Asteroid impact-risk analysis¶

This project focuses on the analysis of asteroid impact risk using machine learning techniques.

The project includes:

Dataset downloading and preparation.
Data cleaning and transformation.
Feature creation.
Exploratory data analysis.
Neural network training.
Model evaluation.
Impact probability predictions.
Visualization and reporting.

Project workflow¶

The complete project follows this workflow:

Dataset from Hugging Face
           |
           v
Dataset downloading and cleaning
           |
           v
Feature creation
           |
           v
Neural network training
           |
           v
Model evaluation
           |
           v
Impact probability predictions

Project outputs¶

The trained models are stored in models/.

The metrics, training history, and visualizations are stored in reports/.

To install and run the project, see Installation.