DSN Pre-Bootcamp Hackathon: Expresso Churn Prediction Challenge by Data Science Nigeria
Helping Nigeria
Knowledge
Completed (over 5 years ago)
Classification
Prediction
671 joined
358 active
Start
Aug 08, 20
Close
Aug 22, 20
Reveal
Aug 22, 20
2nd Place Solution and also a score of 0.246643, which is slightly better than the current 1st place score.
Connect · 26 Aug 2020, 02:27 · 4
Congratulations to everyone who participated.
Here's the link to my solution
https://github.com/KolatimiDave/Expresso-Customer-Churn-Prediction I'm open to questions.
Star repo if you like my solution
Lets connect on linkedIn: www.linkedin.com/in/olukolatimidavid
Thanks for sharing your notebook @KolatimiDave. I've really learnt a lot from you and others who have been throwing their notebooks in. And also the blend at the end of the framework is really cool... I'll make some trials on my results as well and see where it pushes...
Thanks for the solution. Great work!
Quick question, the values you used in multiplying your blend, how did you obtain them.
I chose the weights of each model based on their individual score on the leaderboard
sorry, is it based on the private leaderboard or public leaderboard?