Table 1.

Optimized hyperparameters applied to the models.

Hyperparameter	Search Space	Model 1 (CNN)	Model 2 (CNN-LSTM)
Number of filters in the first CNN layer	64, 128, 256	64	64
Filter length of the first CNN layer	3–10	5	5
Number of cells in the first LSTM layer	64, 128, 256	-	256
Batch size	32, 64, 128	64	32
Learning rate	0.001–0.00001	0.0001	0.0001

CNN: convolutional neural network, LSTM: long short-term memory.