Result
- class glotaran.project.result.Result(number_of_function_evaluations: int, success: bool, termination_reason: str, glotaran_version: str, free_parameter_labels: list[str], scheme: Scheme, initial_parameters: Parameters, optimized_parameters: Parameters, parameter_history: ParameterHistory, optimization_history: OptimizationHistory, data: Mapping[str, xr.Dataset], additional_penalty: list[np.ndarray] | None = None, cost: ArrayLike | None = None, chi_square: float | None = None, covariance_matrix: ArrayLike | None = None, degrees_of_freedom: int | None = None, number_of_clps: int | None = None, jacobian: ArrayLike | list | None = None, number_of_residuals: int | None = None, number_of_jacobian_evaluations: int | None = None, number_of_free_parameters: int | None = None, optimality: float | None = None, reduced_chi_square: float | None = None, root_mean_square_error: float | None = None)[source]
Bases:
object
The result of a global analysis.
Attributes Summary
A vector with the value for each additional penalty, or None
The chi-square of the optimization.
The final cost.
Covariance matrix.
Degrees of freedom in optimization .
Modified Jacobian matrix at the solution
Return the model used to fit result.
Number of conditionally linear parameters .
Return the number of values in the residual vector .
Number of free parameters in optimization
The number of jacobian evaluations.
Return the number of free parameters in optimization .
Number of values in the residual vector .
The reduced chi-square of the optimization.
The root mean square error the optimization.
The number of function evaluations.
Indicates if the optimization was successful.
The reason (message when) the optimizer terminated
The glotaran version used to create the result.
List of labels of the free parameters used in optimization.
The parameter history.
The optimization history.
The resulting data as a dictionary of xarray.Dataset.
Methods Summary
Create dataset for clp guidance.
Return a new scheme from the Result object with optimized parameters.
Create a
Result
instance from the specs defined in a file.Format the model as a markdown text.
Recrate a result from the initial parameters.
Save the result to given folder.
Verify a result.
Methods Documentation
- additional_penalty: list[np.ndarray] | None = None
A vector with the value for each additional penalty, or None
- covariance_matrix: ArrayLike | None = None
Covariance matrix.
The rows and columns are corresponding to
free_parameter_labels
.
- create_clp_guide_dataset(clp_label: str, dataset_name: str) Dataset [source]
Create dataset for clp guidance.
- Parameters:
- Returns:
DataArray containing the clp guide, with
clp_label
dimension replaced by the model dimensions first value.- Return type:
xr.Dataset
- Raises:
ValueError – If
dataset_name
is not in result.ValueError – If
clp_labels
is not in result.
Examples
Extracting the clp guide from an optimization result object.
from glotaran.io import save_dataset clp_guide = result.create_clp_guide_dataset("species_1", "dataset_1") save_dataset(clp_guide, "clp_guide__result_dataset_1__species_1.nc")
- data: Mapping[str, xr.Dataset]
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.
- get_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:
- initial_parameters: Parameters
- jacobian: ArrayLike | list | None = None
Modified Jacobian matrix at the solution
See also:
scipy.optimize.least_squares()
- loader(format_name: str | None = None, **kwargs: Any) Result
Create a
Result
instance from the specs defined in a file.- Parameters:
result_path (StrOrPath) – Path containing the result data.
format_name (str | None) – Format the result is in, if not provided and it is a file it will be inferred from the file extension.
**kwargs (Any) – Additional keyword arguments passes to the
load_result
implementation of the project io plugin.
- Returns:
Result
instance created from the saved format.- Return type:
- markdown(with_model: bool = True, *, base_heading_level: int = 1, wrap_model_in_details: bool = False) MarkdownStr [source]
Format the model as a markdown text.
- Parameters:
- Returns:
MarkdownStr – The scheme as markdown string.
- Return type:
- property model: Model
Return the model used to fit result.
- Returns:
The model instance.
- Return type:
- property number_of_data_points: int | None
Return the number of values in the residual vector .
Deprecated since it returned the wrong value.
- Returns:
Number of values in the residual vector .
- Return type:
int | None
- property number_of_parameters: int | None
Return the number of free parameters in optimization .
- Returns:
Number of free parameters in optimization .
- Return type:
int | None
- optimization_history: OptimizationHistory
The optimization history.
- optimized_parameters: Parameters
- parameter_history: ParameterHistory
The parameter history.
- recreate() Result [source]
Recrate a result from the initial parameters.
- Returns:
The recreated result.
- Return type:
- save(path: StrOrPath, saving_options: SavingOptions = SavingOptions(data_filter=None, data_format='nc', parameter_format='csv', report=True)) list[str] [source]
Save the result to given folder.
- Parameters:
path (StrOrPath) – The path to the folder in which to save the result.
saving_options (SavingOptions) – Options for the saved result.
- Returns:
Paths to all the saved files.
- Return type:
- source_path: StrOrPath = 'result.yml'