Struct smartcore::ensemble::random_forest_classifier::RandomForestClassifierParameters

pub struct RandomForestClassifierParameters {
    pub criterion: SplitCriterion,
    pub max_depth: Option<u16>,
    pub min_samples_leaf: usize,
    pub min_samples_split: usize,
    pub n_trees: u16,
    pub m: Option<usize>,
    pub keep_samples: bool,
    pub seed: u64,
}

Parameters of the Random Forest algorithm. Some parameters here are passed directly into base estimator.

Fields

criterion: SplitCriterion

Split criteria to use when building a tree. See Decision Tree Classifier

max_depth: Option<u16>

Tree max depth. See Decision Tree Classifier

min_samples_leaf: usize

The minimum number of samples required to be at a leaf node. See Decision Tree Classifier

min_samples_split: usize

The minimum number of samples required to split an internal node. See Decision Tree Classifier

n_trees: u16

The number of trees in the forest.

m: Option<usize>

Number of random sample of predictors to use as split candidates.

keep_samples: bool

Whether to keep samples used for tree generation. This is required for OOB prediction.

seed: u64

Seed used for bootstrap sampling and feature selection for each tree.

Implementations

impl RandomForestClassifierParameters

pub fn with_criterion(self, criterion: SplitCriterion) -> Self

Split criteria to use when building a tree. See Decision Tree Classifier

pub fn with_max_depth(self, max_depth: u16) -> Self

Tree max depth. See Decision Tree Classifier

pub fn with_min_samples_leaf(self, min_samples_leaf: usize) -> Self

The minimum number of samples required to be at a leaf node. See Decision Tree Classifier

pub fn with_min_samples_split(self, min_samples_split: usize) -> Self

The minimum number of samples required to split an internal node. See Decision Tree Classifier

pub fn with_n_trees(self, n_trees: u16) -> Self

The number of trees in the forest.

pub fn with_m(self, m: usize) -> Self

Number of random sample of predictors to use as split candidates.

pub fn with_keep_samples(self, keep_samples: bool) -> Self

Whether to keep samples used for tree generation. This is required for OOB prediction.

pub fn with_seed(self, seed: u64) -> Self

Seed used for bootstrap sampling and feature selection for each tree.

Trait Implementations

impl Clone for RandomForestClassifierParameters

fn clone(&self) -> RandomForestClassifierParameters

Returns a copy of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for RandomForestClassifierParameters

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for RandomForestClassifierParameters

fn default() -> Self

Returns the “default value” for a type.

impl<TX: Number + FloatNumber + PartialOrd, TY: Number + Ord, X: Array2<TX>, Y: Array1<TY>> SupervisedEstimator<X, Y, RandomForestClassifierParameters> for RandomForestClassifier<TX, TY, X, Y>

fn new() -> Self

Empty constructor, instantiate an empty estimator. Object is dropped as soon as fit() is called. used to pass around the correct fit() implementation. by calling ::fit(). mostly used to be used with model_selection::cross_validate(...)

fn fit( x: &X, y: &Y, parameters: RandomForestClassifierParameters ) -> Result<Self, Failed>

Fit a model to a training dataset, estimate model’s parameters.