Research Article
Employing Machine Learning-Based Predictive Analytical Approaches to Classify Autism Spectrum Disorder Types
Figure 3
Impact of different hyperparameters on the performance of DNN. (a) Epochs and batch_size. (b) Optimization algorithms. (c) Learning rate and momentum. (d) Weight initialization. (e) Activation function. (f) Dropout regularization.
| (a) |
| (b) |
| (c) |
| (d) |
| (e) |
| (f) |