.

Parameters: with_model – If True, the model will be printed with initial and optimized parameters filled in.

property model¶: The model for analysis.

property nnls¶: If True non-negative least squares optimization is used instead of the default variable projection.

property number_of_data_points¶: Number of data points $N$ .

property number_of_function_evaluations¶: The number of function evaluations.

property number_of_jacobian_evaluations¶: The number of jacobian evaluations.

property number_of_variables¶: Number of variables in optimization $N_{vars}$

property optimized_parameters¶: The optimized parameters.

property reduced_chi_square¶

The reduced chi-square of the optimization.

$\chi^2_{red}= {\chi^2} / {(N - N_{vars})}$ .

property root_mean_square_error¶

The root mean square error the optimization.

$rms = \sqrt{\chi^2_{red}}$

save(path: str) → list[str][source]¶

Saves the result to given folder.

Returns a list with paths of all saved items.

The following files are saved:

result.md: The result with the model formatted as markdown text.
optimized_parameters.csv: The optimized parameter as csv file.
{dataset_label}.nc: The result data for each dataset as NetCDF file.

Parameters: path – The path to the folder in which to save the result.

property scheme¶: The scheme for analysis.

property success¶: Indicates if the optimization was successful.

property termination_reason¶: The reason of the termination of the process.