Krishikarm

AI-powered farming intelligence platform combining satellite data with voice AI.

Krishikarm provides personalized agricultural advice to Indian farmers through real-time satellite analysis, multilingual voice interaction, and live market prices. The platform is built around KisanNet v3, a cross-attention transformer trained on 82,000+ real satellite observations across 127 Indian districts with 96.8% accuracy.

---

Features

KisanNet v3 (Trained Model)

• Cross-attention transformer with 42 input features
• Trained on real satellite data from NASA POWER, Open-Meteo, and SoilGrids
• Predicts crop distress score, intervention window, and 5-level risk classification
• Inference time: ~4ms on GPU
• Available as a standalone Python package: pip install krishikarm

Voice-First Interface

• Speech-to-text and text-to-speech powered by Sarvam AI
• Supports 11 Indian languages: Hindi, Kannada, Telugu, Tamil, Marathi, Bengali, Gujarati, Punjabi, Odia, Malayalam, and English
• Designed for accessibility in rural areas

Satellite Intelligence

• Real-time data fusion from NASA GIBS, ISRO Bhuvan, Open-Meteo, and SoilGrids
• Visual map layers: NDVI (crop health), soil moisture, temperature, true color

Krishikarm Buddy

• Context-aware AI assistant for farming queries
• Knows your farm location, soil type, and weather forecast
• WhatsApp integration for sending reports and advisories

Mandi Marketplace

• Real-time commodity prices from data.gov.in (AgMarkNet)
• Covers all major agricultural markets across India

---

Tech Stack

Layer	Technologies
Frontend	Vite, Vanilla JS, Leaflet, Chart.js, TensorFlow.js
Backend	FastAPI, Python, SQLite, Sarvam AI
AI/ML	PyTorch (training), ONNX (inference), Transformer architecture
Data	NASA POWER, ISRO Bhuvan, Open-Meteo, AgMarkNet, SoilGrids

---

Setup

Prerequisites

• Node.js (v18+)
• Python (v3.9+)

1. Clone

git clone https://github.com/varshinicb1/Krishikarm.git
cd Krishikarm

2. Backend

cd backend
python -m venv venv
.\venv\Scripts\activate        # Windows
# source venv/bin/activate     # Linux/Mac

pip install -r requirements.txt

cp .env.example .env
# Edit .env and set SARVAM_API_KEY and MANDI_API_KEY

python server.py

3. Frontend

npm install
npm run dev

The backend runs at http://localhost:8000, the frontend at http://localhost:5173.

---

Python Package

The trained model is also available as a standalone package:

pip install krishikarm

from krishikarm import Predictor

p = Predictor()
result = p.predict({
    "lat": 12.97, "lon": 77.59,
    "T2M": 32, "RH2M": 65, "PREC": 2.5,
    "state": "KA", "irrig": "borewell",
    "crops": ["rice"],
})
print(result["risk_label"])  # "Healthy"

See PKG_README.md for full API documentation.

---

Environment Variables

Variable	Description	Required
SARVAM_API_KEY	Sarvam AI key for voice/translation	Yes
MANDI_API_KEY	data.gov.in key for market prices	Yes

---

License

MIT. See LICENSE for details.