Journal of Big Data

Table 1 Hyperparameter space used during the grid search

From: A new deep learning architecture with inductive bias balance for transformer oil temperature forecasting

Hyperparameter	Values
Window size	24, 48, 72, 96, 120
Embedding size	8, 16
# Neurons for each layer	8, 16, 32, 64
# Layers	1, 2, 3
Kernel size for each layer	3, 5, 7

Back to article page