Update README.md

README.md

@@ -1,3 +1,51 @@
+---
+# Model Card for Hybrid ViT - Diabetic Retinopathy Classification
+
+library_name: pytorch
+tags:
+  - classification
+  - diabetic-retinopathy
+  - hybrid vit
+  - efficientnet
+  - deep-learning
+  - medical-ai
+model_name: Hybrid_ViT
+model_description: "A Hybrid Vision Transformer (ViT) with EfficientNetB0 as the backbone, trained to classify Diabetic Retinopathy severity into 5 categories."
+dataset:
+  name: "APTOS 2019 Blindness Detection"
+  url: "https://www.kaggle.com/competitions/aptos2019-blindness-detection"
+  description: "A dataset containing retinal fundus images for classifying Diabetic Retinopathy severity."
+
+metrics:
+  accuracy: 85.4
+  loss: 0.23
+  validation_accuracy: 84.5
+  validation_loss: 0.25
+
+training:
+  epochs: 5
+  batch_size: 32
+  optimizer: Adam
+  learning_rate: 1e-4
+  scheduler: 
+    type: StepLR
+    step_size: 10
+    gamma: 0.5
+  loss_function: CrossEntropyLoss
+  k_folds: 5
+  device: "CUDA"
+model_architecture:
+  backbone: EfficientNetB0
+  transformer: Vision Transformer (ViT)
+  num_classes: 5
+  input_size: 224x224
+  framework: PyTorch
+
+
+author: "Pavan Kumar Ambadapudi>"
+repository: "https://huggingface.co/PavanKumarAmbadapudi/DiabeticRetinopathy_Hybrid-ViT"
+
+---
 
 
 # 🏥 Diabetic Retinopathy Severity Classification
@@ -82,21 +130,6 @@ def predict_model_Hybrid(model, image_tensor):
         return model_predictions
 print("Hybrid ViT ", predict_model_Hybrid(model, image_tensor))
 ```
-## 📊 Training Details
-This model was trained on **APTOS 2019 Blindness Detection** dataset using **5-Fold Cross-Validation** to ensure better generalization. The training process involved EfficientNet B0 as a feature extractor combined with a Vision Transformer (ViT) classification head.
-
-## 🛠️ Hyperparameters
-| Parameter     | Value |
-|--------------|-------|
-| **Image Size**  | 224x224 |
-| **Batch Size**  | 32 |
-| **Epochs**      | 5 |
-| **K-Folds**     | 5 |
-| **Learning Rate** | 1e-4 |
-| **Optimizer**  | Adam |
-| **Scheduler**  | StepLR (Step=10, Gamma=0.5) |
-| **Loss Function** | CrossEntropyLoss |
-| **Device** | `CUDA` (if available) |
 
 ## 📬 Contact
 For any queries, reach out to me at: