Table 2 Benchmarking results of different models on the ISO-NE dataset.

From: A comparative evaluation of gradient-based optimization algorithms for short-term load forecasting using deep residual networks

Model

MAPE

MAE

MSE

RMSE

NMSE

R

R2

CNN

0.024224

0.015815

0.000595

0.024393

0.039482

0.980696

0.960518

LSTM

0.023277

0.015163

0.000542

0.023283

0.035968

0.982602

0.964032

Transformer

0.021925

0.014352

0.000517

0.022749

0.034337

0.982763

0.965663

Transformer-LSTM

0.017280

0.011185

0.000311

0.017624

0.020609

0.989707

0.979391

CNN-LSTM-MMA

0.019086

0.012157

0.000348

0.018645

0.023067

0.988523

0.976933

DRN

0.017182

0.011138

0.000308

0.017548

0.020432

0.989767

0.979568

Back to article page