updated readme.md with training information and new test results

README.md

@@ -35,3 +35,46 @@ results:
       accuracy: 0.9761
       fraction: 3958/4055
 ---
+
+# Model Training Results
+
+## Run Summary
+
+| Metric | Value |
+|--------|-------|
+| **Best Model Epoch** | 1 |
+| **Best Validation Accuracy** | 99.605% |
+| **Final Best Accuracy** | 99.605% |
+| **Training Accuracy** | 99.829% |
+| **Training Loss** | 0.00525 |
+| **Validation Accuracy** | 99.605% |
+| **Validation Loss** | 0.01301 |
+| **Step** | 1800 |
+| **Step Accuracy** | 99.837% |
+| **Step Loss** | 0.00009 |
+
+## Test Results
+
+| Metric | Value |
+|--------|-------|
+| **Overall Accuracy** | 96.74% |
+| **Overall Accuracy Fraction** | 7861/8126 |
+| **Real Images Accuracy** | 97.25% (3959/4071) |
+| **Fake Images Accuracy** | 96.23% (3902/4055) |
+
+### Confusion Matrix
+
+| True Label | Predicted Label | Count |
+|------------|----------------|-------|
+| fake | fake | 3902 |
+| fake | real | 153 |
+| real | real | 3959 |
+| real | fake | 112 |
+
+## Key Achievements
+
+- **Exceptional Training Performance**: Achieved 99.6% validation accuracy in just 1 epoch
+- **Strong Test Performance**: 96.74% accuracy on test set with balanced performance across classes
+- **Fast Convergence**: Best model found at epoch 1, demonstrating effective fine-tuning
+- **Low Loss**: Final step loss of 9e-05 indicates excellent model convergence
+- **Robust Classification**: Real images (97.25%) and fake images (96.23%) both classified with high accuracy