bart.particle_tree

Module: bart.particle_tree

Inheritance diagram for ISLP.bart.particle_tree:

digraph inheritancefdbc56115d { bgcolor=transparent; rankdir=LR; size="8.0, 12.0"; "bart.particle_tree.ParticleTree" [URL="#ISLP.bart.particle_tree.ParticleTree",fillcolor=white,fontname="Vera Sans, DejaVu Sans, Liberation Sans, Arial, Helvetica, sans",fontsize=10,height=0.25,shape=box,style="setlinewidth(0.5),filled",target="_top",tooltip="Particle tree"]; }

Class

ParticleTree

class ISLP.bart.particle_tree.ParticleTree(tree, resid, log_weight, split_prob, min_depth, X_missing, ssv, available_predictors, m, sigmasq, mu_prior_mean, mu_prior_var, random_state)

Bases: object

Particle tree

Methods

increment_loglikelihood
marginal_loglikelihood
sample_tree_sequential
sample_values
set_resid

__init__(tree, resid, log_weight, split_prob, min_depth, X_missing, ssv, available_predictors, m, sigmasq, mu_prior_mean, mu_prior_var, random_state)

increment_loglikelihood(left_node, right_node)

marginal_loglikelihood()

sample_tree_sequential(X, X_quantiles, resid)

sample_values(resid)

set_resid(resid)

Functions

ISLP.bart.particle_tree.discrete_uniform_sampler(upper_value, random_state)

Draw an integer from the uniform distribution with bounds [0, upper_value). This is the same and np.random.randit(upper_value) but faster.

ISLP.bart.particle_tree.get_new_idx_data_points(split_value, idx_data_points, X_select)

ISLP.bart.particle_tree.grow_tree(tree, index_leaf_node, split_prior, available_predictors, X, X_quantiles, X_missing, random_state)