Update README.md

README.md

@@ -1,17 +1,24 @@
 ---
-title: "Summ Small: Medical Dialogue to SOAP Summarizer"
-emoji: "📄"
-colorFrom: "green"
-colorTo: "pink"
-sdk: "static"
+title: 'Summ Small: Medical Dialogue to SOAP Summarizer'
+emoji: 📄
+colorFrom: green
+colorTo: pink
+sdk: static
 pinned: false
+license: mit
+datasets:
+- omi-health/medical-dialogue-to-soap-summary
+language:
+- en
+metrics:
+- rouge
 ---
 
 # Model Card for Summ (3B) Small
 
 ## Model Description
 
-Summ Small is a powerful language model specifically designed to generate SOAP summaries from medical dialogues. It is a fine-tuned version of the Microsoft/Phi-3-mini-4k-instruct model using the Omi Health/medical-dialogue-to-soap-summary dataset. Despite its 'small' label, which anticipates future larger versions, this model delivers superior performance compared to larger models like GPT-4.
+Summ Small is a powerful language model specifically designed to generate SOAP summaries from medical dialogues. It is a fine-tuned version of the [Microsoft/Phi-3-mini-4k-instruct](https://huggingface.co/microsoft/Phi-3-mini-4k-instruct) using the [Omi Health/medical-dialogue-to-soap-summary dataset](https://huggingface.co/datasets/omi-health/medical-dialogue-to-soap-summary). This model demonstrates superior performance compared to larger models like GPT-4.
 
 ## Intended Use
 
@@ -27,7 +34,18 @@ Training was conducted on NVIDIA A100 GPUs, ensuring efficient processing and mo
 
 ## Evaluation
 
-Preliminary evaluation using Rouge-1 metrics shows Summ Small achieving a score of 70, outperforming other models like GPT4Turbo (69), LLama3 8B Instruct (59), and GPT3.5 (54).
+The performance of Summ Small has been evaluated using several Rouge metrics as follows:
+
+| Model                    | ROUGE-1 | ROUGE-2 | ROUGE-L | ROUGE-LSUM |
+|--------------------------|---------|---------|---------|------------|
+| **Omi-Sum 3B Small**     | **70**  | **45**  | **53**  | **62**     |
+| GPT4Turbo                | 69      | 46      | 54      | 63         |
+| LLama3 8B Instruct       | 59      | 32      | 39      | 51         |
+| GPT3.5                   | 54      | 30      | 40      | 49         |
+| Phi-3 3B mini 4k instruct| 55      | 32      | 40      | 50         |
+| Phi2 basic               | 41      | 19      | 26      | 36         |
+
+These results showcase Summ Small's capabilities in generating accurate SOAP summaries compared to other leading models.
 
 ## Limitations
 
@@ -37,14 +55,10 @@ While Summ Small demonstrates promising results, the training data is completely
 
 The Summ Small model is released under the MIT License, which permits broad use with fewer restrictions, making it accessible for both commercial and non-commercial use.
 
-## How to Use
-
-For those interested in experimenting with or further developing Summ Small, access can be granted upon request, with terms aligning with usage rights and ethical considerations.
-
 ## Ethical Considerations
 
 Users are urged to consider the ethical implications of AI in healthcare and ensure that any deployment of such models prioritizes patient safety and data privacy.
 
 ## Contact
 
-For more information or to request access to Summ Small, please contact [info@omi.health](mailto:info@omi.health).
+For more information or to request access to Summ Small, please contact [info@omi.health](mailto:info@omi.health).