Struct Binarizer 

pub struct Binarizer<F> { /* private fields */ }

A stateless feature binarizer.

Values strictly greater than threshold become 1.0; all other values become 0.0. The default threshold is 0.0.

This transformer is stateless — no fitting is needed. Call Transform::transform directly.

Examples

use ferrolearn_preprocess::binarizer::Binarizer;
use ferrolearn_core::traits::Transform;
use ndarray::array;

let binarizer = Binarizer::<f64>::new(0.5);
let x = array![[0.0, 0.5, 1.0]];
let out = binarizer.transform(&x).unwrap();
// out = [[0.0, 0.0, 1.0]]

Implementations

impl<F: Float + Send + Sync + 'static> Binarizer<F>

pub fn new(threshold: F) -> Self

Create a new Binarizer with the given threshold (and the default copy = true).

sklearn constrains threshold to Interval(Real, None, None, closed="neither") on binarize (_data.py:2114-2115) — an OPEN interval (-inf, inf) that EXCLUDES NaN/±inf. A non-finite threshold is NOT rejected by new (no validation at construction, matching sklearn’s __init__, which stores params unchecked); it is rejected later by Fit::fit / Transform::transform / binarize (InvalidParameter), matching sklearn’s _fit_context / @validate_params raising InvalidParameterError at fit/binarize.

pub fn threshold(&self) -> F

Return the configured threshold.

pub fn with_copy(self, copy: bool) -> Self

Set the copy parameter (sklearn Binarizer(copy=...), _data.py:2253, _parameter_constraints {copy:["boolean"]} :2250).

This is an ACCEPT-AND-DOCUMENT no-op: ferrolearn’s Transform always returns a freshly allocated array, so copy has no observable effect on the output. It is retained for API parity with scikit-learn.

pub fn copy(&self) -> bool

Return the configured copy flag (sklearn Binarizer.copy).

Trait Implementations

impl<F: Clone> Clone for Binarizer<F>

fn clone(&self) -> Binarizer<F>

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl<F: Debug> Debug for Binarizer<F>

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

Formats the value using the given formatter.

impl<F: Float + Send + Sync + 'static> Default for Binarizer<F>

fn default() -> Self

Returns the “default value” for a type.

impl<F: Float + Send + Sync + 'static> Fit<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>, ()> for Binarizer<F>

fn fit(&self, x: &Array2<F>, _y: &()) -> Result<FittedBinarizer<F>, FerroError>

Validate the input and record n_features_in_, returning a FittedBinarizer.

Binarizer is stateless: like scikit-learn’s Binarizer.fit (sklearn/preprocessing/_data.py:2257-2278, “Only validates estimator’s parameters”), this learns NO statistics. It runs the SAME check_array validation as Transform::transform / binarize (REQ-9, via the shared [validate_binarize_input] helper) and records n_features_in_ = x.ncols(). sklearn’s _validate_data uses the default force_all_finite=True, so NaN/±inf are REJECTED in fit (Binarizer().fit([[nan]]) / [[inf]] raise ValueError). sklearn’s _fit_context validates _parameter_constraints (:2249) BEFORE the data, and threshold is constrained to Interval(Real, None, None, closed="neither") on binarize (_data.py:2114) — an OPEN interval (-inf, inf) that EXCLUDES NaN/±inf. A non-finite threshold is therefore rejected here (param-check first, matching _fit_context).

Errors

Returns FerroError::InvalidParameter if threshold is non-finite (NaN/±inf, sklearn Interval(Real, None, None, closed="neither"), _data.py:2114), FerroError::InsufficientSamples for zero rows, and FerroError::InvalidParameter for zero features or any non-finite value (NaN, +inf, -inf) — matching check_array (sklearn/utils/validation.py:1084, :1093, :1063) as routed through Binarizer.fit -> _validate_data (_data.py:2277).

type Fitted = FittedBinarizer<F>

The fitted model type returned by fit.

type Error = FerroError

The error type returned by fit.

impl<F: Float + Send + Sync + 'static> Transform<ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>> for Binarizer<F>

fn transform(&self, x: &Array2<F>) -> Result<Array2<F>, FerroError>

Apply the threshold: values > threshold become 1.0, others become 0.0.

Errors

Returns FerroError::InsufficientSamples if x has zero rows. This mirrors scikit-learn’s Binarizer.transform (sklearn/preprocessing/_data.py:2301), whose _validate_data -> check_array min-samples check raises ValueError: Found array with 0 sample(s) ... while a minimum of 1 is required by Binarizer.

Returns FerroError::InvalidParameter if x has zero features (columns). This mirrors scikit-learn’s Binarizer.transform (sklearn/preprocessing/_data.py:2301), whose _validate_data -> check_array min-features check (utils/validation.py:1093, ensure_min_features=1) raises ValueError: Found array with 0 feature(s) (shape=(3, 0)) while a minimum of 1 is required by Binarizer.

Returns FerroError::InvalidParameter if x contains any non-finite value (NaN, +inf, or -inf). This mirrors scikit-learn’s Binarizer.transform (sklearn/preprocessing/_data.py:2301), which validates input via check_array(force_all_finite=True) and raises ValueError: Input X contains NaN. / Input X contains infinity ... before applying the threshold comparison.

type Output = ArrayBase<OwnedRepr<F>, Dim<[usize; 2]>>

The transformed output type.

type Error = FerroError

The error type returned by transform.