Training Performance
Our model was trained for 10 epochs with detailed performance metrics tracked throughout the training process.
Training and Validation Accuracy
Training and Validation Loss
Confusion Matrix
Training Configuration
- Optimizer: Adam
- Loss Function: Categorical Crossentropy
- Batch Size: 32
- Epochs: 10
Training Results
| Epoch | Training Loss | Training Accuracy | Validation Loss | Validation Accuracy |
|---|---|---|---|---|
| 1 | 0.8655 | 88.98% | 0.6782 | 57.57% |
| 2 | 0.1696 | 92.87% | 1.2812 | 29.26% |
| 3 | 0.1261 | 95.03% | 1.3782 | 38.20% |
| 4 | 0.0902 | 96.63% | 3.0074 | 21.79% |
| 5 | 0.0537 | 98.09% | 2.4005 | 23.74% |
| 6 | 0.0393 | 98.77% | 5.1723 | 22.66% |
| 7 | 0.0264 | 99.11% | 5.7586 | 25.73% |
| 8 | 0.0893 | 96.78% | 6.3777 | 28.41% |
| 9 | 0.0281 | 99.15% | 7.9073 | 18.74% |
| 10 | 0.0263 | 99.13% | 8.0507 | 20.47% |
Analysis
The training results show that while the model achieves high training accuracy (99.13%), there are signs of overfitting as indicated by the increasing validation loss and decreasing validation accuracy. This suggests that while the model performs well on the training data, it may struggle with generalization to new, unseen data.
Evaluation Process
- Model evaluation on separate test dataset
- Confusion matrix visualization for detailed performance analysis
- Test dataset located in ./TestEvaluation directory
- Evaluation metrics include accuracy, precision, and recall