Tabular Kaggle Competitions are a monthly playground competitions that are a great place for begginer ML engineers to practice and improve their skills on small datasets.
The template for the code base is based on AAMLP book by Abhishek Thakur.
For each model, I have listed best pre-processing based on public leaderboard score.
| Model | Pre-processing | Public leaderboard |
|---|---|---|
| Decision Tree | ||
| SVM | ||
| Random forest | None | 0.96787 |
| XGBoost | None | 0.86727 |
| MLP | ||
| ResNet | ||
| Transformer |
TODO: Feature engineering ideas explored
What worked and what didn't and why?
What insights and learnings did you learn from this project?