— Tools and Techniques —
Opendoor buys and sells thousands of homes each month and it's critical for them to be able to predict how long a house will take to sell. Here's how they use survival analysis to model their business.
AutoML-Zero aims to discover machine learning algorithms from scratch, starting from empty or random programs and using only basic math operations. This repo is a nice introduction to the project.
This tutorial walks through the end-to-end implementation of a custom built lead-scoring model using MLflow, AWS Sagemaker, and Booklet.ai.
If you think your R code runs slow, it might just be that you're doing it wrong. This post by Thomas Lin Pedersen explores the design of a vectorised function to help make your R code run faster.
— Data Viz —
This interactive article by Kevin Simler explores the effectiveness of social distancing when a population is confronted with an infectious disease. This is very well done and includes a variety of scenarios and considerations.
If you’re a data visualization designer or a data engineer and are interested in helping health organizations create clear communications of the COVID-19 pandemic, this initiative is worth checking out.