Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

GaussianNb

Classifier sklearn-compatible 📊 Naive Bayes

Gaussian Naive Bayes — likelihood modelled as Gaussian per class per feature. / Naive Bayes Gaussien — vraisemblance modélisée comme Gaussienne par classe et feature.

⚡ Rust-native ✓ sklearn parity
Quick start — Python 
import seraplot as sp
from sklearn.datasets import load_iris
X, y = load_iris(return_X_y=True)
gnb = sp.GaussianNB()
gnb.fit(X, y)
print(gnb.score(X, y))
💡
EN — Drop-in replacement: sp.GaussianNb has the same API as sklearn.
FR — Remplacement direct : même API que sklearn, changez l'import.

API Reference

JSON function name

ml_gaussian_nb — aliases: gaussian_nb

Python class
sp.GaussianNb()
Constructor Parameters

No constructor parameters.

Returns

JSON with predictions.

Algorithm

$$P(x_j | y=c) = \frac{1}{\sqrt{2\pi\sigma_{cj}^2}} \exp!\left(-\frac{(x_j-\mu_{cj})^2}{2\sigma_{cj}^2}\right)$$

Example
import seraplot as sp
from sklearn.datasets import load_iris
X, y = load_iris(return_X_y=True)
gnb = sp.GaussianNB()
gnb.fit(X, y)
print(gnb.score(X, y))

Référence API

Nom de fonction JSON

ml_gaussian_nb — alias : gaussian_nb

Classe Python
sp.GaussianNb()
Paramètres du constructeur

Aucun paramètre de constructeur.

Retourne

JSON avec predictions.

Algorithme

$$P(x_j | y=c) = \frac{1}{\sqrt{2\pi\sigma_{cj}^2}} \exp!\left(-\frac{(x_j-\mu_{cj})^2}{2\sigma_{cj}^2}\right)$$

Exemple
import seraplot as sp
from sklearn.datasets import load_iris
X, y = load_iris(return_X_y=True)
gnb = sp.GaussianNB()
gnb.fit(X, y)
print(gnb.score(X, y))

MultinomialNb

Classifier sklearn-compatible 📊 Naive Bayes

Multinomial Naive Bayes — for count/frequency features (text, bag-of-words). / Naive Bayes Multinomial — pour features de comptage/fréquence (texte, sac de mots).

⚡ Rust-native ✓ sklearn parity
Quick start — Python 
import seraplot as sp, numpy as np
X = np.random.randint(0, 10, size=(300, 5)).astype(float)
y = (X[:, 0] > 5).astype(int)
mnb = sp.MultinomialNB(alpha=1.0)
mnb.fit(X, y)
print(mnb.score(X, y))
💡
EN — Drop-in replacement: sp.MultinomialNb has the same API as sklearn.
FR — Remplacement direct : même API que sklearn, changez l'import.

API Reference

JSON function name

ml_multinomial_nb — aliases: multinomial_nb

Python class
sp.MultinomialNb(alpha=1.0)
Constructor Parameters
ParameterTypeDefaultDescription
alphafloat1.0Additive (Laplace) smoothing.
Returns

JSON with predictions.

Algorithm

$$P(x|y=c) = \frac{N_{cy} + \alpha}{N_c + \alpha p}$$

Example
import seraplot as sp, numpy as np
X = np.random.randint(0, 10, size=(300, 5)).astype(float)
y = (X[:, 0] > 5).astype(int)
mnb = sp.MultinomialNB(alpha=1.0)
mnb.fit(X, y)
print(mnb.score(X, y))

Référence API

Nom de fonction JSON

ml_multinomial_nb — alias : multinomial_nb

Classe Python
sp.MultinomialNb(alpha=1.0)
Paramètres du constructeur
ParamètreTypeDéfautDescription
alphafloat1.0Lissage additif (Laplace).
Retourne

JSON avec predictions.

Algorithme

$$P(x|y=c) = \frac{N_{cy} + \alpha}{N_c + \alpha p}$$

Exemple
import seraplot as sp, numpy as np
X = np.random.randint(0, 10, size=(300, 5)).astype(float)
y = (X[:, 0] > 5).astype(int)
mnb = sp.MultinomialNB(alpha=1.0)
mnb.fit(X, y)
print(mnb.score(X, y))

BernoulliNb

Classifier sklearn-compatible 📊 Naive Bayes

Bernoulli Naive Bayes — for binary/boolean features. / Naive Bayes Bernoulli — pour features binaires/booléennes.

⚡ Rust-native ✓ sklearn parity
Quick start — Python 
import seraplot as sp, numpy as np
X = (np.random.randn(400, 6) > 0).astype(float)
y = (X[:, 0] & X[:, 1]).astype(int)
bnb = sp.BernoulliNB(alpha=1.0)
bnb.fit(X, y)
print(bnb.score(X, y))
💡
EN — Drop-in replacement: sp.BernoulliNb has the same API as sklearn.
FR — Remplacement direct : même API que sklearn, changez l'import.

API Reference

JSON function name

ml_bernoulli_nb — aliases: bernoulli_nb

Python class
sp.BernoulliNb(alpha=1.0, binarize=0.0)
Constructor Parameters
ParameterTypeDefaultDescription
alphafloat1.0Additive (Laplace) smoothing.
binarizefloat0.0Threshold for binarising features.
Returns

JSON with predictions.

Algorithm

$$P(x_j|y=c) = p_{cj}^{x_j}(1-p_{cj})^{1-x_j}$$

Example
import seraplot as sp, numpy as np
X = (np.random.randn(400, 6) > 0).astype(float)
y = (X[:, 0] & X[:, 1]).astype(int)
bnb = sp.BernoulliNB(alpha=1.0)
bnb.fit(X, y)
print(bnb.score(X, y))

Référence API

Nom de fonction JSON

ml_bernoulli_nb — alias : bernoulli_nb

Classe Python
sp.BernoulliNb(alpha=1.0, binarize=0.0)
Paramètres du constructeur
ParamètreTypeDéfautDescription
alphafloat1.0Lissage additif (Laplace).
binarizefloat0.0Seuil pour binariser les features.
Retourne

JSON avec predictions.

Algorithme

$$P(x_j|y=c) = p_{cj}^{x_j}(1-p_{cj})^{1-x_j}$$

Exemple
import seraplot as sp, numpy as np
X = (np.random.randn(400, 6) > 0).astype(float)
y = (X[:, 0] & X[:, 1]).astype(int)
bnb = sp.BernoulliNB(alpha=1.0)
bnb.fit(X, y)
print(bnb.score(X, y))