objectif

Sauvegarder et recharger un modèle entraîné.

code minimal

from sklearn.linear_model import LogisticRegression
from sklearn.datasets import load_iris
from joblib import dump, load
X, y = load_iris(return_X_y=True)
clf = LogisticRegression(max_iter=1000).fit(X, y)
dump(clf, "model.joblib")
m2 = load("model.joblib")
print(hasattr(m2, "predict"))

utilisation

from joblib import load
print(load("model.joblib").predict([[5,3,1,0]]).shape[0])

variante(s) utile(s)

from joblib import dump
dump({"meta":"ok"}, "meta.joblib")
print("ok")

notes

• Inclure la Pipeline complète (préprocess + modèle).