This is my final Machine Learning and Data Mining exam's answer.

This dataset contains 30,000 datas on asteroid objects, while the original version (more than 100 MB in size) contains more than 800,000 asteroid object data. Most of the objects in the two datasets are near-Earth asteroids (NEO, Near Earth Object).

My work on this dataset is concerned with cleaning data and predicting the diameter of an asteroid.

If you have any suggestions that would improve my code, please feel free to suggest it as an issue. ") Thank you