piglot.utils package

Submodules

piglot.utils.assorted module

Assorted utilities.

change_cwd(path: str)

Context manager to temporarily change the current working directory.

Adapted from https://stackoverflow.com/a/75049063

Parameters

pathstr

New working directory.

filter_close_points(data: ndarray, tol: float) → ndarray

Remove points from an array that are too close to each other.

Parameters

datanp.ndarray

Data array.

tolfloat

Tolerance to use during removal.

Returns

np.ndarray

Reduced array.

pretty_time(elapsed_sec: float) → str

Return a human-readable representation of a given elapsed time

Parameters

elapsed_secfloat

Elapsed time, in seconds

Returns

str

Pretty elapsed time string

read_custom_module(config: Dict[str, Any], cls: Type[T]) → Type[T]

Read a custom module from a configuration spec.

Parameters

configDict[str, Any]

Configuration of the custom module.

clsType

Base class of the module to load.

Returns

Type

Custom module type read from the script.

reverse_pretty_time(time_str: str) → float

Return an elapsed time from its human-readable representation

Parameters

time_strstr

Pretty elapsed time string

Returns

str

Elapsed time, in seconds

stats_interp_to_common_grid(responses: List[Tuple[ndarray, ndarray]]) → Dict[str, ndarray]

Interpolate a set of response to a common grid and compute several statistics.

Parameters

responsesList[Tuple[np.ndarray, np.ndarray]]

List of responses to interpolate.

Returns

Dict[str, np.ndarray]

Results for the interpolated responses.

piglot.utils.responses module

Module for reducing the number of points in a response function

class ResamplingLoss(ref_x: ndarray, ref_y: ndarray, n_points: int)

Bases: object

Loss for resampling a response function

errors_interps(x_new: ndarray, y_new: ndarray, x_ref: ndarray, y_ref: ndarray) → ndarray

Compute the error associated with removing each point from the grid

Parameters

x_newnp.ndarray

New time grid

y_newnp.ndarray

Values on the new grid

x_refnp.ndarray

Old time grid

y_refnp.ndarray

Values on the old grid

Returns

np.ndarray

Error associated with removing each point

interpolate_response(x_resp: ndarray, y_resp: ndarray, x_grid: ndarray) → ndarray

Interpolate a response function.

Parameters

x_respnp.ndarray

Original time grid.

y_respnp.ndarray

Values in the original grid.

x_gridnp.ndarray

New time grid.

Returns

np.ndarray

Values on the new grid.

reduce_response(x_old: ndarray, y_old: ndarray, tol: float) → Tuple[int, float, Tuple[ndarray, ndarray]]

Reduce the number of points in a response function

Parameters

x_oldnp.ndarray

Original time grid

y_oldnp.ndarray

Values in the original grid

tolfloat

Maximum acceptable error

Returns

Tuple[int, float, Tuple[np.ndarray, np.ndarray]]

Number of points, error, and new grid

piglot.utils.solver_utils module

Utilities for the solver module.

class VerbosityManager(verbosity: str | None, output_dir: str)

Bases: object

Class to manage output streams based on verbosity levels.

AVAILABLE_VERBOSITIES = ['none', 'file', 'error', 'all']

DEFAULT_VERBOSITY = 'none'

flush() → None

Flush the solver outputs, if needed.

prepare() → None

Set up the verbosity level for the solver.

extract_parameters(input_file: str) → ParameterSet

Extract a ParameterSet from an input file.

Parameters

input_filestr

Input file path.

Returns

ParameterSet

Set of parameters discovered in the input file.

Raises

RuntimeError

When a repeated pattern is found.

RuntimeError

When a pattern is referenced but never defined.

find_keyword(file: str, keyword: str) → str

Finds the first line where a keyword is defined.

Parameters

filestr

Path for the input file.

keywordstr

Keyword to locate.

Returns

str

Line containing the keyword.

Raises

RuntimeError

If the keyword is not found.

get_case_name(input_file: str) → str

Extracts the name of a given case.

Parameters

input_filestr

Path for the input file.

Returns

str

Name of the case.

has_keyword(input_file: str, keyword: str) → bool

Checks whether an input file contains a given keyword.

Parameters

input_filestr

Path for the input file.

keywordstr

Keyword to locate.

Returns

bool

Whether the input file contains the keyword or not.

has_parameter(input_file: str, parameter: str) → bool

Checks whether an input file contains a given parameter.

Parameters

input_filestr

Path for the input file.

parameterstr

parameter to locate.

Returns

bool

Whether the input file contains the parameter or not.

load_module_from_file(filename: str, attribute) → object

Loads a module from a given file.

Parameters

filenamestr

Path for the file.

attributestr

Attribute to load from the module.

Returns

object

Module loaded from the file.

piglot.utils.surrogate module

Module for surrogate models.

get_model(x_data: ndarray, y_data: ndarray, var_data: ndarray | None = None, noisy: bool = False) → SingleTaskGP

Get a GP regression model for the current data.

Parameters

x_datanp.ndarray

Input features.

y_datanp.ndarray

Output outcomes.

var_datanp.ndarray, optional

Observation variance, by default None

noisybool, optional

Whether to use a noise-infering GPs or fixed noise ones, by default False

Returns

SingleTaskGP

GP regression model.

optmise_posterior_mean(model: SingleTaskGP, bounds: ndarray) → ndarray

Optimise the posterior mean of the GP model.

Parameters

modelSingleTaskGP

GP model.

boundsnp.ndarray

Bounds for the optimisation problem.

Returns

np.ndarray

Optimal point.

piglot.utils.yaml_parser module

Module for parsing the YAML configuration file.

class UniqueKeyLoader(stream)

Bases: SafeLoader

YAML loader that checks for duplicate keys in mappings.

Adapted from https://gist.github.com/pypt/94d747fe5180851196eb.

construct_mapping(node, deep=False)

parse_config_file(config_file: str) → Dict[str, Any]

Parses the YAML configuration file.

Parameters

config_filestr

Path to the configuration file.

Returns

Dict[str, Any]

Dictionary with the YAML data.

Raises

RuntimeError

When the YAML parsing fails.

Module contents

Module with various utility functions and classes.