— In the News —
California recently passed a consumer privacy law that could force significant changes on companies that work with personal data. Technically, the law only applies to California residents but on a practical level, the new law will likely have broad impacts across the Internet. This article in the Harvard Business Review highlights what to expect.
— Profiles —
Cross-culture living + music + youth + TensorFlow + whale conservation. The contrasts here are amazing.
— Sponsored Link —
On July 24-25 Metis will host its 2nd annual Demystifying Data Science event - a free, online conference - featuring 28 interactive data science talks from industry-leading speakers. Keynotes include Lillian Pierson, CEO of Data-Mania LLC and author of Data Science for Dummies, and Beth Comstock, author and former Vice Chair of GE. Register now!
— Tools and Techniques —
Machine learning talent is reportedly in short supply but what is it that these experts really do? Answering that will help guide which skills to teach, which tools to build, and which processes to automate. This post by Rachel Thomas offers a practical perspective of the field, including an overview of common organizational issues that hinder machine learning efforts.
Seedbank is a collection of interactive machine learning examples that you can run, edit, extend, and grow into your own projects and ideas. Everything runs in the form of Colab notebooks, which means you can work with Seedbank entirely in a browser. The examples cover a broad range of topics from data analysis problems to art projects.
How random is random? Well, it depends. This guide explores a variety of ways to generate random data and discusses how each approach compares in terms of security, versatility, and speed. Includes suggestions for specific use cases and a summary of worthwhile resources.
Reading papers is a good way to keep up with fast-moving fields like deep learning but it can be time-consuming and take a lot of work. Here's how to make the best use of your reading time.
DJ Patil, Hilary Mason and Mike Loukides have started a series of articles on data ethics. In this second part, they explore the value of oaths and offer a more practical and effective alternative.
— Data Viz —
plotly.py 3.0.0 is a major release with lots of new features, including first class Jupyter widgets, fast typed arrays, deep figure validation, offline static image export, typeahead, and huge performance improvements. This is an awesome library to have in your visualization toolbox.
Here's a nice collection of things not to do in your data visualizations. Each caveat includes examples, a description of the issues, and linked references.
— Conferences —
Looking for a conference to attend, sponsor or submit a talk to? This latest list from Tryolabs includes 28 machine learning conferences around the world, including cities throughout the U.S, London, Toronto and Belgium.