Expand source code

from copy import deepcopy
from typing import List, Optional

import numpy as np

from .data import Data
from .splitcondition import CombinedCondition, SplitCondition


class Split:
    """
    The Split class represents the conditioned data at any point in the decision tree
    It contains the logic for:

     - Maintaining a record of which rows of the covariate matrix are in the split
     - Being able to easily access a `Data` object with the relevant rows
     - Applying `SplitConditions` to further break up the data
    """

    def __init__(self, data: Data, combined_condition: Optional[CombinedCondition]=None):
        self._data = data
        if combined_condition is None:
            combined_condition = CombinedCondition(self._data.X.variables, [])
        self._combined_condition = combined_condition

    @property
    def data(self):
        return self._data

    def combined_condition(self):
        return self._combined_condition

    def condition(self, X: np.ndarray=None) -> np.array:
        if X is None:
            return ~self._data.mask
        else:
            return self.out_of_sample_condition(X)

    def out_of_sample_condition(self, X: np.ndarray) -> np.ndarray:
        return self._combined_condition.condition(X)

    def out_of_sample_conditioner(self) -> CombinedCondition:
        return self._combined_condition

    def __add__(self, other: SplitCondition):

        return Split(self._data + other,
                     self._combined_condition + other)

    def most_recent_split_condition(self) -> Optional[SplitCondition]:
        return self._combined_condition.most_recent_split_condition()

Classes

class Split (data: Data, combined_condition: Optional[CombinedCondition] = None)

The Split class represents the conditioned data at any point in the decision tree It contains the logic for:

Maintaining a record of which rows of the covariate matrix are in the split
Being able to easily access a Data object with the relevant rows
Applying SplitConditions to further break up the data

Expand source code

class Split:
    """
    The Split class represents the conditioned data at any point in the decision tree
    It contains the logic for:

     - Maintaining a record of which rows of the covariate matrix are in the split
     - Being able to easily access a `Data` object with the relevant rows
     - Applying `SplitConditions` to further break up the data
    """

    def __init__(self, data: Data, combined_condition: Optional[CombinedCondition]=None):
        self._data = data
        if combined_condition is None:
            combined_condition = CombinedCondition(self._data.X.variables, [])
        self._combined_condition = combined_condition

    @property
    def data(self):
        return self._data

    def combined_condition(self):
        return self._combined_condition

    def condition(self, X: np.ndarray=None) -> np.array:
        if X is None:
            return ~self._data.mask
        else:
            return self.out_of_sample_condition(X)

    def out_of_sample_condition(self, X: np.ndarray) -> np.ndarray:
        return self._combined_condition.condition(X)

    def out_of_sample_conditioner(self) -> CombinedCondition:
        return self._combined_condition

    def __add__(self, other: SplitCondition):

        return Split(self._data + other,
                     self._combined_condition + other)

    def most_recent_split_condition(self) -> Optional[SplitCondition]:
        return self._combined_condition.most_recent_split_condition()

Instance variables

var data

Expand source code

@property
def data(self):
    return self._data

Methods

def combined_condition(self)

Expand source code

def combined_condition(self):
    return self._combined_condition

def condition(self, X: numpy.ndarray = None) ‑>

Expand source code

def condition(self, X: np.ndarray=None) -> np.array:
    if X is None:
        return ~self._data.mask
    else:
        return self.out_of_sample_condition(X)

def most_recent_split_condition(self) ‑> Optional[SplitCondition]

Expand source code

def most_recent_split_condition(self) -> Optional[SplitCondition]:
    return self._combined_condition.most_recent_split_condition()

def out_of_sample_condition(self, X: numpy.ndarray) ‑> numpy.ndarray

Expand source code

def out_of_sample_condition(self, X: np.ndarray) -> np.ndarray:
    return self._combined_condition.condition(X)

def out_of_sample_conditioner(self) ‑> CombinedCondition

Expand source code

def out_of_sample_conditioner(self) -> CombinedCondition:
    return self._combined_condition