Table 3 Hyperparameter tuned for the modified models.
Sr. no | Hyperparameter | CNN | VGG16 | ResNet50 | CNNAlexnet | InceptionV3 |
|---|---|---|---|---|---|---|
1. | Activation Layer | ReLu, Softmax | ReLu, Softmax | ReLu, Softmax | ReLu, Softmax | ReLu, Softmax |
2. | Dense Layer | 128 | 1024 | 1024 | 4096 | 1024 |
3. | Dropout | 0.5 | 0.2 | 0.2 | 0.5 | 0.2 |
4. | Learning Rate | 0.0001 | 0.0001 | 0.0001 | 0.0001 | 0.0001 |
5. | Batch Size | 156 | 156 | 156 | 156 | 156 |
6. | Optimizer | Adam | Adam | Adam | Adam | Adam |
7. | Pooling | MaxPooling2D | – | – | MaxPooling2D | – |
8. | Epochs | 25 | 25 | 25 | 25 | 25 |
9. | Loss | Sparse Categorical Cross Entropy | Sparse Categorical Cross Entropy | Sparse Categorical Cross Entropy | Sparse Categorical Cross Entropy | Sparse Categorical Cross Entropy |
10. | Weights | – | Imagenet | Imagenet | – | Imagenet |
