LightGBM Predicting Water Pump Functionality

Water Pump functionality prediction. Based on the DrivenData challenge Pump it Up: Data Mining the Water Table

See the exploration in Streamlit Cloud 🎈

For better performance, grab the code from github repo

Project Walkthrough

Using data from Taarifa and the Tanzanian Ministry of Water, can we predict which pumps are functional, which need some repairs, and which don’t work at all?

Predict one of these three classes based on a number of variables about what kind of pump is operating, when it was installed, and how it is managed.

A smart understanding of which waterpoints will fail can improve maintenance operations and ensure that clean, potable water is available to communities across Tanzania.

Data Sources

  • Training and test data from drivendata, along with submission format

driven data downloads

Feature Exploration and Engineering

Much of this was guided by the DrivenData Competition forum, specifically this user’s EDA + Catboost example (I haven’t tried out all of his data processing steps… yet)


This package was created with Cookiecutter and the gerardrbentley/cookiecutter-streamlit project template.

DrivenData Platform

@misc{, doi = {10.48550/ARXIV.1606.07781},

url = {},

author = {Bull, Peter and Slavitt, Isaac and Lipstein, Greg},

keywords = {Human-Computer Interaction (cs.HC), Computers and Society (cs.CY), Social and Information Networks (cs.SI), Machine Learning (stat.ML), FOS: Computer and information sciences, FOS: Computer and information sciences},

title = {Harnessing the Power of the Crowd to Increase Capacity for Data Science in the Social Sector},

publisher = {arXiv},

year = {2016},

copyright = {Creative Commons Attribution 4.0 International} }