Struct FtrlRegressor 

pub struct FtrlRegressor { /* private fields */ }

FTRL regressor with squared loss.

Learns y ≈ w · x + b incrementally. The gradient of the squared loss w.r.t. the prediction is prediction - target, so each feature’s gradient is (prediction - target) * x_i.

Examples

use rill_ml::models::{FtrlConfig, FtrlRegressor};
use rill_ml::sparse::SparseFeatures;
use rill_ml::SparseRegressor;

let mut model = FtrlRegressor::new(FtrlConfig::default()).unwrap();
let sf = SparseFeatures::from_sorted(vec![(0, 1.0), (1, 2.0)]).unwrap();
let _pred = model.predict(&sf).unwrap();
model.learn(&sf, 3.0).unwrap();

Implementations

impl FtrlRegressor

pub fn new(config: FtrlConfig) -> Result<Self, RillError>

Create a new FTRL regressor.

Returns an error if the configuration is invalid.

pub const fn config(&self) -> &FtrlConfig

The model configuration.

pub fn weights(&self) -> Vec<(FeatureId, f64)>

Return the current non-zero feature weights, sorted by FeatureId.

Features whose FTRL weight is exactly zero (due to L1 soft-thresholding or never having been updated) are excluded.

pub fn intercept(&self) -> f64

Compute the current intercept (bias) weight.

pub fn feature_count(&self) -> usize

Number of distinct features the model has seen.

Trait Implementations

impl Clone for FtrlRegressor

fn clone(&self) -> FtrlRegressor

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for FtrlRegressor

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

Formats the value using the given formatter.

impl SparseRegressor for FtrlRegressor

fn samples_seen(&self) -> u64

How many training samples the model has seen so far.

fn predict(&self, features: &SparseFeatures) -> Result<f64, RillError>

Predict the target for the given sparse features.

fn learn( &mut self, features: &SparseFeatures, target: f64, ) -> Result<(), RillError>

Update the model using a single labeled sparse sample.

fn reset(&mut self)

Reset the model to its initial state.