PRACTICE Beginner Challenge
Can you predict a measure of wealth for different locations across Africa?
Prize
Knowledge
Time
Ended ~1 year ago
Participants
107 active · 553 enrolled
Helping
Africa
The right algorithm ?
Data · 13 Mar 2022, 17:43 · 5
which algorithm should i use ? i used RandomForestRegressor
honestly I'm not sure. but in the jupyter starter notebook they recommended using xgboost and neural networks saying it's better than randomforest in this task.
they also used catboostregressor at the end.
Try go with XGBoost, CatBoost and LGBM. They would give you the best results.
Just be sure to tune the hyperparameters properly.
Also, consider using the frequency encoder for the categorical variables.
I tried xgboost but it froze my computer for some reason.. both on jupyter and on pycharm :(
That's weird.
Try put the hyperparameter
n_jobs = 1
In the definition of xgboost
lgbm, catboost , xgboost ,gradient boosting regressor all did well . i'm yet to properly tune the hyperparameters so you can start with these