Struct AdaBoostParams 

pub struct AdaBoostParams<P, R>(/* private fields */);

A helper struct for building AdaBoost hyperparameters.

This struct follows the builder pattern, allowing you to chain method calls to configure the AdaBoost algorithm before fitting.

Example

use linfa_ensemble::AdaBoostParams;
use linfa_trees::DecisionTree;

let params = AdaBoostParams::new(DecisionTree::<f64, usize>::params().max_depth(Some(1)))
    .n_estimators(100)
    .learning_rate(0.5);

Implementations

impl<P> AdaBoostParams<P, ThreadRng>

pub fn new(model_params: P) -> AdaBoostParams<P, ThreadRng>

Create a new AdaBoost parameter set with default values and a thread-local RNG.

Arguments

• model_params - The parameters for the base learner (e.g., DecisionTreeParams)

Default Values

• n_estimators: 50
• learning_rate: 1.0

impl<P, R: Rng + Clone> AdaBoostParams<P, R>

pub fn new_fixed_rng(model_params: P, rng: R) -> AdaBoostParams<P, R>

Create a new AdaBoost parameter set with a fixed RNG for reproducibility.

Arguments

• model_params - The parameters for the base learner
• rng - A seeded random number generator for reproducible results

Example

use linfa_ensemble::AdaBoostParams;
use linfa_trees::DecisionTree;
use ndarray_rand::rand::SeedableRng;
use rand::rngs::SmallRng;

let rng = SmallRng::seed_from_u64(42);
let params = AdaBoostParams::new_fixed_rng(
    DecisionTree::<f64, usize>::params().max_depth(Some(1)),
    rng
);

pub fn n_estimators(self, n_estimators: usize) -> Self

Set the maximum number of weak learners to train.

Arguments

• n_estimators - Must be at least 1. Typical values: 50-500

Notes

Higher values generally lead to better training performance but:

• Increase training time linearly
• May lead to overfitting
• Should be balanced with learning_rate

pub fn learning_rate(self, learning_rate: f64) -> Self

Set the learning rate (shrinkage parameter).

Arguments

• learning_rate - Must be positive. Typical values: 0.01 to 2.0

Notes

• Values < 1.0 provide regularization and often improve generalization
• Lower values require more estimators to achieve similar performance
• A common strategy is to use learning_rate=0.1 with n_estimators=500

Trait Implementations

impl<P: Clone, R: Clone> Clone for AdaBoostParams<P, R>

fn clone(&self) -> AdaBoostParams<P, R>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<P: Copy, R: Copy> Copy for AdaBoostParams<P, R>

impl<P: Debug, R: Debug> Debug for AdaBoostParams<P, R>

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

Formats the value using the given formatter.

impl<P, R> ParamGuard for AdaBoostParams<P, R>

type Checked = AdaBoostValidParams<P, R>

The checked hyperparameters

type Error = Error

Error type resulting from failed hyperparameter checking

fn check_ref(&self) -> Result<&Self::Checked>

Checks the hyperparameters and returns a reference to the checked hyperparameters if successful

fn check(self) -> Result<Self::Checked>

Checks the hyperparameters and returns the checked hyperparameters if successful

fn check_unwrap(self) -> Self::Checked

where Self: Sized,

Calls check() and unwraps the result

impl<P: PartialEq, R: PartialEq> PartialEq for AdaBoostParams<P, R>

fn eq(&self, other: &AdaBoostParams<P, R>) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl<P, R> StructuralPartialEq for AdaBoostParams<P, R>