Implementation:Scikit learn Scikit learn VotingClassifier Init

Template:Metadata

Overview

Concrete tool for combining multiple classifiers via majority or soft voting provided by scikit-learn. The Template:Code fits clones of the provided base estimators and aggregates their predictions using either hard voting (majority class label) or soft voting (averaged predicted probabilities). It supports optional per-classifier weighting to emphasize stronger models.

Constructor Signature

from sklearn.ensemble import VotingClassifier

VotingClassifier(
    estimators,
    *,
    voting="hard",
    weights=None,
    n_jobs=None,
    flatten_transform=True,
    verbose=False,
)

Parameters

• estimators (list of (str, estimator) tuples) -- Base estimators to be stacked together. Each element is a tuple of a name string and an estimator instance. An estimator can be set to Template:Code using Template:Code to exclude it from the ensemble.
• voting ({"hard", "soft"}, default="hard") -- If "hard", uses predicted class labels for majority rule voting. If "soft", predicts the class label based on the argmax of the sums of the predicted probabilities (recommended for well-calibrated classifiers).
• weights (array-like of shape (n_classifiers,), default=None) -- Sequence of weights to apply to the predicted class labels (hard voting) or class probabilities (soft voting). If None, uniform weights are used.
• n_jobs (int, default=None) -- Number of jobs to run in parallel for Template:Code. None means 1 unless in a Template:Code context. -1 means using all processors.
• flatten_transform (bool, default=True) -- Only affects the shape of Template:Code output when Template:Code. If True, returns a matrix of shape Template:Code. If False, returns shape Template:Code.
• verbose (bool, default=False) -- If True, prints the time elapsed while fitting each estimator.

Fitted Attributes

• estimators_ -- The collection of fitted sub-estimators (excludes any set to "drop").
• named_estimators_ -- A Template:Code object allowing access to fitted sub-estimators by name.
• le_ -- A Template:Code transformer used to encode labels during fit and decode during prediction.
• classes_ -- The class labels (ndarray of shape Template:Code).
• n_features_in_ -- Number of features seen during Template:Code (only defined if the underlying classifiers expose this attribute).
• feature_names_in_ -- Names of features seen during Template:Code (only defined if the underlying estimators expose this attribute).

Example Usage

import numpy as np
from sklearn.linear_model import LogisticRegression
from sklearn.naive_bayes import GaussianNB
from sklearn.ensemble import RandomForestClassifier, VotingClassifier

clf1 = LogisticRegression(random_state=1)
clf2 = RandomForestClassifier(n_estimators=50, random_state=1)
clf3 = GaussianNB()

X = np.array([[-1, -1], [-2, -1], [-3, -2], [1, 1], [2, 1], [3, 2]])
y = np.array([1, 1, 1, 2, 2, 2])

eclf = VotingClassifier(
    estimators=[("lr", clf1), ("rf", clf2), ("gnb", clf3)],
    voting="soft",
)
eclf.fit(X, y)
print(eclf.predict(X))
# [1 1 1 2 2 2]

Source Location

Template:Code, class Template:Code (lines 194-540).

Related Pages

• Principle:Scikit_learn_Scikit_learn_Voting_Ensemble