Release Description for Richter's Predictor: Modeling Earthquake Damage
Version 1.0.0 - Initial Release
Release Date: 1/16/2024
I am excited to announce the first release of the Richter's Predictor: Modeling Earthquake Damage project. This release marks the beginning of our journey to provide an effective tool for predicting earthquake damage in Nepal, based on the data from the devastating 2015 earthquake.
Key Features:
- Data Analysis and Preprocessing: Implementation of comprehensive data analysis and preprocessing steps to understand and prepare the Nepal Earthquake dataset.
- Model Development: Initial development of predictive models using machine learning techniques to estimate earthquake damage levels.
- Jupyter Notebooks: Detailed Jupyter notebooks are provided to showcase the data analysis, preprocessing, and model-building steps.
What's Included:
- Jupyter notebooks with complete analysis and model development.
- XGBoost, LGBM, TensorFlow (TF) Neural Network, TensorFlow Decision Forest (TFDF), and more!
- Initial predictive models with baseline performance metrics.
- Documentation on data sources and initial findings.
Acknowledgments:
We want to thank the DrivenData community for providing the dataset and challenge that inspired this project.
Looking Ahead:
In upcoming releases, we aim to improve our models' accuracy, incorporate more sophisticated machine learning algorithms, and deploy the model as a web-based application for easier access and use.
For more information, queries, or contributions, please visit our Issues page.