AutoInland Vehicle Insurance Claim Challenge
$1,000 USD
Can you predict if a client will submit a vehicle insurance claim in the next 3 months?
835 data scientists enrolled, 373 on the leaderboard
InsuranceFinancial ServicesPredictionStructured
26 March—27 June
Ends in 2 months
LightAutoML baseline -> 0.3252
published 1 May 2021, 19:23

Hello colleagues,

I have created the baseline for this competition based on our LightAutoML with the score 0.3252.

https://github.com/alexmryzhkov/Zindi_lightautoml_baseline

Feel free to use and make it better 😎

P.S. Special thanks goes to @kingabzpro who use LightAutoML on his own before this baseline - that's really cool to hear..

Alex

Nice baseline and FE. Model OOF scores ~ 0.5, test scores ~0.4, and only scores 0.33 on public. Why this happens? Does it due to best threshold shifts between trainset and testset? My model scores 0.42 on CV but only scores 0.29 on public, it is a bit strange