Releases: neurodata/scikit-learn
v1.2.3
Renamed sklearn to sklearn_fork, so that it is compatible with installing scikit-learn main under the sklearn namespace. Now, one can install both scikit-learn-tree and scikit-learn in their packages.
It is recommended that one use sklearn_fork for all things related to decision tree models such as:
RandomForest*ExtraTrees*- and any importable item from the
tree/submodule
and then use sklearn for anything else.
v1.2.2
A v1.2 in-line release with upstream scikit-learn.
This version is backwards compatible with v1.2 of scikit-learn with the exception of the changelog mentioned in the README. These changes are mostly to private API with the exception of the max_bins parameter added to the Forest based models in sklearn/ensemble/. Currently the max_bins feature on this fork is experimental and should be used with caution.
The details of the changes are listed below.
Cython Internal Private API:
Note, the Cython API for scikit-learn is still not a publicly supported API, so it may
change without warning.
- leaf and split nodes: These nodes are treated the same way and there is no internal
API for setting them differently. Quantile trees and causal trees inherently generalize
how leaf nodes are set. - Criterion class: The criterion class currently assumes a supervised learning interface.
- Our fix: We implement a
BaseCriterionobject that provides an abstract API for unsupervised criterion.
- Our fix: We implement a
- Splitter class: The splitter clas currently assumes a supervised learning interface and
does not provide a way of generalizing the way split candidates are proposed.- Our fix: We implement a
BaseSplitterobject that provides an abstract API for unsupervised splitters and also implement an API to allow generalizations of theSplitRecordstruct andSplitter.node_splitfunction. For example, this enables oblique splits to be considered.
- Our fix: We implement a
- Tree class: The tree class currently assumes a supervised learning interface and does not
provide a way of generalizing the type of tree.- Our fix: We implementa
BaseTreeobject that provides an abstract API for general tree models and also implement an API that allows generalization of the type of tree. For example, oblique trees are trivially implementable as an extension now.
- Our fix: We implementa
- stopping conditions for splitter: Currently, the
Splitter.node_splitfunction has various
stopping conditions for the splitter based on hyperparameters. It is plausible that these conditions
may be extended. For example, in causal trees, one may want the splitter to also account for
a minimal degree of heterogeneity (i.e. variance) in its children nodes.
Python API:
sklearn.tree.BaseDecisionTreeassumes the underlying tree model is supervised: They
parameter is required to be passed in, which is not necessary for general tree-based models.
For example, an unsupervised tree may pass iny=None.- Our fix: We fix this API, so the
BaseDecisionTreeis subclassable by unsupervised tree models that do not requireyto be defined.
- Our fix: We fix this API, so the
sklearn.tree.BaseDecisionTreedoes not provide a way to generalize theCriterion,Splitter
andTreeCython classes used: The current codebase requires users to define custom
criterion and/or splitters outside the instantiation of theBaseDecisionTree. This prevents
users from generalizing theCriterionandSplitterand creating a neat Python API wrapper.
Moreover, theTreeclass is not customizable.- Our fix: We internally implement a private function to actually build the entire tree,
BaseDecisionTree._build_tree, which can be overridden in subclasses that customize the criterion, splitter, or tree, or any combination of them.
- Our fix: We internally implement a private function to actually build the entire tree,
sklearn.ensemble.BaseForestand its subclass algorithms are slow whenn_samplesis very high. Binning
features into a histogram, which is the basis of "LightGBM" and "HistGradientBoostingClassifier" is a computational
trick that can both significantly increase runtime efficiency, but also help prevent overfitting in trees, since
the sorting in "BestSplitter" is done on bins rather than the continuous feature values. This would enable
random forests and their variants to scale to millions of samples.- Our fix: We added a
max_bins=Nonekeyword argument to theBaseForestclass, and all its subclasses. The default behavior is no binning. The current implementation is not necessarily efficient. There are several improvements to be made. See below.
- Our fix: We added a
Overall, the existing tree models, such as sklearn.tree.DecisionTreeClassifier <https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html>_
and sklearn.ensemble.RandomForestClassifier <https://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html#sklearn.ensemble.RandomForestClassifier>_ all work exactly the same as they
would in scikit-learn main, but these extensions enable 3rd-party packages to extend
the Cython/Python API easily.
v1.1 Refactored Tree (bug fixes)
Enable scikit-learn refactored trees to operate cleanly at the Python interface
v1.0-refactoredtrees
Refactored scikit-learn tree submodule to enable the following features:
- separate leaf/split node setting in the Cython Tree
- oblique splits that is the Forest-RC algorithm implemented by Breiman 2001
- implement an abstract base class for Criterion, Splitter and Tree in Cython tree submodule
- modularize the Python tree class to allow for different Cython tree implementations