Smart Irrigation Need Prediction System

A Machine Learning based web application that predicts irrigation requirements for crops using soil, weather, crop, and field conditions.

Live Demo

https://irrigation-need-4.onrender.com

---

Features

• Predicts irrigation need as:

  • Low
  • Medium
  • High

• Uses inputs such as:

  • Soil Type
  • Soil Moisture
  • Soil pH
  • Organic Carbon
  • Temperature
  • Humidity
  • Rainfall
  • Sunlight Hours
  • Crop Type
  • Growth Stage
  • Irrigation Type
  • Water Source
  • Region

• Mobile responsive web interface

• Flask-based deployment

• Random Forest Machine Learning model

• Preprocessing Pipeline using Scikit-learn

---

Machine Learning Workflow

Data Preprocessing

• Missing value handling using SimpleImputer
• Numerical feature scaling using StandardScaler
• Categorical encoding using OneHotEncoder
• Combined using ColumnTransformer

Model Used

• Random Forest Classifier

Final Performance

Metric	Score
Training Accuracy	93.83%
Testing Accuracy	93.05%
F1 Score	86.46%

Model tuned using:

• n_estimators
• max_depth

---

Tech Stack

• Python
• Flask
• Scikit-learn
• Pandas
• NumPy
• HTML
• CSS
• Render

---

Project Structure

Irrigation_Need/
│
├── app.py
├── requirements.txt
├── runtime.txt
├── best_model_pipeline.pkl
│
├── templates/
│   ├── home.html
│   └── index.html
│
├── Model_training.ipynb
├── EDA.ipynb
└── README.md

---

Installation

Clone repository:

git clone https://github.com/vedddev/Irrigation_Need.git
cd Irrigation_Need

Install dependencies:

pip install -r requirements.txt

Run locally:

python app.py

Open:

http://127.0.0.1:5000

---

Deployment

Deployed using Render.

Start Command:

gunicorn app:app

Python Version:

3.11.9

---

Future Improvements

• Real-time weather API integration
• IoT sensor integration
• Smart irrigation scheduling
• Crop recommendation support
• Farmer advisory chatbot

---

Author

Vedant Shelake

GitHub: https://github.com/vedddev

---

License

This project is for educational and research purposes.