Your Guide to Quality Data Science Products

Convolutional neural networks and deep learning don't have to be difficult subjects. Read this article if topics like input and output channels, kernels, strides, and receptive fields feel a bit fuzzy to you.

PP-YOLO is a machine learning object detection framework based on the YOLO algorithm. In this article, we’ll explain what PP-YOLO is, why it is an improvement over YOLOv4, and show you how to use PP-YOLO for object detection.

Recent research on weight poisoning attacks on pre-trained models in the area of NLP may transfer to the context of Computer Vision, where the usage of pre-trained weights is widespread.

Learn best practices for setting up a data science team and kicking off a data analytics project – remote or otherwise. I'll cover Scrum methodology, Asana, GitHub, Docker, renv, linter, and a variety of other tools and workflows.

xspliner is an R package that helps explain black box ML models. In this presentation, you will learn what PDP curves and GLMs are and how you can calculate them based on black box models. I'll then show you a specific use-case for xspliner.

In this presentation we present how to make use of the latest technical advancements in computer vision with an R setup even if they are originally implemented in Python.

Appsilon engineers Krystian Igras, Marcin Dubel, and Jędrzej Świeżewski, PhD will be giving virtual presentations on Friday, June 19th. Learn about xspliner, making production-ready R code, and using R for Machine Learning projects. 

Appsilon created a dashboard in collaboration the International Institute for Applied Systems Analysis to deliver information about COVID-19 to policymakers. We implemented the solution with Python and Svelte.

Introduction to object detection and image classification featuring the YOLO algorithm and its Darknet implementation

Our submission to the Pandemic Response Hackathon (CoronaRank) is inspired by Google’s PageRank and utilizes geolocation data in the Apache Parquet format from Veraset for effective exposure risk assessment using R and Markov Chain modeling.

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.

This is the first of two blog posts about our recent participation in the Pandemic Response Hackathon. Our project (CoronaRank) was one of only 5 projects out of 230 submissions chosen to present at the closing ceremony.