Table 4 RNN architecture using SimpleRNN neurons for CarboxE site identification.

From: Computational identification of 4-carboxyglutamate sites to supplement physiological studies using deep learning

No

Layer

No. of weights

1

Embedding layer

\((2320 = 460)\)

2

R1: Recurrent layer with 14 simple RNN units and dropout regularization with 20% probability

\((3514= 490)\)

3

Dense layer with 8 units

\((14+1)8 = 120\)

4

Output layer

\((8+1)1 = 9\)

Back to article page