pcntoolkit.math_functions.basis_function
========================================

.. py:module:: pcntoolkit.math_functions.basis_function


Classes
-------

.. autoapisummary::

   pcntoolkit.math_functions.basis_function.BasisFunction
   pcntoolkit.math_functions.basis_function.BsplineBasisFunction
   pcntoolkit.math_functions.basis_function.CompositeBasisFunction
   pcntoolkit.math_functions.basis_function.FractionalPolynomialBasisFunction
   pcntoolkit.math_functions.basis_function.LinearBasisFunction
   pcntoolkit.math_functions.basis_function.PolynomialBasisFunction


Functions
---------

.. autoapisummary::

   pcntoolkit.math_functions.basis_function.create_basis_function


Module Contents
---------------

.. py:class:: BasisFunction(basis_column: int = 0, **kwargs)

   Bases: :py:obj:`abc.ABC`


   Helper class that provides a standard way to create an ABC using
   inheritance.


   .. py:method:: fit(X: numpy.ndarray) -> None


   .. py:method:: from_args(name: str, args: dict) -> BasisFunction
      :classmethod:



   .. py:method:: from_dict(my_dict: dict, version: str | None = None) -> BasisFunction
      :classmethod:



   .. py:method:: to_dict() -> dict


   .. py:method:: transform(X: numpy.ndarray) -> numpy.ndarray


   .. py:attribute:: basis_column
      :value: 0



   .. py:attribute:: basis_name
      :type:  str


   .. py:attribute:: compute_max
      :type:  bool


   .. py:attribute:: compute_min
      :type:  bool


   .. py:property:: dimension
      :abstractmethod:



   .. py:attribute:: is_fitted
      :type:  bool


   .. py:attribute:: max
      :type:  float | None


   .. py:attribute:: min
      :type:  float | None


.. py:class:: BsplineBasisFunction(basis_column: int = 0, degree: int = 3, nknots: int = 5, left_expand: float = 0.05, right_expand: float = 0.05, knot_method: str = 'uniform', knots: numpy.ndarray | list = None, **kwargs)

   Bases: :py:obj:`BasisFunction`


   Helper class that provides a standard way to create an ABC using
   inheritance.


   .. py:method:: to_dict() -> dict


   .. py:attribute:: basis_name
      :value: 'bspline'



   .. py:attribute:: degree
      :value: 3



   .. py:property:: dimension


   .. py:attribute:: knot_method
      :value: 'uniform'



   .. py:attribute:: left_expand
      :value: 0.05



   .. py:attribute:: nknots
      :value: 5



   .. py:attribute:: right_expand
      :value: 0.05



.. py:class:: CompositeBasisFunction(parts)

   Bases: :py:obj:`BasisFunction`


   Helper class that provides a standard way to create an ABC using
   inheritance.


   .. py:method:: fit(X)


   .. py:method:: to_dict()


   .. py:method:: transform(X)


   .. py:property:: dimension


   .. py:attribute:: parts


.. py:class:: FractionalPolynomialBasisFunction(basis_column: int = 0, order: int = 3, powers: list | tuple | str | None = 'default', power_set: list | tuple | None = None, eps: float = 1e-08, **kwargs)

   Bases: :py:obj:`BasisFunction`


   Fractional polynomial basis function for modelling smooth nonlinear
   effects.

   The input must be strictly positive (do not standardize the covariates).
   Power convention:
       p = 0      -> log(x)
       p != 0     -> x**p

   Repeated powers:
       [p, p, p]  -> x**p, x**p * log(x), x**p * log(x)**2

   Initialise the fractional polynomial basis function.

   :param basis_column: Column index to transform.
   :type basis_column: :py:class:`int`, *default* ``0``
   :param order: Fractional polynomial order. Must be 1, 2, or 3.
   :type order: :py:class:`int`, *default* ``3``
   :param powers:
   :type powers: :py:class:`list`, :py:class:`tuple`, :py:class:`str`, or :py:obj:`None`, *default* ``"default"``
   :param power_set: Allowed fractional polynomial powers.
   :type power_set: :py:class:`list`, :py:class:`tuple`, or :py:obj:`None`, *default* :py:obj:`None`
   :param eps: Numerical stability constant.
   :type eps: :py:class:`float`, *default* ``1e-8``


   .. py:method:: to_dict() -> dict

      Serialize the basis function configuration.



   .. py:attribute:: AGE_FP_POWER_PRESETS


   .. py:attribute:: DEFAULT_POWER_SET


   .. py:attribute:: basis_name
      :value: 'fractional_polynomial'



   .. py:property:: dimension
      :type: int


      Number of generated basis columns.


   .. py:attribute:: eps


   .. py:attribute:: order
      :value: 3



   .. py:attribute:: power_set


.. py:class:: LinearBasisFunction(basis_column: int = 0, **kwargs)

   Bases: :py:obj:`BasisFunction`


   Helper class that provides a standard way to create an ABC using
   inheritance.


   .. py:attribute:: basis_name
      :value: 'linear'



   .. py:property:: dimension


.. py:class:: PolynomialBasisFunction(basis_column: int = 0, degree: int = 3, **kwargs)

   Bases: :py:obj:`BasisFunction`


   Helper class that provides a standard way to create an ABC using
   inheritance.


   .. py:attribute:: basis_name
      :value: 'poly'



   .. py:attribute:: degree
      :value: 3



   .. py:property:: dimension


.. py:function:: create_basis_function(basis_type: str | dict | None, basis_column: int = 0, **kwargs) -> BasisFunction