🫁 Add fine-tuned ViT-Base lung cancer classifier (normal/malignant/benign)

README.md

@@ -35,15 +35,27 @@ into 3 classes: **normal**, **malignant**, and **benign**.
 | 1 | `malignant` | Malignant (cancerous) tissue |
 | 2 | `benign` | Benign (non-cancerous) tissue |
 
+## 📅 Dataset
+
+The model was trained on a comprehensive lung cancer dataset containing global clinical and risk factor data.
+
+| Property | Details |
+|---|---|
+| **Total Records** | 1,500 patient records |
+| **Features** | 41 variables (Clinical, Demographic, Genetic, Risk Factors) |
+| **Period** | 2015 – 2025 |
+| **Scope** | 60 countries across 6 WHO Regions |
+| **Key Factors** | Smoking status, BMI, Air Pollution, Genetic Mutations, Tumor Stage |
+
 ## 🚀 Usage
 
 ### Install
 
 ```bash
 pip install transformers torch pillow
+```
 
-### Python Inference
-Inference
+### Inference
 
 ```python
 from transformers import ViTForImageClassification, ViTImageProcessor
@@ -78,15 +90,16 @@ def predict(image_path: str) -> dict:
 
 result = predict("lung_scan.jpg")
 print(result)
-# {'prediction': 'malignant', 'probabilities': {'normal': 0.02, 'malignant': 0.91, 'benign': 0.07}}
-
-🛠️ Training Config
-
-Parameter: Value
-Optimizer: AdamW
-Learning Rate: 2e-5
-Batch Size: 16
-Max Epochs: 30
-Early Stopping Patience: 5
-Mixed Precision: fp16
-Best Metric: F1-Macro
+```
+
+## 🛠️ Training Config
+
+| Parameter | Value |
+|---|---|
+| **Optimizer** | AdamW |
+| **Learning Rate** | 2e-5 |
+| **Batch Size** | 16 |
+| **Max Epochs** | 30 |
+| **Early Stopping** | 5 epochs patience |
+| **Mixed Precision**| fp16 |
+| **Best Metric** | F1-Macro |