— In the News —
Building a data science team? Jeremy Stanley and Daniel Tunkelang discuss why, when, where and how. Jeremy is VP Data Science at Instacart and Daniel Tunkelang is a highly regarded data science leader and advisor. Must Read.
As Silicon Valley fights for talent, universities struggle to hold on to their stars.
Riley Newman is Head of Data Science at Airbnb. In this podcast, Riley discusses why it’s crucial for startups to invest in data early, the difficulties of scale, and how data science can find the voice of the customer. There are lots of good insights here and the discussion is transcribed if you prefer to read.
— Sponsored Link —
Data ScienceTech Institute is the first applied school of Data Science and Big Data in Europe with campuses in Paris and Nice! We offer the most intensive programs available to take up the Big Data challenge!
All our courses are taught in English, programs are available on-campus or through our HD telepresence system. You can also choose to follow only one or more courses if you wish to enhance your knowledge in Data Science or Big Data.
— Tools and Techniques —
Awesome! Tinker with a real neural network in your browser.
It’s one thing to be interested in Machine Learning, it’s another thing altogether to actually start working in the field. This post will help you understand both the mindset and the specific skills you’ll need to start working as a Machine Learning engineer.
Nice exploration of MTA arrival times. Includes MTA API usage, analysis, data visualizations, and a GitHub repo of code.
missingno provides a set of easy-to-use utilities that give you a quick visual summary of the completeness (or lack thereof) of your dataset. It's built using matplotlib, so it's fast, and takes any pandas.DataFrame as input.
— Resources —
The Deep Learning textbook was recently completed and is available online for free. This is a great resource that's intended to help students and practitioners enter the field.
— Data Viz —
Data visualizations are everywhere these days and for good reason. Among other things, visualizations can help make complex data understandable. But as Jake Porway, executive director of DataKind, explains here, "Data visualization without rigorous analysis is at best just rhetoric and, at worst, incredibly harmful."
This project from the M.I.T. Media Lab bills itself as “the most comprehensive visualization of U.S. public data." There are some occasional gaps but overall, it does a great job of pulling together lots of data from disparate datasets. This project is open-source and includes an API.
Use R Markdown to publish a group of related data visualizations as a dashboard.