模型融合stacking交叉验证 feature_names mismatch

在做模型融合的时候提示feature_names mismatch 实在是找不出哪里出错了  请大神帮忙看看

from sklearn.base import clone
def stacking(base_learners, meta_learner, X, y, generator):
    print("Fitting final base learners...", end="")
    train_base_learners(base_learners, X, y, verbose=False)
    print("done")
    print("Generating cross-validated predictions...")
    cv_preds, cv_y = [], []
    for i, (train_idx, test_idx) in enumerate(generator.split(X)):
        fold_xtrain, fold_ytrain = X[train_idx, :], y[train_idx]
        fold_xtest, fold_ytest = X[test_idx, :], y[test_idx]
        fold_base_learners = {name: clone(model)
                              for name, model in base_learners.items()}
        train_base_learners(
            fold_base_learners, fold_xtrain, fold_ytrain, verbose=False)
        fold_P_base = predict_base_learners(
            fold_base_learners, fold_xtest, verbose=False)
        cv_preds.append(fold_P_base)
        cv_y.append(fold_ytest)
        print("Fold %i done" % (i + 1))
    print("CV-predictions done")
    cv_preds = np.vstack(cv_preds)
    cv_y = np.hstack(cv_y)
    print("Fitting meta learner...", end="")
    meta_learner.fit(cv_preds, cv_y)
    print("done")
    return base_learners, meta_learner
from sklearn.model_selection import KFold
cv_base_learners, cv_meta_learner = stacking(
    get_models(), clone(meta_learner), xtrain.values, ytrain.values, KFold(2))
P_pred, p = ensemble_predict(cv_base_learners, cv_meta_learner, xtest, verbose=False)
print("\nEnsemble ROC-AUC score: %.3f" % roc_auc_score(ytest, p))

更适合发去python 论坛 。

你这个是R？