Update README.md

README.md

@@ -42,7 +42,7 @@ A fine-tuned DeBERTa-v3 model for classifying marine incident types based on acc
 
 This model is a fine-tuned version of `microsoft/deberta-v3-base` specifically designed to classify marine incidents into 11 different categories. It was trained on the MAIB incident reports dataset and achieves high performance in maritime safety incident classification.
 
-- **Developed by**: MAIB Incident Type Classifier Team
+- **Developed by**: Ilia Munaev
 - **Model type**: Text Classification
 - **Language(s)**: English
 - **License**: Apache 2.0
@@ -58,21 +58,16 @@ The model achieves the following performance metrics on the test set:
 | **Weighted F1-Score** | 89.0% |
 | **Macro F1-Score** | 70.2% |
 
-### Per-Class Performance
-
-| Incident Type | Precision | Recall | F1-Score | Support |
-|---------------|-----------|--------|----------|---------|
-| Accident to person(s) | 0.97 | 0.96 | 0.96 | 67 |
-| Capsizing / Listing | 0.75 | 0.86 | 0.80 | 7 |
-| Collision | 0.76 | 0.87 | 0.81 | 15 |
-| Contact | 0.62 | 0.57 | 0.59 | 23 |
-| Damage / Loss Of Equipment | 0.94 | 0.92 | 0.93 | 72 |
-| Fire / Explosion | 0.95 | 0.95 | 0.95 | 20 |
-| Flooding / Foundering | 1.00 | 0.83 | 0.91 | 12 |
-| Grounding / Stranding | 0.90 | 0.90 | 0.90 | 31 |
-| Hull Failure | 0.00 | 0.00 | 0.00 | 0 |
-| Loss Of Control | 0.83 | 0.90 | 0.86 | 42 |
-| Non-accidental Event | 0.00 | 0.00 | 0.00 | 0 |
+
+## Evaluation Results
+
+The model evaluation:
+
+- **Confusion Matrix**: Shows classification accuracy across all incident types
+- **Per-Class F1 Scores**: Displays F1 performance for each incident category
+
+<img src="confusion_matrix.png" alt="Confusion Matrix" width="600">
+<img src="per_class_f1.png" alt="Per-Class F1 Scores" width="600">
 
 ## Intended Use
 
@@ -200,17 +195,6 @@ The model classifies incidents into the following 11 categories:
 9. **Loss Of Control** - Loss of steering or propulsion control
 10. **Non-accidental Event** - Events not classified as accidents
 
-## Evaluation Results
-
-The model evaluation includes comprehensive visualizations:
-
-- **Confusion Matrix**: Shows classification accuracy across all incident types
-- **Per-Class F1 Scores**: Displays F1 performance for each incident category
-
-![Confusion Matrix](confusion_matrix.png)
-
-![Per-Class F1 Scores](per_class_f1.png)
-
 ## Limitations and Bias
 
 ### Known Limitations
@@ -231,7 +215,7 @@ The model evaluation includes comprehensive visualizations:
 ```bibtex
 @software{maib_classifier,
   title={MAIB Incident Type Classifier},
-  author={MAIB Incident Type Classifier Team},
+  author={Ilia Munaev},
   year={2024},
   url={https://huggingface.co/your-username/maib-incident-classifier}
 }