My solution for the 7th place / 245 in the Umoja Hack 2022 challenge
Umoja Hack Africa is a yearly hackathon organized by Zindi, a data science platform for African data scientist, the hackathon contain 3 challenges : Beginner, Intermediate & Andvanced, this is my solution for the Intermediate challenge which allowed me to avoid overfitting and rank 7th in the final leaderboard while I was 25th in the public leaderboard.
The solution is based on a slightly tuned LGBM regressor model with basic feature engineering, but the model was trained on log(1+y) instead of the target directly in order to reduce the optimisation space and give good results, this improved the score a lot.
Besides, I've set up a good cross validation strategy with 10 folds, the CV did follow the private Leaderboard, this is my best solution according to the local CV.
If you enjoyed this solution don't forget to star the repo :)