jpt.learning.pruning

Pruning criteria for JPT learning.

Submodules

• jpt.learning.pruning.xi

Classes

XiPruningCriterion

Prune-or-split callback based on xi.

Package Contents

class jpt.learning.pruning.XiPruningCriterion(alpha: float = 0.05, min_n: int = 30)

Prune-or-split callback based on xi.

Tests whether any feature-target pair shows significant functional dependence in the current partition. Under H0 (independence), sqrt(n) * xi ~ N(0, 2/5).

Parameters:

• alpha – significance level

• min_n – minimum samples to apply test

z_crit: float

min_n: int = 30

__call__(jpt: JPT, partition: JPTPartition, indices: numpy.ndarray, data: numpy.ndarray) → bool

Return True to prune (stop splitting).

Parameters:

• jpt – the JPT being learned

• partition – current data partition

• indices – index buffer mapping positions to data rows

• data – the full training data array (n_samples x n_vars)

Returns:

True if the node should become a leaf