Software Engineering
<- Return to all blogs
-
Using Noisy Labels to Train Deep Learning Models on Satellite Imagery
How do noisy labels affect the accuracy of a deep learning model? We added different amounts of noise to the SpaceNet Vegas buildings dataset and trained some models to find out.
-
Using Noisy Labels to Train Deep Learning Models on Satellite Imagery
How do noisy labels affect the accuracy of a deep learning model? We added different amounts of noise to the SpaceNet Vegas buildings dataset and trained some models to find out.
-
An Introduction to Machine Learning Accuracy Metrics
How accurate of a machine learning model can you make? It depends on how we decide to define “accurate”.
-
E84 @ 2019 ESIP Summer Meeting
We’re passionate about Earth Science Data and helping to solve the challenges in that field. At this year’s ESIP Summer Meeting, we have five E84 team members speaking and moderating across six different sessions, with an additional four of us out there to chat and support the event. Here’s where you can find us if…
-
Cloud-native Apps Force Developers to Put Security First
Data stored in the cloud is an attractive target for hackers. High profile breaches like the recent exposure of images of license plates at a U.S. port of entry have become commonplace. These events justifiably alarm consumers. Every incident cements distrust toward the tech industry, causing technology users to demand better security from software creators.…
-
Introducing WMS and WCS Support for GeoTrellis and Raster Foundry
Raster Foundry and GeoTrellis now support WMS and WCS standards, enabling streamlined workflows that utilize multiple data sources.
-
MUB Monitor: Helping Water Utilities Track and Respond to Upstream Threats
In partnership with Downstream Strategies, we created MUB Monitor – a comprehensive GIS-based tool that helps water professionals track spills, assess watershed threats, and make informed decisions.
-
E84 โค๏ธ Gatsby & JAMstack โ Introducing Gatsby [Spark]
The JAMstack generally has become ubiquitous as a solid and reliable architecture for building applications for the web. As much as we didn’t know it, the JAMstack has been part of E84’s blood before the term was coined, as we worked through application challenges focusing on serving static applications and websites using AWS S3 and…
-
Lessons in Functional API Development from Haskell’s Servant and Http4s
As a 10% time project, we replicated a small piece of the Raster Foundry backend code in Haskell, using servant for the REST interface and postgresql-simple for database interaction.
-
Write Merge Requests Like You're Posting to Instagram
Merge requests (or pull requests) are a huge part of a teamโs development process. Itโs the main gatekeeper preventing developers from throwing whatever they want into the default branch. It’s also a reference to the history and understanding of the changes someone is making to an application, giving code reviewers and testers more confidence in…
-
Using Cloud-Optimized GeoTIFFs (COGs) for More Efficient Web App Architecture
Cloud-Optimized GeoTIFFs (COGs) are geoTIFFs hosted on a cloud or file server, and are optimized for remote reads. They proved useful in a recent project.
