The objective of this challenge is to build a machine learning model to predict a measure of wealth as measured in household surveys using readily available inputs., have a look on Zindi.
This Repository contains an Unofficial Winning Solution Code for this Challenge . Enjoy !
Solution Code Pipeline :
1. Data Processing :
1.1 Apply frequency Encoder to : 'Year','country','urban_or_rural' Columns .
1.2 Combine similar Features ( we sum them )
2. Modeling :
Using Cross validation strategy with K=10 , We trained an LGBM Model .
Look for the team name : AKCE
Rank : 1/141
| Name | Zindi ID | Github ID |
|---|---|---|
| Azer KSOURI | @ASSAZZIN | @ASSAZZIN-01 |
| Caleb Emelike | @CalebEmelike | @CalebEmelike |