Appsilon is officially announcing two new R packages: shiny.fluent and shiny.react. shiny.fluent brings Microsoft’s Fluent UI to Shiny, and shiny.react integrates the React JavaScript library with Shiny.

In this article, we focus on the technical aspects of the machine learning solution we implemented for the xView2 competition. We used PyTorch to build our models for satellite image analysis and fast.ai to develop their critical parts.

We recently took part in Hakuna-ma Data, a competition organised by DrivenData in partnership with Microsoft’s AI for Earth, which asked participants to build an algorithm for wildlife detection that would generalise well across time and locations...

Building Deep Learning vision models satellite imagery to assist in identifying and labeling property damage from natural disasters.

The algorithms we build affect lives. We have an ethical obligation not to teach our machines to be prejudiced. Here are some lessons and examples that will help you achieve this in your machine learning model assembly and use.

We need to build a culture where AI is built with a focus on the entire outcome, not just metrics or profit.

AI will transform the insurance sector. Two techniques that will have the biggest impact are convolutional neural networks (CNNs) and recurrent neural networks (RNNs). The 3rd key to successful AI implementation: productionization

Our CTO used transfer learning to build an object detection model on a small dataset - the model can recognize certain animals in images.

Testing automation can help reduce the risk of errors, avoid overhead and increase the quality of your R project. Read to learn more about CircleCI.

What's the best way to achieve fast data lookups in R? In this post, we compare dplyr and data.table, and also explore what data.table indexes bring to the table.