🌾 Crop Yield Prediction

📌 Project Description

This project predicts crop yield (in tonnes/hectare) based on historical data including weather (rainfall, temperature), soil properties (pH, nitrogen, etc.), and crop type. It uses machine learning models and includes a web-based frontend for user interaction and data visualization.

---

🎯 Objectives

• Predict yield based on soil and weather parameters.
• Enable user-friendly input via frontend.
• Display predictions and data insights using visualizations.
• Aid farmers and agriculture planners in better decision-making.

---

🧰 Tech Stack

Area	Technology
Programming	Python 3
Machine Learning	scikit-learn, pandas, NumPy
Visualization	matplotlib, seaborn, Plotly
Backend	Flask
Frontend	HTML, CSS, JavaScript (Bootstrap)
Model	Linear Regression / Random Forest
Deployment	Localhost / Flask server

---

📂 Dataset Details

1. crop_data.csv
   • Columns:
     • Crop
     • Rainfall (mm)
     • Temperature (°C)
     • pH
     • Nitrogen (N)
     • Phosphorus (P)
     • Potassium (K)
     • Yield (tonnes/hectare)

📌 Note: Dataset can be collected from Kaggle, ICAR (India), or FAO sources.

---

🚀 Features

• ✅ Crop yield prediction using trained ML model
• ✅ Frontend form for user input
• ✅ Visualization of dataset and predictions
• ✅ Downloadable prediction report (optional)
• ✅ Responsive UI using Bootstrap

---

🖼️ Sample Visualizations

• Correlation Heatmap
• Crop-wise Yield Bar Plot
• Scatter Plot: Rainfall vs Yield

---

🧪 How to Run the Project

1. Clone the Repository

git clone https://github.com/yourusername/crop-yield-predictor.git
cd crop-yield-predictor

2. Install Dependencies

pip install -r requirements.txt

3. Run Flask App

python app.py

4. Open in Browser

Visit: http://127.0.0.1:5000

---

🗂️ Project Structure

├── static/
│   └── style.css
├── templates/
│   ├── index.html
│   ├── result.html
│   └── visualize.html
├── crop_data.csv
├── app.py
├── model.pkl
├── requirements.txt
└── README.md

---

🧠 Machine Learning Model

• Model Used: Random Forest Regressor (can switch to Linear Regression)
• Accuracy: ~90% (varies with dataset)
• Target Variable: Yield (tonnes/hectare)

---

📊 Frontend Preview

• Home Page: Enter soil, weather, crop details
• Prediction Page: Displays predicted yield
• Visualization Page: Shows charts (bar, heatmap, scatter)

---

🛠️ Sample Input (From UI)

Crop: Rice
Rainfall: 120 mm
Temperature: 28°C
pH: 6.5
N: 80
P: 45
K: 40

✅ Sample Output:

Predicted Yield: 3.8 tonnes/hectare

---

📈 Sample Charts

1. Bar chart of yield by crop
2. Heatmap showing correlation of features
3. Scatter plot for rainfall vs yield

---

👤 Author

• Name: Charan Basava
• GitHub: CharanBasava

---

📄 License

This project is open-source and available under the MIT License.