def fast_objective(
max_depth,
max_leaf,
l1,
l2,
min_samples_leaf,
learning_rate,
):
max_leaf = int(max_leaf)
max_depth = int(max_depth)
min_samples_leaf = int(min_samples_leaf)
assert type(max_leaf) == int
assert type(max_depth) == int
assert type(min_samples_leaf) == int
model = FastRGFClassifier(
max_leaf=max_leaf,
max_depth=max_depth,
l1=l1,
l2=l2,
min_samples_leaf=min_samples_leaf,
learning_rate=learning_rate,
)
model.fit(train_m, label_m)
pred_proba = model.predict_proba(train_val)
score = roc_auc_score(label_val, pred_proba[:, 1])
return score
def run_model(X_train, y_train, X_val, y_val, params):
l1 = params["l1"]
l2 = params["l2"]
learning_rate = params["learning_rate"]
max_leaf = int(params["max_leaf"])
max_depth = int(params["max_depth"])
min_samples_leaf = int(params["min_samples_leaf"])
model = FastRGFClassifier(
max_leaf=max_leaf,
max_depth=max_depth,
l1=l1,
l2=l2,
min_samples_leaf=min_samples_leaf,
learning_rate=learning_rate,
)
model.fit(X_train, y_train)
pred_proba = model.predict_proba(X_val)[:, 1]
score = roc_auc_score(y_val, pred_proba)
return pred_proba, score