如何根据交叉验证分数进行预测？

Question

I am working on building a prediction model. I have managed to reach until getting the cross-validation scores. Now I have no idea how to continue. What function should I use to make predictions using cross-validation scores?

X = data.iloc[:,0:16]
Y = data.iloc[:,16]
validation_size = 0.20
seed = 7
X_train, X_validation, Y_train, Y_validation = model_selection.train_test_split(X, Y,
test_size=validation_size, random_state=seed)
models = [
    ('LR', LogisticRegression()),
    ('CART', DecisionTreeClassifier()),
    ('KNN', KNeighborsClassifier()),
    ('SVM', SVC())
]

results, names = [], []
for name, model in models:
    seed = 32
    scoring = 'accuracy'
    kfold = model_selection.KFold(n_splits=10, random_state=seed)
    cv_results = model_selection.cross_val_score(model, X_train, Y_train, cv=kfold, scoring=scoring)
    results.append(cv_results)
    names.append(name)
    msg = "%s: %f (%f)" % (name, cv_results.mean(), cv_results.std())
    print(msg)

Answer 1

Cross validation is mostly used as a more robust validation scheme to check if your model is performing well or not. After that you can train a model with the whole dataset after being satisfied with your cross validation score or you can use.

sklearn.model_selection.cross_val_predict

Which predicts cross validated estimates. You can check out the documentation for more information.