— Insight —
Palantir helps governments and law enforcement decipher vast amounts of data — to mysterious and, some say, dangerous ends. This is a great profile of the secretive tech giant and its surprisingly lofty aims.
Real-life learnings from five data team iterations: centralized, embedded, full-stack, pods and business domains.
— Tools and Techniques —
Feature stores are used to manage the datasets and pipelines that are needed to get machine learning applications to production. This post walks through the key components of a modern feature store and how they unlock a new kind of collaboration across data science teams.
Pretty cool! This tutorial shows how to create art posters using Open Street Maps data, your own Google Location History and ggplot2. This is a detailed walk-through that explains how to work with map data and shows lots of options along the way.
PandasGUI is a graphical user interface for analyzing Pandas dataframes. This is an early release but it looks promising and definitely worth playing with.
— Resources —
This is a great short-list of introductory machine learning courses that will help you get up to speed with fundamental concepts like clustering, regression, evaluation metrics, etc.
MIT's "Machine Learning in Healthcare," is now available on MIT's OpenCourseWare. The course introduces a variety of topics such as ML for risk stratification, disease progression modeling, precision medicine, diagnosis, subtype discovery, and improving clinical workflows.
— Data Viz —
This nine minute video is an awesome introduction to the Gartner Hype Cycle and explores novel ways of visualizing trends over the past 25 years. Check out the video and then play with the data yourself in this interactive visualization by Amelia Wattenberger >>