Short-Term Daily Univariate Streamflow Forecasting Using Deep Learning Models

Table 3

Keras tuner optimized hyperparameter values for Gummera station with its MSE score.

Hyperparameters	MLP				S-LSTM				Bi-LSTM				GRU
	T + 1	T + 2	T + 3	T + 4	T + 1	T + 2	T + 3	T + 4	T + 1	T + 2	T + 3	T + 4	T + 1	T + 2	T + 3	T + 4
Layer_1_units	25	25	10	25	20	30	15	25	10	25	15	15	5	40	25	15
Dropout_1	0.1	0.0	0.0	0.0	0.0	0.0	0.2	0.2	0.0	0.0	0.1	0.1	0.0	0.0	0.0	0.2
Layer_2_units	15	40	30	25	30	10	35	10	5	20	10	20	30	20	25	15
Learning rate	0.0001	0.0001	0.0001	0.001	0.001	0.001	0.01	0.001	0.001	0.001	0.01	0.001	0.01	0.01	0.0001	0.001
Number of epochs	70	90	80	50	60	50	50	80	60	90	50	30	80	40	50	30
Number of batch sizes	60	60	70	60	80	40	30	50	50	40	20	10	20	30	50	40
Score (MSE)	0.019	0.011	0.012	0.015	0.019	0.011	0.013	0.015	0.018	0.011	0.013	0.013	0.018	0.012	0.013	0.016