Open Access
Table 4
Hyperparameter details, dataset statistics, and cross-validation results.
| Hyperparameter details |
|
|---|---|
| Particulars | Description |
| Number of layers | Total layers (input, hidden, output) |
| Number of neurons per layer | Input layer: 02 neurons |
| Hidden layer: varied for optimum network (06 neurons) | |
| Output layer: 05 neurons | |
| Activation functions | Hyperbolic tangent for hidden layer |
| Linear function for output layer | |
| Training algorithm | Levenberg-Marquardt |
| Iterations | 50 |
| Learning rate | 0.05 |
| Epochs | 300 |
| Loss function | Measures stop training when mean squared error (MSE) reaches 10−5 |
| Batch size | Number of samples per update |
| Regularization | Prevents overfitting |
| Weight initialization | Random |
| Dataset statistics |
|
| Particulars | Description |
| Number of samples | Total number of data points (1000) |
| Number of features | Inputs(2) and Outputs(5) |
| Feature range | Min and Max of input/output values |
| Data extraction | Extract inputs/outputs |
| Data normalization | Zero mean and unit variance. |
| Data splitting | Random (70% Training set, 15% Validation set, 15% Test set |
| Cross-validation results |
|
| Particulars | Description |
| Training performance | Minimum training error (0.1899) |
| Validation performance | Minimum validation error (0.1919) |
| Test performance | Minimum test error (0.126) |
| Goodness of fit | Correlation coefficient (R) |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.