Skip to main content

Table 5 Final LSTM parameters

From: Load forecasting for energy communities: a novel LSTM-XGBoost hybrid model based on smart meter data

Parameter Value
Optimizer Adam
Time steps per day K 96
Past input timesteps as multiple of K 1
Batch size 64
Epochs 75
Steps per epoch 200
Learning rate 0.001
Number of units in LSTM-layer 200
Number of units in dense layers 130