Commons

py2dmat.Info

This class treats the input parameters. This has the following four instance variables.

  • base : dict[str, Any]

    • Parameters for whole program such as the directory where the output will be written.

  • solver : dict[str, Any]

    • Parameters for Solver

  • algorithm : dict[str, Any]

    • Parameters for Algorithm

  • runner : dict[str, Any]

    • Parameters for Runner

An instance of Info is initialized by passing a dict which has the following four sub dictionaries, base, solver, algorithm, and "runner". Each value will be set to the corresponding field of Info.

  • About base

    • Root directory root_dir

      • The default value is "." (the current directory).

      • Value of root_dir will be converted to an absolute path.

      • The leading ~ will be expanded to the user’s home directory.

      • Specifically, the following code is executed

        p = pathlib.Path(base.get("root_dir", "."))
base["root_dir"] = p.expanduser().absolute()

    • Output directory output_dir

      • The default value is ".", that is, the same to root_dir

      • The leading ~ will be expanded to the user’s home directory.

      • If a relative path is given, its origin is root_dir.

      • Specifically, the following code is executed

        p = pathlib.Path(base.get("work_dir", "."))
p = p.expanduser()
base["work_dir"] = base["root_dir"] / p

py2dmat.Message

When Algorithm tries to invoke Solver, an instance of this class is passed from Algorithm to Solver via Runner.

This has the following three instance variables.

  • x: np.ndarray

    • Coordinates of a point \(x\) to calculate \(f(x)\)

  • step: int

    • The index of parameters

    • For example, the index of steps in exchange and the ID of parameter in mapper.

  • set: int

    • Which lap it is

    • For example, min_search has two laps, the first one is optimization and the second one is recalculation the optimal values for each step.

py2dmat.Runner

Runner connects Algorithm and Solver. The constructor of Runner takes solver: Solver, info: Info, and mapping: Callable[[np.ndarray], np.ndarray].

submit(self, message: py2dmat.Message) -> float method invokes the solver and returns the result. To evaluate fx = f(x), use the following code snippet:

message = py2dmat.Message(x, step, set)
fx = runner.submit(message)

submit internally uses mapping for generating a parameter used in Solver, \(y\), from a parameter searched by Algorithm, \(x\), as y = mapping(x). When mapping is omitted in the constructor (or None is passed), an affine mapping (py2dmat.util.mapping.Affine(A,b)) \(y=Ax+b\) is used as mapping. The elements of A and b are defined in info. See Input file for details how/which components of info Runner uses.