Table 1 Effect of the number of neurons in the hidden layer on the S-BiLSTM model.
From: Application of stacked bidirectional LSTM neural networks in reservoir porosity prediction
Number of neurons in the hidden layer | MAE | RMSE | Predictive accuracy (%) |
|---|---|---|---|
32 | 0.0138 | 0.0275 | 62.19 |
64 | 0.0121 | 0.0257 | 67.34 |
96 | 0.0097 | 0.023 | 73.76 |
112 | 0.0073 | 0.0201 | 80.04 |
120 | 0.0048 | 0.0163 | 86.88 |
124 | 0.0042 | 0.0151 | 88.49 |
126 | 0.0031 | 0.013 | 91.51 |
128 | 0.0027 | 0.0121 | 92.75 |
130 | 0.0028 | 0.0123 | 92.38 |
132 | 0.0030 | 0.0128 | 92.1 |
136 | 0.0031 | 0.0132 | 91.49 |
144 | 0.0032 | 0.0133 | 91.06 |
160 | 0.0058 | 0.0179 | 84.31 |
256 | 0.0073 | 0.0201 | 80.27 |
512 | 0.008 | 0.021 | 78.12 |
