Lesson 10 of 10
What else could be tried
Next ideas, and what's not allowed.
Outputs will appear here after the full run. The code is final.
Left: what & whyRight: the code and its real output
10. What else could be tried
- Pseudo-labelling: add confident test predictions to training, so the model sees the new operators (hard to validate honestly).
- Seed averaging: train each model with 3 seeds and average (small, reliable gain; costs 3× time).
- Tuning: a small random search on depth / learning rate / regularisation for XGBoost and CatBoost.
- Facilities as text: TF-IDF + SVD over facility codes instead of one-hot columns.
- Not allowed / not used: external data (the original public listings), AutoML, deep-learning frameworks.