时间序列预测的基准模型RELM-RKELM-RESN-RSVR-matlab实现(含完整源码+数据集).zip

recurrent kernel reservoir machine with QPSO for time series prediction model run the algorithm in MATLAB using the following command: 1. run recurrent kernel reservoir machine [average_test_SMAPE,average_test_RMSE,average_test_MSE,average_test_NMSE,average_train_SMAPE,average_train_RMSE,average_train_MSE,average_train_NMSE,Training_RMSE,TrainingAccuracy_SMAPE, Training_MSE,Training_NMSE,Testing_RMSE,TestingAccuracy_SMAPE, Testing_MSE,Testing_NMSE,training_time, testing_time,result,time] = Recurrent_KELM_with_reservoir_computing_2('norlorenz.csv', 0, 'RBF_kernel',9.138,0.02) In this funtion, Recurrent_KELM_with_reservoir_computing_2(Data_File, Elm_Type, kernel_type, kernel_para,a) where, Data_File = 'the file of your data set'; Elm_Type = 0, which is for regression; kernel_type = the type of method; kernel_para = kernel parameter; a = leaking rate of reservoir; 2. Run recurrent kernel reservoir machine with QPSO 2.1, open file named 'qpso_one_max_iteration_loop' 2.2, fill in the specific data_file name for change data set in line 63 and line 146 before runing code; 3.3, direct run code;