Overview
Welcome to the Stroke Prediction Project! This project focuses on predicting strokes efficiently using Polars DataFrames, avoiding the conventional use of Pandas for streamlined data processing and analysis.
Project Highlights
Efficient Data Handling: Utilizing the speed and scalability of Polars DataFrames for efficient data processing.
Advanced Feature Engineering: Exploring and engineering relevant features from health datasets to enhance predictive power.
Model Development: Leveraging machine learning algorithms for stroke prediction, potentially Random Forests or Histogram Based Gradient Boosting.
Threshold Optimization: Investigating and optimizing prediction thresholds to balance precision and recall.
Explore the code in the directory and adapt it to your specific use case.
Contribute Found a bug or have an improvement? Feel free to contribute by opening an issue or creating a pull request!
Project Status This project is actively under development, and contributions are welcome. Check the Issues tab for planned enhancements and known issues.
Acknowledgments Special thanks to the Polars community for their powerful data manipulation library.
Author
Dayana Vincent
Masters in Data Science
Feel free to reach out for questions or collaboration opportunities!