Table 3 The hyper-parameter values in this study.

From: An attention-based multi-residual and BiLSTM architecture for early diagnosis of autism spectrum disorder

Hyper-parameters

Setting

Learning rate (lr)

0.001

Optimizer

\(Adamax(lr=0.001, beta_1=0.9,\)

\(beta_2=0.999, epsilon=1e-08)\)

Kernal size

5

\(Num\_layers\)

4

\(D\_model\) (dimensionality of vectors)

64

\(Num\_units\)

64

\(Batch\_size\)

32

Epochs

50