Update README.md

README.md

@@ -42,8 +42,8 @@ During training, the validation accuracy steadily increased, indicating that the
 ### Test Performance
 After training, the model's performance was evaluated on the test dataset. The test logs report the following metrics:
 
-- Test Accuracy: 80.83%
-- Test Loss: 0.846
+- Test Accuracy: **80.83%**
+- Test Loss: **0.846**
 The test accuracy and loss are consistent with the validation metrics, indicating that the model generalizes well to unseen data. This performance demonstrates the effectiveness of the ResNet architecture for image classification tasks on the CIFAR-10 dataset.
 
 The model achieved competitive performance, achieving over 80% accuracy on both the validation and test sets. These results validate the effectiveness of the ResNet architecture and the training strategy employed in this project.