ml_decision_tree

ml_decision_tree(
    x,
    y,
    newdata,
    seed=None,
    regression=True,
    features=None,
    minimum_split_size=2,
    max_depth=100,
)

Decision tree regression or classification.

Mirrors the C# DecisionTree class: recursively splits the training data on the feature and threshold that most reduce variance (regression) or most increase information gain (classification), then predicts by walking a new observation down the tree.

At the library’s defaults a REGRESSION tree recurses until every leaf holds a single training observation, so it memorizes the training data and generalizes poorly. That is upstream’s behaviour, not a port artifact, and it is why :func:ml_random_forest exists. Set minimum_split_size or max_depth to regularize it.

Parameters

Name	Type	Description	Default
x	array_like	Training predictors, one row per observation.	required
y	array_like	The training response, one value per row of x.	required
newdata	array_like	Predictors to predict for, with the same number of columns as x.	required
seed	int	PRNG seed for the random feature subsets; None uses the computer clock.	None
regression	bool	True fits a regression tree; False a classifier.	True
features	int	The number of random features to consider at each split. None (the default) uses the library’s own max(1, x.shape[1] - 1).	None
minimum_split_size	int	The smallest node the tree will split.	2
max_depth	int	The recursion cap.	100

Returns

Name	Type	Description
	numpy.ndarray	One prediction per row of newdata.

Examples

>>> from corehydropy import ml_decision_tree
>>> x = [1, 2, 3, 4, 5, 6, 100, 101, 102, 103, 104, 105]
>>> y = [10, 10, 10, 10, 10, 10, 100, 100, 100, 100, 100, 100]
>>> ml_decision_tree(x, y, newdata=[3, 104], seed=7).tolist()
[10.0, 100.0]