objectif

Apprentissage incrémental par mini-batchs avec partial_fit.

code minimal

from sklearn.linear_model import SGDClassifier
from sklearn.datasets import make_classification
import numpy as np

X, y = make_classification(n_samples=200, n_features=20, random_state=0)
clf = SGDClassifier(random_state=0)
classes = np.unique(y)
for i in range(0, 200, 50):
    clf.partial_fit(X[i:i+50], y[i:i+50], classes=classes)
print(hasattr(clf, "coef_"))

utilisation

print(clf.predict(X[:5]).tolist())

variante(s) utile(s)

from sklearn.linear_model import SGDRegressor
print(hasattr(SGDRegressor(), "partial_fit"))

notes

• Standardiser les features; régler learning_rate/alpha.