lux.pyuid3.uid3.UId3

class lux.pyuid3.uid3.UId3(max_depth=None, node_size_limit=1, grow_confidence_threshold=0, min_impurity_decrease=0)

A decision tree classifier with customizable parameters for controlling tree growth.

Parameters:

param max_depth:

int or None, default=None The maximum depth of the tree. If None, then nodes are expanded until all leaves are pure or until all leaves contain less than node_size_limit samples.

param node_size_limit:

int, default=1 The minimum number of samples required to split a node further. If the number of samples at a node is less than node_size_limit, the node is not split, and it becomes a leaf.

param grow_confidence_threshold:

float, default=0 The minimum confidence level required for a split to occur. Splits with a confidence level below this threshold are not performed. Confidence level is typically defined by impurity measures such as Gini impurity or entropy.

param min_impurity_decrease:

float, default=0 The minimum decrease in impurity required for a split to occur. A split is only considered if it leads to at least this amount of impurity decrease. If a split does not meet this criterion, it is not performed.

Attributes:

TREE_DEPTH_LIMIT: int or None

The maximum depth of the tree.

NODE_SIZE_LIMIT: int

The minimum number of samples required to split a node further.

GROW_CONFIDENCE_THRESHOLD: float

The minimum confidence level required for a split to occur.

tree: object or None

The decision tree model constructed after fitting the data.

node_size_limit: int

The minimum number of samples required to split a node further.

min_impurity_decrease: float

The minimum decrease in impurity required for a split to occur.

__init__(max_depth=None, node_size_limit=1, grow_confidence_threshold=0, min_impurity_decrease=0)

A decision tree classifier with customizable parameters for controlling tree growth.

Parameters:

param max_depth:

int or None, default=None The maximum depth of the tree. If None, then nodes are expanded until all leaves are pure or until all leaves contain less than node_size_limit samples.

param node_size_limit:

int, default=1 The minimum number of samples required to split a node further. If the number of samples at a node is less than node_size_limit, the node is not split, and it becomes a leaf.

param grow_confidence_threshold:

float, default=0 The minimum confidence level required for a split to occur. Splits with a confidence level below this threshold are not performed. Confidence level is typically defined by impurity measures such as Gini impurity or entropy.

param min_impurity_decrease:

float, default=0 The minimum decrease in impurity required for a split to occur. A split is only considered if it leads to at least this amount of impurity decrease. If a split does not meet this criterion, it is not performed.

Attributes:

TREE_DEPTH_LIMIT: int or None

The maximum depth of the tree.

NODE_SIZE_LIMIT: int

The minimum number of samples required to split a node further.

GROW_CONFIDENCE_THRESHOLD: float

The minimum confidence level required for a split to occur.

tree: object or None

The decision tree model constructed after fitting the data.

node_size_limit: int

The minimum number of samples required to split a node further.

min_impurity_decrease: float

The minimum decrease in impurity required for a split to occur.

Methods

__init__([max_depth, node_size_limit, ...])

A decision tree classifier with customizable parameters for controlling tree growth.

calculate_gains_numeric(stat_for_lt_value, ...)

calculate_split_criterion(values, data, ...)

fit(data[, y, classifier, beta, ...])

Fits pyUID3 tree, optionally using SHAP values calculated for the classifier.

get_metadata_routing()

Get metadata routing of this object.

get_oblique_gains(data, svc_features, ...)

get_params([deep])

Get parameters for this estimator.

predict(X)

set_fit_request(*[, beta, classifier, data, ...])

Configure whether metadata should be requested to be passed to the fit method.

set_params(**params)

Set the parameters of this estimator.

try_attribute_for_split(data, attribute, cl, ...)

Attributes

PARALLEL_ENTRY_FACTOR

fit(data, y=None, *, depth, entropyEvaluator, classifier=None, beta=1, discount_importance=False, prune=False, oblique=False, n_jobs=None)

Fits pyUID3 tree, optionally using SHAP values calculated for the classifier.

Parameters

datapyuid3.Data

Data object containing dataset. It has to be object from pyuid3.Data

ynp.array

Vector containing target values

depthint, optional

This parameter should not be used. It is used internally by recurrent calls to govern the depth of the tree.

entropyEvaluator: pyuid3.EntropyEvaluator

Object responisble for calculating split criterion. Default is UncertainEntropyEvaluator. Although the naming might be confusing, other possibilities are: UncertainGiniEvaluator, UncertainSqrtGiniEvaluator

classifier: optional

A classifier that is designed according to sckit paradigm. It is required from the classifier to have predict_proba function. Default is None

beta: int

Parameter being a weight in harmonic mean between score obtained from EntropyEvaluator and SHAP values. The greater the value the more important are SHAP values when selecting a split. Default is 1.

discount_importance: boolean,

Parameter indicating if the SHAP importances should be calculated resively at every split, or if the importances calculated for the whole data should be used. In the latter case, the importances are discounted by the percentage of reduction in split criterion (e.g. Information Gain). Default it False.

prune: boolean, optional

Define if after training the tree should be pruned. The prounning is done by looking at the change in prediction on a training set. If removing a branch does not change the prediction outcome, the branch is pruned. It will provide more general trees, i.e.rules extracted from branches will have more coverage, but their precission may drop.

oblique: boolean, optional

Define if the tree should assume building linear slipts, instead of simple inequality-based spolits. Deafult False.

n_jobs: int, optional

Number of processess to use when building a tree. Default is None

Returns

pyuid3.Tree

a fitted decision tree

get_metadata_routing()

Get metadata routing of this object.

Please check User Guide on how the routing mechanism works.

Returns

routingMetadataRequest

A MetadataRequest encapsulating routing information.

get_params(deep=True)

Get parameters for this estimator.

Parameters

deepbool, default=True

If True, will return the parameters for this estimator and contained subobjects that are estimators.

Returns

paramsdict

Parameter names mapped to their values.

set_fit_request(*, beta: bool | None | str = '$UNCHANGED$', classifier: bool | None | str = '$UNCHANGED$', data: bool | None | str = '$UNCHANGED$', depth: bool | None | str = '$UNCHANGED$', discount_importance: bool | None | str = '$UNCHANGED$', entropyEvaluator: bool | None | str = '$UNCHANGED$', n_jobs: bool | None | str = '$UNCHANGED$', oblique: bool | None | str = '$UNCHANGED$', prune: bool | None | str = '$UNCHANGED$') → UId3

Configure whether metadata should be requested to be passed to the fit method.

Note that this method is only relevant when this estimator is used as a sub-estimator within a meta-estimator and metadata routing is enabled with enable_metadata_routing=True (see sklearn.set_config()). Please check the User Guide on how the routing mechanism works.

The options for each parameter are:

  • True: metadata is requested, and passed to fit if provided. The request is ignored if metadata is not provided.

  • False: metadata is not requested and the meta-estimator will not pass it to fit.

  • None: metadata is not requested, and the meta-estimator will raise an error if the user provides it.

  • str: metadata should be passed to the meta-estimator with this given alias instead of the original name.

The default (sklearn.utils.metadata_routing.UNCHANGED) retains the existing request. This allows you to change the request for some parameters and not others.

New in version 1.3.

Parameters

betastr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for beta parameter in fit.

classifierstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for classifier parameter in fit.

datastr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for data parameter in fit.

depthstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for depth parameter in fit.

discount_importancestr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for discount_importance parameter in fit.

entropyEvaluatorstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for entropyEvaluator parameter in fit.

n_jobsstr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for n_jobs parameter in fit.

obliquestr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for oblique parameter in fit.

prunestr, True, False, or None, default=sklearn.utils.metadata_routing.UNCHANGED

Metadata routing for prune parameter in fit.

Returns

selfobject

The updated object.

set_params(**params)

Set the parameters of this estimator.

The method works on simple estimators as well as on nested objects (such as Pipeline). The latter have parameters of the form <component>__<parameter> so that it’s possible to update each component of a nested object.

Parameters

**paramsdict

Estimator parameters.

Returns

selfestimator instance

Estimator instance.