Deep-Learning Prediction Model with Serial Two-Level Decomposition Based on Bayesian Optimization

Table 3

The indicators of different models in the same test set.

Model	RMSE	NRMSE	MAE	SMAPE	R
RNN [40]	905.4590	0.0790	677.6044	0.0460	0.9241
LSTM [41]	835.8678	0.0614	630.6519	0.0439	0.9381
GRU [42]	805.0688	0.0707	599.9143	0.0406	0.9471
STL-RNN	851.9837	0.0759	664.7142	0.0452	0.9387
STL-LSTM [43]	771.5973	0.0594	579.3349	0.0391	0.9452
STL-GRU [8]	747.3044	0.0560	554.4263	0.0381	0.9503
W-LSTM [44]	837.0609	0.0648	705.9575	0.0491	0.9343
The proposed method	676.6433	0.0572	486.0197	0.0328	0.9575