A Comparative Study of AutoML Approaches for Short-Term Electric Load Forecasting

Deep learning is increasingly used in short-term load forecasting. However, deep models are difficult to train, and adjusting training hyper-parameters takes time effort. Automated machine (AutoML) can reduce human participation process improve the efficiency of modelling while ensuring accuracy prediction. In this paper, we compare usage three AutoML approaches The experiments on a real-world dataset show that predictive performance AutoGluon outperforms AutoPytorch Auto-Keras, according metrics: MAE, RMSE MAPE. Auto-Keras have similar not easy compare.