def stacking_first(train, train_y, test):

     savepath = './stack_op{}_dt{}_tfidf{}/'.format(args.option, args.data_type, args.tfidf)

     os.makedirs(savepath, exist_ok=True)

     count_kflod = 0

     num_folds = 6

     kf = KFold(n_splits=num_folds, shuffle=True, random_state=10)

     # 测试集上的预测结果

     predict = np.zeros((test.shape[0], config.n_class))

     # k折交叉验证集的预测结果

     oof_predict = np.zeros((train.shape[0], config.n_class))

     scores = []

     f1s = []

     for train_index, test_index in kf.split(train):

         # 训练集划分为6折，每一折都要走一遍。那么第一个是5份的训练集索引，第二个是1份的测试集，此处为验证集是索引

         kfold_X_train = {}

         kfold_X_valid = {}

         # 取数据的标签

         y_train, y_test = train_y[train_index], train_y[test_index]

         # 取数据

         kfold_X_train, kfold_X_valid = train[train_index], train[test_index]

         # 模型的前缀

         model_prefix = savepath + 'DNN' + str(count_kflod)

         if not os.path.exists(model_prefix):

             os.mkdir(model_prefix)

         M = 4  # number of snapshots

         alpha_zero = 1e-3  # initial learning rate

         snap_epoch = 16

         snapshot = SnapshotCallbackBuilder(snap_epoch, M, alpha_zero)

         # 使用训练集的size设定维度，fit一个模型出来

         res_model = get_model(train)

         res_model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])

         # res_model.fit(train_x, train_y, batch_size=BATCH_SIZE, epochs=EPOCH, verbose=1,  class_weight=class_weight)

         res_model.fit(kfold_X_train, y_train, batch_size=BATCH_SIZE, epochs=snap_epoch, verbose=1,

                       validation_data=(kfold_X_valid, y_test),

                       callbacks=snapshot.get_callbacks(model_save_place=model_prefix))

         # 找到这个目录下所有已经训练好的深度学习模型，通过".h5"

         evaluations = []

         for i in os.listdir(model_prefix):

             if '.h5' in i:

                 evaluations.append(i)

         # 给测试集和当前的验证集开辟空间，就是当前折的数据预测结果构建出这么多的数据集[数据个数，类别]

         preds1 = np.zeros((test.shape[0], config.n_class))

         preds2 = np.zeros((len(kfold_X_valid), config.n_class))

         # 遍历每一个模型，用他们分别预测当前折数的验证集和测试集，N个模型的结果求平均

         for run, i in enumerate(evaluations):

             res_model.load_weights(os.path.join(model_prefix, i))

             preds1 += res_model.predict(test, verbose=1) / len(evaluations)

             preds2 += res_model.predict(kfold_X_valid, batch_size=128) / len(evaluations)

         # 测试集上预测结果的加权平均

         predict += preds1 / num_folds

         # 每一折的预测结果放到对应折上的测试集中，用来最后构建训练集

         oof_predict[test_index] = preds2

         # 计算精度和F1

         accuracy = mb.cal_acc(oof_predict[test_index], np.argmax(y_test, axis=1))

         f1 = mb.cal_f_alpha(oof_predict[test_index], np.argmax(y_test, axis=1), n_out=config.n_class)

         print('the kflod cv is : ', str(accuracy))

         print('the kflod f1 is : ', str(f1))

         count_kflod += 1

         # 模型融合的预测结果，存起来，用以以后求平均值

         scores.append(accuracy)

         f1s.append(f1)

     # 指标均值，最为最后的预测结果

     print('total scores is ', np.mean(scores))

     print('total f1 is ', np.mean(f1s))

     return predict