🌱 PlantNet: Advanced Plant Disease Detection

Model Description

PlantNet is a state-of-the-art plant disease detection system designed for production deployment in agricultural applications. This model leverages ensemble deep learning techniques to achieve superior accuracy in identifying plant diseases from leaf images.

🎯 Key Features

• High Accuracy: 97.0% accuracy on PlantVillage test dataset
• Production Ready: Optimized for fast inference and real-world deployment
• Ensemble Architecture: Combines multiple state-of-the-art neural networks
• Multi-Format Support: Available in PyTorch, ONNX, and TorchScript formats
• Comprehensive Coverage: Supports 38 different plant disease classes
• Treatment Recommendations: Provides actionable disease management advice

🏗️ Model Architecture

This model uses an advanced ensemble approach combining:

• ResNet152 (30% weight) - Robust feature extraction with residual connections
• EfficientNet-B4 (25% weight) - Efficient scaling and performance optimization
• Vision Transformer (25% weight) - Attention-based global feature understanding
• Swin Transformer (20% weight) - Hierarchical vision processing

Total Parameters: 235,985,048

📊 Performance Metrics

🌾 Supported Plant Diseases

The model can identify diseases across multiple plant species:

Apple

• Apple Scab
• Black Rot
• Cedar Apple Rust
• Healthy

Corn (Maize)

• Cercospora Leaf Spot (Gray Leaf Spot)
• Common Rust
• Northern Leaf Blight
• Healthy

Tomato

• Bacterial Spot
• Early Blight
• Late Blight
• Leaf Mold
• Septoria Leaf Spot
• Spider Mites (Two-spotted)
• Target Spot
• Yellow Leaf Curl Virus
• Mosaic Virus
• Healthy

Additional Crops

• Cherry, Grape, Peach, Pepper, Potato, Strawberry, and more

🚀 Quick Start

Installation

pip install torch torchvision pillow huggingface-hub

Basic Usage

import torch
from PIL import Image
from torchvision import transforms

# Load model (replace with actual model loading code)
model = torch.hub.load('prof-freakenstein/plantnet-disease-detection', 'model', trust_repo=True)
model.eval()

# Image preprocessing
transform = transforms.Compose([
    transforms.Resize((384, 384)),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])

# Load and process image
image = Image.open('plant_leaf.jpg').convert('RGB')
input_tensor = transform(image).unsqueeze(0)

# Perform inference
with torch.no_grad():
    output = model(input_tensor)
    probabilities = torch.nn.functional.softmax(output, dim=1)
    predicted_class = torch.argmax(probabilities, dim=1)
    confidence = torch.max(probabilities, dim=1)[0]

print(f"Predicted Disease: {predicted_class.item()}")
print(f"Confidence: {confidence.item():.2%}")

Advanced Usage with Top-K Predictions

def predict_top_k(model, image_path, k=3):
    image = Image.open(image_path).convert('RGB')
    input_tensor = transform(image).unsqueeze(0)
    
    with torch.no_grad():
        output = model(input_tensor)
        probabilities = torch.nn.functional.softmax(output, dim=1)
        top_k_probs, top_k_indices = torch.topk(probabilities, k)
    
    results = []
    for i in range(k):
        results.append({
            'class_id': top_k_indices[0][i].item(),
            'confidence': top_k_probs[0][i].item(),
            'class_name': class_names[top_k_indices[0][i].item()]
        })
    
    return results

# Get top 3 predictions
predictions = predict_top_k(model, 'plant_image.jpg', k=3)
for pred in predictions:
    print(f"{pred['class_name']}: {pred['confidence']:.2%}")

🔧 Training Details

• Framework: PyTorch 2.4.1+rocm6.0
• Training Dataset: PlantVillage (79,000+ images)
• Epochs: 200
• Batch Size: 128
• Optimizer: adamw
• Learning Rate: 0.0003
• Mixed Precision: BFloat16 for efficient training
• Data Augmentation: MixUp, CutMix, AutoAugment, RandomErasing

💻 System Requirements

Minimum Requirements

• CPU: 4+ cores
• RAM: 8GB
• Storage: 3GB
• GPU: Optional (CPU inference supported)

Recommended for Optimal Performance

• GPU: 8GB+ VRAM (RTX 3080, V100, A100, MI300X)
• CPU: 8+ cores
• RAM: 16GB+
• Storage: SSD recommended

📈 Model Formats

This repository includes multiple optimized formats:

• pytorch_model.bin: Standard PyTorch model
• model.torchscript.pt: TorchScript for production deployment
• model.onnx: ONNX format for cross-platform compatibility
• model_quantized.pt: Quantized model for edge deployment

🔬 Research & Citation

If you use this model in your research, please cite:

@misc{plantnet2025,
  title={PlantNet: Ensemble Deep Learning for Plant Disease Detection},
  author={Anurag kumar Singh: singh.ask@outlook.com: anurag-bit@github.com},
  year={2025},
  publisher={Hugging Face},
  journal={Hugging Face Model Hub},
  howpublished={\url{https://huggingface.co/prof-freakenstein/plantnet-disease-detection}}
}

📄 License

This model is released under the MIT License. See LICENSE for details.

🤝 Contributing & Support

• Issues: Report bugs and request features on GitHub
• Discussions: Join the community discussions on this model page
• Email: Contact the team for commercial inquiries
• Documentation: Comprehensive guides available in the repository

🙏 Acknowledgments

• PlantVillage dataset creators and contributors
• PyTorch and Hugging Face teams for excellent frameworks
• Open source computer vision and agricultural AI communities
• Agricultural researchers and farmers providing domain expertise

Empowering sustainable agriculture through AI-powered plant disease detection 🌱🔬🚀

Model Version: 1.0.0 Upload Date: 2025-09-19 Framework: PyTorch 2.4.1+rocm6.0