A Novel Hybrid Deep Learning Model for Sugar Price Forecasting Based on Time Series Decomposition

Table 5

Optimal hyperparameters for the TPE-EMD-DNN model.

Hyperparameter	Value
	39
	LeakyReLU
	186
	LeakyReLU
Dropout rate	0.2
Optimizer	Rmsprop
Batch size	32
Learning rate	0.01