AI Datasets

If you are interested in developing hands-on experience with AI and machine learning, we encourage you to experiment with one of these benchmark environmental science datasets. If you want to add your own dataset to the list, please email David John Gagne.

A comprehensive list of machine learning datasets for weather and climate applications can be found at

Name Description Link
AMS Solar Energy Prediction Contest Predict total daily solar irradiance from GEFS and Oklahoma Mesonet Data Kaggle
How Much Did It Rain I Estimate rainfall probability distribution from Dual Pol. radar data. Kaggle
How Much Did It Rain II Estimate rainfall from Dual Pol. radar data. Kaggle
Understanding Clouds from Satellite Images Identify the cloud classification from satellite imagery Kaggle





AI Contest Summaries

2014 Solar Energy Contest

2008 Storm Classification Contest