Stock Price & Churn Prediction Application

(Click on the Image to be redirected to the video)

This repository contains two main components:

1. A Stock Price visualization app using Streamlit
2. A Customer Churn Prediction model using Flask

Project Structure

project/

├── churn_predict/

│ ├── datasets/

│ │ └── churn_data.csv

│ ├── model/

│ │ └── model_C=1.0.bin

│ ├── deploy_churn.ipynb

│ ├── myapp.py

│ ├── Dockerfile

│ └── requirements.txt

└── README.md

Stock Price Visualization App

Features

• Displays Google (GOOGL) stock closing prices
• Shows trading volume visualization
• Interactive date range selection
• Built with Streamlit for easy web interface

Requirements

python streamlit==1.31.0 yfinance==0.2.36 pandas==2.1.0

Running the Stock Price App

streamlit run myapp.py

Churn Prediction Model

Features

• Predicts customer churn probability
• Uses Logistic Regression model
• ROC AUC score of ~0.858 on test set
• REST API endpoint for predictions

Model Details

• Training data includes customer demographics and service usage
• Features include tenure, monthly charges, total charges, and various categorical variables
• Model trained using scikit-learn's LogisticRegression with 5-fold cross-validation

Requirements

python flask==3.0.0 pandas==2.1.0 scikit-learn==1.3.0 gunicorn==21.2.0 numpy==1.24.0

Running Locally

1. Install dependencies pip install -r requirements.txt
2. Run Flask application python predict.py
3. Making Predictions Test the API using curl: bash

curl -X POST -H "Content-Type: application/json" -d '{

"gender": "female",

"seniorcitizen": 0,

"partner": "yes",

"dependents": "no",

"phoneservice": "no",

"multiplelines": "no_phone_service",

"internetservice": "dsl",

"onlinesecurity": "no",

"onlinebackup": "yes",

"deviceprotection": "no",

"techsupport": "no",

"streamingtv": "no",

"streamingmovies": "no",

"contract": "month-to-month",

"paperlessbilling": "yes",

"paymentmethod": "electronic_check",

"tenure": 1,

"monthlycharges": 29.85,

"totalcharges": 29.85

}' http://localhost:9696/predict

5. Running the Streamlit App streamlit run predict_copy.py

Development

The project uses:

• Python 3.8+
• Jupyter Notebook for model development
• Flask for API development
• Docker for containerization
• Streamlit for stock visualization

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/AmazingFeature)
3. Commit your changes (git commit -m 'Add some AmazingFeature')
4. Push to the branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Data source for stock prices: Yahoo Finance
• Churn prediction dataset: Churn_data.csv

Contact:

Zoiba Zia