Class _Program

_program.py:64–766

A program-like representation of the evolved program. This is the underlying data-structure used by the public classes in the :mod:`gplearn.genetic` module. It should not be used directly by the user. Parameters ---------- function_set : list A list of valid functions t

class _Program(object):

    """A program-like representation of the evolved program.

    This is the underlying data-structure used by the public classes in the
    :mod:`gplearn.genetic` module. It should not be used directly by the user.

    Parameters
    ----------
    function_set : list
        A list of valid functions to use in the program. 

    arities : dict
        A dictionary of the form `{arity: [functions]}`. The arity is the
        number of arguments that the function takes, the functions must match
        those in the `function_set` parameter.
        
    paras : dict
        A dictionary of the form `{para: [functions]}`. The para is the type 
        of arguments that the function takes, the functions must match 
        those in the `function_set` parameter.

    init_depth : tuple of two ints
        The range of tree depths for the initial population of naive formulas.
        Individual trees will randomly choose a maximum depth from this range.
        When combined with `init_method='half and half'` this yields the well-
        known 'ramped half and half' initialization method.

    init_method : str
        - 'grow' : Nodes are chosen at random from both functions and
          terminals, allowing for smaller trees than `init_depth` allows. Tends
          to grow asymmetrical trees.
        - 'full' : Functions are chosen until the `init_depth` is reached, and
          then terminals are selected. Tends to grow 'bushy' trees.
        - 'half and half' : Trees are grown through a 50/50 mix of 'full' and
          'grow', making for a mix of tree shapes in the initial population.

    n_features : int
        The number of features in `X`.

    const_range : tuple of two floats
        The range of constants to include in the formulas.

    metric : _Fitness object
        The raw fitness metric.

    p_point_replace : float
        The probability that any given node will be mutated during point
        mutation.

    parsimony_coefficient : float
        This constant penalizes large programs by adjusting their fitness to
        be less favorable for selection. Larger values penalize the program
        more which can control the phenomenon known as 'bloat'. Bloat is when
        evolution is increasing the size of programs without a significant
        increase in fitness, which is costly for computation time and makes for
        a less understandable final result. This parameter may need to be tuned
        over successive runs.