Data Science Things Roundup #5

Time again for the 5th edition of the data science things roundup, named suspiciously similarly to the much more established Data Science Roundup by RJ Metrics (but we won’t worry about that this week).  In previous weeks we’ve seen some pretty cool ML and Data Science libraries, mostly in python, this week we branch out a little bit in more engineering-level projects (databases and deployment).

Space Invaders with Deep-Q Learning

By far the most compelling applications of machine learning for the general public is reinforcement learning.  Models that are playing games like go, Atari or whatever else seem to just be interesting to people.  I can’t say I disagree, and the backing algorithm is in many cases Deep Q Learning.  In this post, the concept of Q learning is introduced and some sample code shows it learning to play space invaders, super fun. Check it out here.

Lessons Learned from a Year of ElasticSearch in Production

At Predikto we are heavy daily users of ElasticSearch.  We’ve had it in production with huge datasets and complex query loads for a couple of years now, but there are always still little things to tweak to get out that last little bit of performance.  This is a great blog post with some of the common lessons learned. Check it out here.

Better Package Management [Python]

Package management and deployment of python projects can be a pain.  Doubly so for c-dependency heavy numeric/ML libraries.  In this post, Vincent Driessen introduces his best-practices for package management in python and supporting open source tools: pip-tools and pip-compile.  I’ve used both of these extensively now and can say that they are as advertised, and worth looking into. Check it out here.

The post Data Science Things Roundup #5 appeared first on Will’s Noise.