/competition-winners

The code for the prize winners in DrivenData competitions.

MIT LicenseMIT



Background

DrivenData runs machine learning competitions to help non-profits, NGOs, governments, and other social impact organizations use data science in service of humanity. Part of DrivenData's mission is to enable data scientists and non-profits to learn from the work that is done in these competitions. To this end, the code submitted by winners is released under an open source license for others to learn from, use, and adapt.

What's in this Repository

This repository contains code prodvided by the winners of competitions hosted on DrivenData. Feel free to browse to any particular competition's winning repository or just clone all of the winning code to learn from!

Winning Submissions

Competition
America's Next Top (Statistical) Model
Box-Plots for Education
Countable Care: Modeling Women's Health Care Decisions
From Fog Nets to Neural Nets
Keeping it Fresh: Predict Restaurant Inspections
Naive Bees Classifier
Senior Data Science: Safe Aging with SPHERE
Pri-matrix Factorization
Pover-T Tests: Predicting Poverty
Random Walk of the Penguins
N+1 Fish, N+2 Fish
Power Laws: Forecasting Energy Consumption
Power Laws: Anomaly Detection
Power Laws: Optimizing Demand-side Strategies
Power Laws: Cold Start Energy Forecasting
Sustainable Industry: Rinse Over Run
Open AI Caribbean Challenge: Mapping Disaster Risk from Aerial Imagery
Hakuna Ma-data: Identify Wildlife on the Serengeti with AI for Earth
Open Cities AI Challenge: Segmenting Buildings for Disaster Resilience
Clog Loss: Advance Alzheimer’s Research with Stall Catchers
TissueNet: Detect Lesions in Cervical Biopsies



Cloning this repository

Note that when any repo containing submodules is cloned, each submodule needs to be initialized else they remain empty.

To avoid explicitly initializing each submodule, use the --recursive flag when cloning:

git clone --recursive https://github.com/drivendata/competition-winners.git.

This will download the competition-winners repository, as well as the contents of every submodule in the winners repository.