Public vs Private leaderboard

Winning a Kaggle Competition in Python

Yauhen Babakhin

Kaggle Grandmaster

Competition metric

Evaluation metric	Type of problem
Area Under the ROC (AUC)	Classification
F1 Score (F1)	Classification
Mean Log Loss (LogLoss)	Classification
Mean Absolute Error (MAE)	Regression
Mean Squared Error (MSE)	Regression
Mean Average Precision at K (MAPK, MAP@K)	Ranking

test split into public and private parts

# Write a submission file to the disk
submission[['id', 'target']].to_csv('submission_1.csv', index=False)

Submission	Public LB MSE	Private LB MSE
submission_1.csv	2.895	?

overfitting example

overfitting example

overfitting example with Public and Private leaderboards

small shake-up example

large shake-up example

Winning a Kaggle Competition in Python