Implementation:Scikit learn Scikit learn OneVsRestClassifier

Knowledge Sources	Scikit_learn Scikit-learn Docs
Domains	Machine Learning, Classification, Meta-Estimators
Last Updated	2026-02-08 15:00 GMT

Overview

Concrete meta-estimator for multiclass and multilabel classification strategies provided by scikit-learn.

Description

The multiclass module provides meta-estimators that extend binary classifiers to handle multiclass and multilabel classification. OneVsRestClassifier fits one binary classifier per class, OneVsOneClassifier fits one classifier per pair of classes, and OutputCodeClassifier uses error-correcting output codes. These meta-estimators support predict_proba, decision_function, and metadata routing.

Usage

Use these meta-estimators when you want to apply custom multiclass strategies, particularly when the base classifier is binary-only. For most cases scikit-learn classifiers handle multiclass natively, but these are useful for experimentation or specialized strategies.

Code Reference

Source Location

• Repository: scikit-learn
• File: sklearn/multiclass.py

Signature

class OneVsRestClassifier(
    MultiOutputMixin, ClassifierMixin, MetaEstimatorMixin, BaseEstimator
):
    def __init__(self, estimator, *, n_jobs=None, verbose=0):
        ...
    def fit(self, X, y, **fit_params):
        ...
    def predict(self, X):
        ...
    def predict_proba(self, X):
        ...

class OneVsOneClassifier(MetaEstimatorMixin, ClassifierMixin, BaseEstimator):
    def __init__(self, estimator, *, n_jobs=None):
        ...

class OutputCodeClassifier(MetaEstimatorMixin, ClassifierMixin, BaseEstimator):
    def __init__(self, estimator, *, code_size=1.5, random_state=None, n_jobs=None):
        ...

Import

from sklearn.multiclass import OneVsRestClassifier, OneVsOneClassifier, OutputCodeClassifier

I/O Contract

Inputs

Name	Type	Required	Description
estimator	estimator instance	Yes	Base classifier (binary or multiclass)
X	array-like of shape (n_samples, n_features)	Yes	Training input samples
y	array-like of shape (n_samples,)	Yes	Target values (multiclass or multilabel)
n_jobs	int	No	Number of parallel jobs for fitting

Outputs

Name	Type	Description
predictions	ndarray of shape (n_samples,)	Predicted class labels
probabilities	ndarray of shape (n_samples, n_classes)	Class probability estimates (if available)

Usage Examples

Basic Usage

from sklearn.multiclass import OneVsRestClassifier
from sklearn.svm import SVC
from sklearn.datasets import make_classification

X, y = make_classification(n_classes=3, n_informative=3, random_state=42)
clf = OneVsRestClassifier(SVC()).fit(X, y)
predictions = clf.predict(X)
print(predictions[:5])

Related Pages

• Principle:Scikit_learn_Scikit_learn_Multiclass_Multioutput