physbo.test_functions.base module

Bases: ABC

Abstract class for test functions.

Test functions are used to evaluate the performance of the optimization algorithms.

Note

Each test function is implemented in the same sense (minimization or maximization) as in the reference it follows, so that f and the docstring can be compared with the literature directly. A subclass declares its sense by the class attribute _is_maximization (False by default, i.e., minimization).

test_maximizer selects the sense of the returned values: with test_maximizer=True (the default) the values returned by __call__ and the reference box always describe a maximization problem (as PHYSBO maximizes objectives), and with test_maximizer=False they always describe a minimization problem. The sign is flipped only when the declared sense and the requested sense differ.

Parameters:

nobj (int) – Number of objectives.
dim (int) – Number of dimensions.
min_X (np.ndarray | list[float] | float) – Minimum value of search space for each dimension.
max_X (np.ndarray | list[float] | float) – Maximum value of search space for each dimension.
test_maximizer (bool, default=True) – If True, the returned values describe a maximization problem (for testing a maximization problem solver such as PHYSBO). If False, they describe a minimization problem.
name (str | None, default=None) – Name of the test function. If None, the class name is used.

constraint(x: ndarray) → ndarray[source]

Evaluate the constraint function at the given point.

Parameters:: x (np.ndarray) – The point at which to evaluate the constraint function. x is a numpy array of shape (n, d), where n is the number of points and d is the dimension of the input space.
Returns:: The boolean values indicating whether the point is valid or not. The output value is a numpy array of shape (n,), where n is the number of points.
Return type:: np.ndarray

property dim: int

Get the number of dimensions of the test function.

Returns:: The number of dimensions of the test function d.
Return type:: int

abstractmethod f(x: ndarray) → ndarray[source]

Evaluate the test function at the given point.

f is written in the sense of the reference (see _is_maximization); the conversion to the requested sense is done by __call__.

Parameters:: x (np.ndarray) – The point at which to evaluate the test function. x is a numpy array of shape (n, d), where n is the number of points and d is the dimension of the input space.
Returns:: f – The value of the test function at the given point. The output value is a numpy array of shape (n, k), where k is the number of objectives.
Return type:: np.ndarray

property is_maximization: bool

Whether the test function is defined as a maximization problem in its reference.

This describes how f is written, not the sense of the returned values (which is selected by test_maximizer).

Returns:: True if the original problem is a maximization problem.
Return type:: bool

make_grid(num_X: int | list[int] | ndarray) → ndarray[source]

Make a grid of points in the search space.

Parameters:: num_X (int | list[int] | np.ndarray) – Number of points in each dimension.
Returns:: The grid of points in the search space. The output is a numpy array of shape (N, d), where N is the number of points and d is the dimension of the search space.
Return type:: np.ndarray

property max_X: ndarray

Get the maximum values of the search space of the test function.

Returns:: The maximum value of the test function for each dimension.
Return type:: np.ndarray

property min_X: ndarray

Get the minimum values of the search space of the test function.

Returns:: The minimum value of the test function for each dimension.
Return type:: np.ndarray

property name: str

Get the name of the test function.

Returns:: The name of the test function.
Return type:: str

property nobj: int

Get the number of objectives of the test function.

Returns:: The number of objectives of the test function k.
Return type:: int

set_name(name: str)[source]

Set the name of the test function.

Parameters:: name (str) – The name of the test function.

property test_maximizer: bool

Whether the returned values describe a maximization problem.

Returns:: True if __call__ returns values of a maximization problem.
Return type:: bool