Time-Series Forecasting of a Typical PWR Undergoing Large Break LOCA

Table 3

Dictionary of hyperparameters.

Hyperparameters	Search space
Number of neurons in 1^st layer	13, 25, 50
Number of neurons in a final layer	1
Number of hidden layers	1, 2, 3
Optimizer	Adam, Nadam, SGD, RMSprop
Activation functions	ReLU, Tanh, Sigmoid, Softmax
Recurrent activation functions	Sigmoid, ReLU, Tanh
Dropout	0, 0.1, 0.2
Batch size	64, 100, 200
Number of iterations	15, 20, 35, 40
Kernel regularizers l1	1 × 10^−4, 1 × 10^−5, 1 × 10^−6