Re-posted from the Azure blog channel
Whats new on arXiv
Robust Estimation of Data-Dependent Causal Effects based on Observing a Single Time-Series
Cosmos DB for Data Science
Cosmos DB is a snazzy new(ish) Microsoft Azure product. I was able to go to Microsoft Office in London for three days of training on the database service, which was really well structured and well run, with a lot of knowledgeable Microsoft bods around to pass on their considerable knowledge. This post will extract out some key features and benefits of the service, and then discuss how this fit’s into a data scientists role.
Distilled News
An End-to-End Guide to Understand the Math behind XGBoost
Multithreaded in the Wild
Hello Stitch Fix followers, check out where our fellow Stitch Fixers are speaking this month.
Get started with automated metadata extraction using the AWS Media Analysis Solution
You can easily get started extracting meaningful metadata from your media files by using the Media Analysis Solution on AWS.
How to scrape data from a website using Python
Introduction
Whats new on arXiv
Future-Prediction-Based Model for Neural Machine Translation
Whats new on arXiv
NNCubes: Learned Structures for Visual Data Exploration
Visualization in the 1980s, just before the rise of computers
Graham Douglas, a data journalist at The Economist, looks back on the days when getting data and visualizing it was tedious from start to finish: