This repository hosts a web application developed with Gradio that predicts customer churn for a telecommunications company. By inputting customer information like tenure, monthly charges, and total charges, the app provides predictions about the likelihood of a customer churning.
-
Clone the Repository:
git clone https://github.com/your-username/churn-prediction-gradio.git -
Install Dependencies:
pip install gradio numpy scikit-learn -
Run the Gradio Interface:
gradio src/app.py -
Access the App:
Open a web browser and navigate to the URL displayed in the terminal (usually http://127.0.0.1:7860). The Churn Prediction Web App interface will be visible.
-
Interact with the App:
Adjust the sliders for tenure, monthly charges, and total charges to observe real-time churn predictions.
The predictive model utilized in this app is a fine-tuned Logistic Regression model. It assesses whether a customer is likely to churn, considering input characteristics like tenure, monthly charges, and total charges. The model was trained using a processed dataset with the Scikit-learn library.
This project serves as a demonstration of deploying a machine learning model via Gradio. It can be used as a blueprint for developing similar predictive web applications.
- Your Name
- GitHub: https://github.com/rasmodev
- LinkedIn: https://www.linkedin.com/in/rasmo-/
This project operates under the MIT License. To understand the terms and conditions, refer to the LICENSE file. Make sure to replace "your-username" with your actual GitHub username and provide accurate URLs for your GitHub profile and LinkedIn profile. Additionally, include the app_screenshot.png file (a screenshot of your app) in the repository, and if needed, upload the app.py file containing the Gradio code, along with any pertinent model weights or preprocessing steps.