Assembling your first ensemble
It's time to build your first ensemble model! The Pokémon dataset from the previous exercise has been loaded and split into train and test sets.
Your job is to leverage the voting ensemble technique using the sklearn API. It's up to you to instantiate the individual models and pass them as parameters to build your first voting classifier.
Este exercício faz parte do curso
Ensemble Methods in Python
Instruções do exercício
- Instantiate a
KNeighborsClassifiercalledclf_knnwith 5 neighbors (specified usingn_neighbors). - Instantiate a
"balanced"LogisticRegressioncalledclf_lr(specified usingclass_weight). - Instantiate a
DecisionTreeClassifiercalledclf_dtwithmin_samples_leaf = 3andmin_samples_split = 9. - Build a
VotingClassifierusing the parameterestimatorsto specify the following list of (str, estimator) tuples:'knn',clf_knn,'lr',clf_lr, and'dt',clf_dt.
Exercício interativo prático
Experimente este exercício completando este código de exemplo.
# Instantiate the individual models
clf_knn = ____
clf_lr = ____
clf_dt = ____(____, ____, random_state=500)
# Create and fit the voting classifier
clf_vote = ____(
estimators=[('____', ____), ('____', ____), ('____', ____)]
)
clf_vote.fit(X_train, y_train)