Upload 4 files

README.md

@@ -1,12 +1,69 @@
 ---
-title: Swapnith
-emoji: 💻
-colorFrom: pink
-colorTo: pink
+title: Plant Disease Detection
+emoji: 🌱
+colorFrom: green
+colorTo: blue
 sdk: gradio
-sdk_version: 6.1.0
+sdk_version: 4.44.0
 app_file: app.py
 pinned: false
+license: mit
 ---
 
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+# 🌱 Plant Disease Detection AI
+
+An AI-powered plant disease detection system that can identify diseases in various crops including Chilli, Pepper Bell, Potato, Tomato, and GroundNut plants.
+
+## Features
+
+- **19 Disease Classes**: Comprehensive detection across multiple plant types
+- **TinyViT Model**: Efficient Vision Transformer for fast inference
+- **Treatment Recommendations**: Detailed information about causes, symptoms, and treatments
+- **User-Friendly Interface**: Simple drag-and-drop image upload
+
+## Supported Plants & Diseases
+
+### 🌶️ Chilli
+- Healthy
+- Leaf Curl Virus
+
+### 🫑 Pepper Bell
+- Healthy
+- Bacterial Spot
+
+### 🥔 Potato
+- Healthy
+- Early Blight
+- Late Blight
+
+### 🍅 Tomato
+- Healthy
+- Bacterial Spot
+- Early Blight
+- Late Blight
+- Leaf Mold
+- Mosaic Virus
+- Septoria Leaf Spot
+- Target Spot
+- Two Spotted Spider Mite
+- Yellow Leaf Curl Virus
+
+### 🥜 GroundNut
+- Healthy
+- Rust
+
+## How to Use
+
+1. Upload an image of a plant leaf
+2. Click "Analyze Plant Disease"
+3. Get instant results with:
+   - Disease identification
+   - Confidence score
+   - Treatment recommendations
+
+## Model Details
+
+- **Architecture**: TinyViT (WinKawaks/vit-tiny-patch16-224)
+- **Parameters**: ~5.5M parameters
+- **Input Size**: 224x224 RGB images
+- **Training**: Knowledge distillation from larger teacher model

app.py

@@ -0,0 +1,235 @@
+import gradio as gr
+import torch
+import torch.nn as nn
+from torchvision import transforms
+from PIL import Image
+import os
+from transformers import ViTForImageClassification
+
+# Disease class mappings - 19 classes from your trained model
+DISEASE_CLASSES = [
+    "Chilli - Healthy",                 # 00
+    "Chilli - Leaf Curl Virus",          # 01
+    "Pepper Bell - Bacterial Spot",      # 02
+    "Pepper Bell - Healthy",             # 03
+    "Potato - Early Blight",             # 04
+    "Potato - Healthy",                  # 05
+    "Potato - Late Blight",              # 06
+    "Tomato - Bacterial Spot",           # 07
+    "Tomato - Early Blight",             # 08
+    "Tomato - Healthy",                  # 09
+    "Tomato - Late Blight",              # 10
+    "Tomato - Leaf Mold",                # 11
+    "Tomato - Mosaic Virus",             # 12
+    "Tomato - Septoria Leaf Spot",       # 13
+    "Tomato - Target Spot",              # 14
+    "Tomato - Two Spotted Spider Mite",  # 15
+    "Tomato - Yellow Leaf Curl Virus",   # 16
+    "GroundNut - Healthy",               # 17
+    "GroundNut - Rust"                   # 18
+]
+
+# Disease information database (shortened for demo)
+DISEASE_INFO = {
+    "Chilli - Healthy": {
+        "description": "The chilli plant appears healthy with no visible signs of disease.",
+        "treatment": "Continue good agricultural practices and regular monitoring."
+    },
+    "Chilli - Leaf Curl Virus": {
+        "description": "Leaf curl virus causes leaves to curl, wrinkle, and become distorted.",
+        "treatment": "Remove infected plants, control whiteflies with neem oil, use yellow sticky traps."
+    },
+    "Tomato - Early Blight": {
+        "description": "Early blight causes characteristic target-spot patterns on older leaves.",
+        "treatment": "Apply fungicides (chlorothalonil, mancozeb), remove infected leaves, improve air circulation."
+    },
+    "Potato - Late Blight": {
+        "description": "Late blight is a devastating disease that can destroy entire crops rapidly.",
+        "treatment": "Apply systemic fungicides immediately, destroy infected plants, improve air circulation."
+    },
+    # Add more as needed...
+}
+
+# Global variables
+model = None
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+# Image preprocessing
+transform = transforms.Compose([
+    transforms.Resize((224, 224)),
+    transforms.ToTensor(),
+    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
+])
+
+def load_model():
+    """Load the TinyViT student model"""
+    global model
+    
+    # Look for model file
+    model_paths = [
+        "best_student_attention_kd.pth",
+        "model/best_student_attention_kd.pth"
+    ]
+    
+    model_path = None
+    for path in model_paths:
+        if os.path.exists(path):
+            model_path = path
+            break
+    
+    if model_path is None:
+        raise FileNotFoundError("Model file not found")
+    
+    try:
+        print("Loading TinyViT student model...")
+        
+        # Initialize TinyViT model architecture
+        model = ViTForImageClassification.from_pretrained(
+            "WinKawaks/vit-tiny-patch16-224",
+            num_labels=len(DISEASE_CLASSES),
+            ignore_mismatched_sizes=True
+        )
+        
+        # Load trained weights
+        checkpoint = torch.load(model_path, map_location=device)
+        
+        if isinstance(checkpoint, dict) and 'model_state_dict' in checkpoint:
+            model.load_state_dict(checkpoint['model_state_dict'])
+        else:
+            model.load_state_dict(checkpoint)
+        
+        model.to(device)
+        model.eval()
+        
+        print(f"✓ Model loaded successfully on {device}")
+        return True
+        
+    except Exception as e:
+        print(f"Error loading model: {e}")
+        return False
+
+def predict_disease(image):
+    """Predict plant disease from image"""
+    if model is None:
+        return "❌ Model not loaded", "", ""
+    
+    try:
+        # Preprocess image
+        if image is None:
+            return "❌ No image provided", "", ""
+        
+        # Convert to RGB if needed
+        if image.mode != 'RGB':
+            image = image.convert('RGB')
+        
+        image_tensor = transform(image).unsqueeze(0).to(device)
+        
+        # Make prediction
+        with torch.no_grad():
+            outputs = model(image_tensor)
+            logits = outputs.logits
+            probabilities = torch.nn.functional.softmax(logits, dim=1)
+            confidence, predicted_idx = torch.max(probabilities, 1)
+            
+            predicted_class = DISEASE_CLASSES[predicted_idx.item()]
+            confidence_score = confidence.item()
+        
+        # Parse results
+        parts = predicted_class.split(" - ")
+        crop_type = parts[0] if len(parts) > 0 else "Unknown"
+        disease = parts[1] if len(parts) > 1 else predicted_class
+        status = "🟢 Healthy" if "healthy" in disease.lower() else "��� Diseased"
+        
+        # Get disease info
+        disease_info = DISEASE_INFO.get(predicted_class, {
+            "description": f"Information about {disease}.",
+            "treatment": "Consult with a plant pathologist or agricultural extension service."
+        })
+        
+        # Format results
+        result = f"""
+## 🌱 **Crop Type:** {crop_type}
+## 🦠 **Disease:** {disease}
+## 📊 **Status:** {status}
+## 🎯 **Confidence:** {confidence_score:.2%}
+
+### 📝 **Description:**
+{disease_info['description']}
+
+### 💊 **Treatment:**
+{disease_info['treatment']}
+        """
+        
+        return result, predicted_class, f"Confidence: {confidence_score:.2%}"
+        
+    except Exception as e:
+        return f"❌ Error processing image: {str(e)}", "", ""
+
+# Load model on startup
+print("Initializing Plant Disease Detection Model...")
+model_loaded = load_model()
+
+if not model_loaded:
+    print("⚠️ Model failed to load - running in demo mode")
+
+# Create Gradio interface
+with gr.Blocks(title="🌱 Plant Disease Detection", theme=gr.themes.Soft()) as demo:
+    gr.Markdown("""
+    # 🌱 Plant Disease Detection AI
+    
+    Upload an image of a plant leaf to detect diseases and get treatment recommendations.
+    
+    **Supported Plants:** Chilli, Pepper Bell, Potato, Tomato, GroundNut
+    **Supported Diseases:** 19 different disease classes including healthy plants
+    """)
+    
+    with gr.Row():
+        with gr.Column():
+            image_input = gr.Image(
+                type="pil",
+                label="📸 Upload Plant Image",
+                height=400
+            )
+            
+            predict_btn = gr.Button(
+                "🔍 Analyze Plant Disease",
+                variant="primary",
+                size="lg"
+            )
+            
+        with gr.Column():
+            result_output = gr.Markdown(
+                label="📋 Analysis Results",
+                value="Upload an image and click 'Analyze' to get started!"
+            )
+    
+    with gr.Row():
+        with gr.Column():
+            predicted_class_output = gr.Textbox(
+                label="🏷️ Predicted Class",
+                interactive=False
+            )
+        with gr.Column():
+            confidence_output = gr.Textbox(
+                label="📊 Confidence Score",
+                interactive=False
+            )
+    
+    # Example images (you can add these)
+    gr.Markdown("### 📸 Example Images")
+    gr.Markdown("Try uploading images of plant leaves with various diseases or healthy plants.")
+    
+    # Connect the prediction function
+    predict_btn.click(
+        fn=predict_disease,
+        inputs=[image_input],
+        outputs=[result_output, predicted_class_output, confidence_output]
+    )
+
+# Launch the app
+if __name__ == "__main__":
+    demo.launch(
+        server_name="0.0.0.0",
+        server_port=7860,
+        share=False
+    )

best_student_attention_kd.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:64435d4efb8325520d8a7702176008344447566e9380d76458e6e68d2651d86d
+size 22196075

requirements.txt

@@ -0,0 +1,5 @@
+gradio==4.44.0
+torch==2.1.0
+torchvision==0.16.0
+transformers==4.35.2
+Pillow==10.1.0