Trait Fit

pub trait Fit<DM: DesignMatrix, T: AsMultiTargets, E: Error> {
    type Object;

    // Required method
    fn fit(&self, dataset: &DatasetBase<DM, T>) -> Result<Self::Object, E>;
}

The Fit trait allows an estimator to be fitted to a dataset (a combination of design matrix and targets). More formally, the model estimates coefficients that minimizes an objective.

Required Associated Types

type Object

Required Methods

fn fit(&self, dataset: &DatasetBase<DM, T>) -> Result<Self::Object, E>

Implementors

impl<DM, T, E, P> Fit<DM, T, E> for P

where DM: DesignMatrix, T: AsMultiTargets, P: ParamGuard, P::Checked: Fit<DM, T, E>, E: Error + From<P::Error>,

Performs checking step and calls fit on the checked hyperparameters. If checking failed, the checking error is converted to the original error type of Fit and returned.

type Object = <<P as ParamGuard>::Checked as Fit<DM, T, E>>::Object

impl<F: Float, S: Data<Elem = F>, T: AsMultiTargets<Elem = F>> Fit<ArrayBase<S, Dim<[usize; 2]>>, T, EstimatorError> for BlockMCPValidParams<F>

This implements the block coordinate descent optimization procedure for multi-task problems and dense design matrices.

type Object = BlockMCPEstimator<F>

impl<F: Float, S: Data<Elem = F>, T: AsMultiTargets<Elem = F>> Fit<ArrayBase<S, Dim<[usize; 2]>>, T, EstimatorError> for MultiTaskElasticNetValidParams<F>

This implements the coordinate descent optimization procedure for single-task problems and dense design matrices.

type Object = MultiTaskElasticNet<F>

impl<F: Float, S: Data<Elem = F>, T: AsMultiTargets<Elem = F>> Fit<ArrayBase<S, Dim<[usize; 2]>>, T, EstimatorError> for MultiTaskLassoValidParams<F>

This implements the block coordinate descent optimization procedure for multi-task problem and dense design matrices.

type Object = MultiTaskLasso<F>

impl<F: Float, S: Data<Elem = F>, T: AsSingleTargets<Elem = F>> Fit<ArrayBase<S, Dim<[usize; 2]>>, T, EstimatorError> for ElasticNetValidParams<F>

This implements the coordinate descent optimization procedure for single-task problems and dense design matrices.

type Object = ElasticNet<F>

impl<F: Float, S: Data<Elem = F>, T: AsSingleTargets<Elem = F>> Fit<ArrayBase<S, Dim<[usize; 2]>>, T, EstimatorError> for LassoValidParams<F>

This implements the coordinate descent optimization procedure for single-task problem and dense design matrices.

type Object = Lasso<F>

impl<F: Float, S: Data<Elem = F>, T: AsSingleTargets<Elem = F>> Fit<ArrayBase<S, Dim<[usize; 2]>>, T, EstimatorError> for MCPValidParams<F>

This implements the coordinate descent optimization procedure for single-task problems and dense design matrices.

type Object = MCPEstimator<F>

impl<F: Float, T: AsMultiTargets<Elem = F>> Fit<CSCArray<'_, F>, T, EstimatorError> for BlockMCPValidParams<F>

This implements the block coordinate descent optimization procedure for multi-task problems and sparse design matrices.

type Object = BlockMCPEstimator<F>

impl<F: Float, T: AsMultiTargets<Elem = F>> Fit<CSCArray<'_, F>, T, EstimatorError> for MultiTaskElasticNetValidParams<F>

This implements the coordinate descent optimization procedure for single-task problems and sparse design matrices.

type Object = MultiTaskElasticNet<F>

impl<F: Float, T: AsMultiTargets<Elem = F>> Fit<CSCArray<'_, F>, T, EstimatorError> for MultiTaskLassoValidParams<F>

This implements the block coordinate descent optimization procedure for multi-task problem and sparse design matrices.

type Object = MultiTaskLasso<F>

impl<F: Float, T: AsSingleTargets<Elem = F>> Fit<CSCArray<'_, F>, T, EstimatorError> for ElasticNetValidParams<F>

This implements the coordinate descent optimization procedure for single-task problems and sparse design matrices.

type Object = ElasticNet<F>

impl<F: Float, T: AsSingleTargets<Elem = F>> Fit<CSCArray<'_, F>, T, EstimatorError> for LassoValidParams<F>

This implements the coordinate descent optimization procedure for single-task problem and sparse design matrices.

type Object = Lasso<F>

impl<F: Float, T: AsSingleTargets<Elem = F>> Fit<CSCArray<'_, F>, T, EstimatorError> for MCPValidParams<F>

This implements the coordinate descent optimization procedure for single-task problems and sparse design matrices.

type Object = MCPEstimator<F>