gradient boosting d'arbre

[AdrianPaulCarrieres]

Je crois que ça va un peu de paire avec #12 mais pas sûr. Pareil que les autres modèles, il faudra expliquer pourquoi on veut essayer ce modèle #11

[AmineADD]

Which models are promising?

I.Random Forest

from sklearn.ensemble import RandomForestClassifier
from sklearn import metrics 

classifierRandomForest = RandomForestClassifier(max_depth=2,n_estimators=2,random_state=100,criterion='entropy')

def calculate_accuracy(classifier , X_train, X_test , y_train , y_test , modelName ):
    classifier.fit(X_train,y_train)
    y_pred_train = classifier.predict(X_train)
    y_pred_test = classifier.predict(X_test)
    accuracy_train = metrics.accuracy_score(y_train,y_pred_train)
    accuracy_test = metrics.accuracy_score(y_test,y_pred_test)
    print(modelName , 'Train accuracy:','{:.3f}'.format(accuracy_train),'Test accuracy','{:.3f}'.format(accuracy_test))
    return accuracy_train,accuracy_test,classifier

accuracy_train, accuracy_test , trained_classifier = calculate_accuracy(classifierRandomForest,X_train,X_test,y_train,y_test,modelName="Random Forest")
metrics.ConfusionMatrixDisplay.from_estimator(trained_classifier,X_test,y_test)