jpt.learning.c45
================

.. py:module:: jpt.learning.c45


Attributes
----------

.. autoapisummary::

   jpt.learning.c45.logger
   jpt.learning.c45.DISCRIMINATIVE
   jpt.learning.c45.GENERATIVE
   jpt.learning.c45._locals


Classes
-------

.. autoapisummary::

   jpt.learning.c45.JPTPartition
   jpt.learning.c45.C45Algorithm


Functions
---------

.. autoapisummary::

   jpt.learning.c45._initialize_worker_process
   jpt.learning.c45.learn_prior
   jpt.learning.c45.c45split


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

.. py:data:: logger

.. py:data:: DISCRIMINATIVE
   :value: 'discriminative'


.. py:data:: GENERATIVE
   :value: 'generative'


.. py:data:: _locals

.. py:function:: _initialize_worker_process()

.. py:class:: JPTPartition(data: Optional[numpy.ndarray], start: int, end: int, node_idx: int, parent_idx: Optional[int], child_idx: Optional[int], path: List[Set or Interval], min_samples_leaf: int, depth: int, min_eval_samples: int = 0)

   Represents a partition of the input data during JPT learning.

   :param data:        the indices for the training samples used to calculate the gain.
   :param start:       the starting index in the data.
   :param end:         the stopping index in the data.
   :param node_idx:    the node of the current iteration
   :param parent_idx:  the parent node of the current iteration, initially ``None``.
   :param child_idx:   the index of the child in the current iteration.
   :param depth:       the depth of the tree in the current recursion level.


   .. py:attribute:: data


   .. py:attribute:: start


   .. py:attribute:: end


   .. py:attribute:: node_idx


   .. py:attribute:: parent_idx


   .. py:attribute:: child_idx


   .. py:attribute:: depth


   .. py:attribute:: min_samples_leaf


   .. py:attribute:: min_eval_samples
      :value: 0



   .. py:attribute:: path


   .. py:property:: n_samples


.. py:function:: learn_prior(variable: jpt.variables.Variable, column: int)

.. py:function:: c45split(partition: JPTPartition, prune_or_split: Callable = None) -> Tuple[Dict[str, Any], JPTPartition, Optional[JPTPartition], Optional[JPTPartition]]

   Creates a node in the decision tree according to the C4.5 algorithm


.. py:class:: C45Algorithm(jpt: jpt.trees.JPT)

   .. py:attribute:: jpt


   .. py:attribute:: lock
      :value: None



   .. py:attribute:: c45queue
      :value: None



   .. py:attribute:: finish
      :value: None



   .. py:attribute:: _progressbar
      :value: None



   .. py:attribute:: _prune_or_split
      :value: None



   .. py:attribute:: queue_length
      :value: 0



   .. py:attribute:: indices
      :value: None



   .. py:attribute:: min_samples_leaf
      :value: None



   .. py:attribute:: min_eval_samples
      :value: 0



   .. py:attribute:: _node_counter
      :value: 0



   .. py:method:: _node_created(args: Tuple) -> None


   .. py:method:: learn(data: pandas.DataFrame = None, keep_samples: bool = False, close_convex_gaps: bool = True, verbose: bool = False, prune_or_split: Optional[Callable] = None, multicore: Optional[int] = None, split_validation_mask: Optional[numpy.ndarray] = None, split_validation_mode: str = 'both') -> None

      Fit the jpt to ``data``.

      :param data:    The training examples (assumed in row-shape)
      :type data:     [[str or float or bool]]; (according to `self.variables`)
      :param rows:    The training examples (assumed in row-shape)
      :type rows:     [[str or float or bool]]; (according to `self.variables`)
      :param columns: The training examples (assumed in column-shape)
      :type columns:  [[str or float or bool]]; (according to `self.variables`)
      :param keep_samples: If true, stores the indices of the original data samples in the leaf nodes. For debugging
                      purposes only. Default is false.
      :param close_convex_gaps:
      :param prune_or_split:
      :param multicore: The number of cores to use for learning. If ``None``, all cores available will be used.
      :param verbose:

      :return: the fitted model



   .. py:method:: postprocess_leaves() -> None

      Postprocess leaves such that the convex hull that is
      postulated from this tree has likelihood > 0 for every
      point inside the hull.