🌱 Crop Recommendation System

📌 Overview

This project leverages Machine Learning to recommend the most suitable crop to grow based on key agricultural factors:

• Soil nutrients: Nitrogen (N), Phosphorus (P), Potassium (K)
• Environmental conditions: pH, Rainfall, Temperature, Humidity

The model was trained and evaluated on the Crop Recommendation Dataset and deployed as an API using Flask on Render.

---

✨ Features

• ✅ Data preprocessing with Label Encoding and StandardScaler
• ✅ Multiple ML models tested (Decision Tree, Random Forest, Logistic Regression, etc.)
• ✅ XGBoost chosen as the best-performing model
• ✅ Saved trained model + scaler + label encoder for reproducibility
• ✅ REST API built with Flask for deployment
• ✅ Hosted live on Render

---

📂 Project Structure

''' ├── notebooks/ │ └── crop_recommendation.ipynb # Training notebook ├── models/ │ ├── best_model_xgboost.pkl # Saved trained model (XGBoost) │ ├── scaler.pkl # StandardScaler object │ └── label_encoder.pkl # LabelEncoder object ├── app.py # Flask API for deployment ├── requirements.txt # Dependencies ├── README.md # Project documentation '''

---

🚀 Live API

• Base URL: https://crop-recommendation-2-u8wx.onrender.com
• Health Check: / → Returns a JSON welcome message
• Prediction Endpoint: /predict

🔧 Usage

Endpoint: /predict

• Method: POST
• Headers:
  Content-Type: application/json

Example Request

{
  "N": 90,
  "P": 42,
  "K": 43,
  "temperature": 20.5,
  "humidity": 80,
  "ph": 6.5,
  "rainfall": 200
}