— Insight —
Recent advances in machine learning have resulted in AIs that can actually write for you. The technology isn't perfect but it's impressive. Just check out the predicted text that precedes each section in this long-read from The New Yorker. Where will predictive text take us n
There's a lot of talk about AI but for a lot of reasons, government and social sectors are being left behind. In this post, Jennifer Pahlka takes a look at why that is and how the deep the issues go.
— Tools and Techniques —
sotabench is a new service from Papers with Code that aims to benchmark every open source ML model. It integrates directly with GitHub and results are automatically compared to papers and other models.
This post in Adrian Colyer's, The Morning Paper, takes a look at lessons learned from 150 successful machine learning models at Booking.com.
According to this analysis by Horace He, TensorFlow is currently the platform of choice in industry but researchers are abandoning TensorFlow in droves. This article explores where they're all going and why. There's also a useful discussion on Hacker News that goes along with this article.
If you're interested in exploring TensorFlow 2.0 + Keras, this Colab notebook by Francois Chollet is a great place to start. This is a very well-organized crash course that starts with the basics and includes lots of examples along the way.
Upskill yourself, work on production-level machine learning projects with 1:1 mentorship from machine learning experts and personalized career coaching. Go through a proven approach that raises the average salary of our graduates by $27,000. Get a job or your money back.
— Resources —
Great resource that goes along with the STAT 545 course at the University of British Columbia. This is well organized and comprehensive.
— Data Viz —
In its premier issue, the Sports Data Viz Digest highlights a variety of noteworthy visualization projects across the world of sports. Includes linked references to interactives and related tutorials.
This is a beautiful tutorial by Christian Burkhart that shows how to create violin plots with ggplot2.
— Conferences & Events —
Scale by the Bay - Learn from top Data Engineers from Netflix, Spotify, Twitter, DataStax, Databricks how to build an end to end data pipeline and seamlessly integrate deep neural networks with traditional software development. November 13-15. Data Elixir readers save 15% with this code: DATAELIXIR15