sklearn的分层抽样

目的

比较sklearn的StratifiedShuffleSplit与train_test_split的区别

StratifiedShuffleSplit

                                    # 分层抽样 训练测试
                                    from sklearn.model_selection import StratifiedShuffleSplit
                                    sss = StratifiedShuffleSplit(n_splits=1, test_size=0.2, random_state=randoms)
                                  
    
                                    X=X_new3.copy()
                                    y=y_df.copy()
                                    for train_index, test_index in sss.split(X, y):  # 这里循环的次数由n_splits决定，前面指定的5
                                            #print("TRAIN:", train_index, "TEST:", test_index)
                                            X_train, X_test = X.iloc[train_index,:], X.iloc[test_index,:]
                                            y_train, y_test = y.iloc[train_index,:], y.iloc[test_index,:]
                                    #  分层抽样 训练验证集
                                    sss = StratifiedShuffleSplit(n_splits=1, test_size=0.25, random_state=randoms)
                                    X=X_train.copy()
                                    y=y_train.copy()
                                    for train_index, test_index in sss.split(X, y):  # 这里循环的次数由n_splits决定，前面指定的5
                                            #print("TRAIN:", train_index, "TEST:", test_index)
                                            X_train, X_validate = X.iloc[train_index,:], X.iloc[test_index,:]
                                            y_train, y_validate = y.iloc[train_index,:], y.iloc[test_index,:]

train_test_split

X_train1, X_test1, y_train1, y_test1 = train_test_split(X_new3, y_df, test_size=0.2, random_state=8,stratify=y_df)

二者拆分的数据交集

二者拆分而成的样本一模一样