Learn with Tech Resources

We announced the Radiant Earth Spot the Crop Data Challenge winners to predict crop types in Western Cape, South Africa using satellite image time-series. We sat down with Raphael Kiminya, the winner of the track that used Sentinel-2 multispectral data as input to his model, and MG Ferreira and Tien-Dung LE that used both Sentinel-2 and Sentinel-1 (radar) data as input. Read how they've approached the challenge and join us for a live webinar on Jan 19 at 8:30 AM EST. They will dive deeper into their winning solutions.

An accurate and timely diagnostic model for tropical cyclone intensity can save lives and property. Join this global competition to predict tropical cyclone wind speed in near-real-time for a chance to win up to $13,000 in cash + $18,000 in Azure credits.  This competition ended on Feb. 2, 2021

This training of trainers bootcamp will cover topics related to use of machine learning techniques on Earth observations. Participants are required to attend all training sessions and work on assignments during the bootcamp. There will be a total of 5 days of training sessions within the two week period. Each participant is then expected to train five other individuals within four months after the bootcamp and report that back to the organizing team. The follow up training is a requirement to receive your certificate of participation in the bootcamp.

Webinar | September 1 | 9 AM EDT

Speakers Joe Morrison, Azavea's GroundWork product specialist, and Hamed Alemohammad, Radiant Earth Foundation's chief data scientist, will talk about how to generate your training data using GroundWork and export and discover it in Radiant MLHub.

Radiant Earth Foundation hosted an international expert workshop to discuss how best to use machine learning techniques on NASA’s Earth Observation data and address environmental challenges. This workshop was sponsored by the NASA Earth Science Data Systems (ESDS). 

• Read the Workshop Report (PDF) to learn about the challenges, potential solutions, and best practices for using EO data in ML workflows identified by expert participants 
• Watch the recorded expert presentations

In this online meeting, the proposed FIVE best practice criteria that make ground reference data ready for Earth observation machine learning models are discussed. Co-hosted by Radiant Earth and CGIAR, the meeting took place on Tuesday, April 21 at 10:00 AM ET.  You can read Radiant Earth's short write up on the proposed criteria, or click below to watch the recording of the discussions.