DataList¶

class DataList¶

Bases: object

A class to handle input-output pairs of a black-box function.

Methods

`__init__`	Initialize the data list class.
`append`	Append an input-output pair.
`copy`	Deepcopy an instantce of the class.
`is_unique`
`save`	Save the data if the filepath is set either in the `DataList.__init__` or with the `DataList.set_output_path` method.
`set_output_path`	Set a output file path.
`to_df`	Convert and return the data to `pandas.DataFrame`.
`to_solution_dict`	Convert to the solution dict.
`to_structured_solution`	Convert the i-th data (input) to `StructuredSolution`.
`to_structured_solution_list`	Convert the data (all input vectors) to a list of `StructuredSolution`.

Attributes

`abs_y_max`	The absolute maximum value in the output values in the current dataset.
`values`	Return input vectors and output values.
`variable_names`	Names of the variables.
`x`	The list of the input vectors.
`y`	The list of the output values.

__getitem__(i: int) → tuple[list[bool | int | float], int | float]¶

Return a tuple of the input vector and output value.

Parameters:: i (int) – Index.
Returns:: A tuple of the input vector and output value.
Return type:: tuple[list[bool | int | float], int | float]

Initialize the data list class.

Parameters:

x (list[list[bool | int | float]] | None, optional) – A list of the input vectors. Defaults to None.
y (list[int | float] | None, optional) – A list of the output values. Defaults to None.
variable_names (list[str] | None, optional) – A list of the variable names. Defaults to None.
filepath (str | None, optional) – A filepath to save the data with save() method. Defaults to None.

Raises:

ValueError – If only one of (x, y) is specified.
ValueError – If x and y do not have the same length.
ValueError – If at least one of the data (x, y) or variable_names is not specified.

__iter__() → Iterator[tuple]¶

A tuple of the input vector and output value.

Yields:: tuple[list[bool | int | float], int | float] – A tuple of the input vector and output value.

__len__() → int¶

Length of the data list.

Returns:: The length.
Return type:: int

append(value: tuple[list[bool | int | float], int | float]) → None¶

Append an input-output pair.

Parameters:: value (tuple[list[bool | int | float], int | float]) – A tuple of input and output.

copy() → DataList¶

Deepcopy an instantce of the class.

Returns:: The copied data.
Return type:: DataList

is_unique(solution_dict: FlatSolutionDict) → bool¶

save() → None¶: Save the data if the filepath is set either in the DataList.__init__ or with the DataList.set_output_path method.

set_output_path(filepath: str | None) → None¶

Set a output file path.

Parameters:: filepath (str | None) – A file path.

to_df() → DataFrame¶

Convert and return the data to pandas.DataFrame.

Returns:: The converted data.
Return type:: pandas.DataFrame

to_solution_dict(i: int) → FlatSolutionDict¶

Convert to the solution dict. This can be used for multi-objective optimization where there is no one specific Variables assigned for the multiple objectives.

Parameters:: i (int) – Index of the sample.
Returns:: The resulting solution dict.
Return type:: FlatSolutionDict

to_structured_solution(variables: Variables, i: int) → StructuredSolution¶

Convert the i-th data (input) to StructuredSolution.

Parameters:

variables (Variables) – A Variables instance that is relevant to the black-box objective function class that this DataList is for.
i (int) – Index of the sample.

Returns:

A converted input vector.

Return type:

StructuredSolution

to_structured_solution_list(variables: Variables) → list[StructuredSolution]¶

Convert the data (all input vectors) to a list of StructuredSolution.

Parameters:: variables (Variables) – A Variables instance that is relevant to the black-box objective function class that this DataList is for.
Returns:: A list of converted input vectors.
Return type:: list[StructuredSolution]

__dict__ = mappingproxy({'__module__': 'amplify_bbopt.data_list', '__doc__': 'A class to handle input-output pairs of a black-box function.', '__init__': <function DataList.__init__>, 'variable_names': <property object>, 'append': <function DataList.append>, '__len__': <function DataList.__len__>, '__getitem__': <function DataList.__getitem__>, '__iter__': <function DataList.__iter__>, 'to_df': <function DataList.to_df>, 'to_solution_dict': <function DataList.to_solution_dict>, 'to_structured_solution': <function DataList.to_structured_solution>, 'to_structured_solution_list': <function DataList.to_structured_solution_list>, 'save': <function DataList.save>, 'set_output_path': <function DataList.set_output_path>, 'is_unique': <function DataList.is_unique>, 'values': <property object>, 'abs_y_max': <property object>, 'x': <property object>, 'y': <property object>, 'copy': <function DataList.copy>, '__dict__': <attribute '__dict__' of 'DataList' objects>, '__weakref__': <attribute '__weakref__' of 'DataList' objects>, '__annotations__': {'_x': 'list[list[bool | int | float]]', '_y': 'list[int | float]'}})¶

__weakref__¶: list of weak references to the object (if defined)

property abs_y_max: float¶: The absolute maximum value in the output values in the current dataset.

property values: tuple[list[list[bool | int | float]], list[int | float]]¶

Return input vectors and output values.

Returns:: A tuple of the lists of the input vectors and the corresponding output values.
Return type:: tuple[list[list[bool | int | float]], list[int | float]]

property variable_names: list[str] | None¶: Names of the variables.

property x: list[list[bool | int | float]]¶: The list of the input vectors.

property y: list[int | float]¶: The list of the output values. None elements are replaced with 0.5 * sys.float_info.max.