Struct IterativeImputer 

pub struct IterativeImputer { /* private fields */ }

Iterative Imputer using the MICE (Multiple Imputation by Chained Equations) algorithm

This transformer iteratively models each feature with missing values as a function of other features. The algorithm performs multiple rounds of imputation where each feature is predicted using the other features in a round-robin fashion.

MICE is particularly useful when:

• There are multiple features with missing values
• The missing patterns are complex
• You want to model relationships between features

Implementations

impl IterativeImputer

pub fn new( max_iter: usize, tolerance: f64, initial_strategy: ImputeStrategy, missingvalues: f64, alpha: f64, ) -> Self

Creates a new IterativeImputer

Arguments

• max_iter - Maximum number of iterations
• tolerance - Convergence tolerance
• initial_strategy - Strategy for initial imputation
• missingvalues - Value representing missing data
• alpha - Regularization parameter for regression

Returns

• A new IterativeImputer instance

pub fn with_defaults() -> Self

Creates an IterativeImputer with default parameters

Uses 10 iterations, 1e-3 tolerance, mean initial strategy, NaN missing values, and 1e-6 regularization.

pub fn with_max_iter(_maxiter: usize) -> Self

Creates an IterativeImputer with specified max iterations and defaults for other parameters

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

Set the random seed for reproducible results

pub fn with_alpha(self, alpha: f64) -> Self

Set the regularization parameter

pub fn with_min_improvement(self, minimprovement: f64) -> Self

Set the minimum improvement threshold

pub fn fit<S>(&mut self, x: &ArrayBase<S, Ix2>) -> Result<()>

where S: Data, S::Elem: Float + NumCast,

Fits the IterativeImputer to the input data

Arguments

• x - The input data, shape (n_samples, n_features)

Returns

• Result<()> - Ok if successful, Err otherwise

pub fn transform<S>(&self, x: &ArrayBase<S, Ix2>) -> Result<Array2<f64>>

where S: Data, S::Elem: Float + NumCast,

Transforms the input data by imputing missing values using MICE

Arguments

• x - The input data, shape (n_samples, n_features)

Returns

• Result<Array2<f64>> - Transformed data with missing values imputed

pub fn fit_transform<S>(&mut self, x: &ArrayBase<S, Ix2>) -> Result<Array2<f64>>

where S: Data, S::Elem: Float + NumCast,

Fits the imputer and transforms the data in one step

Arguments

• x - The input data, shape (n_samples, n_features)

Returns

• Result<Array2<f64>> - Transformed data with missing values imputed