QYZ: LaTeX, R and Redis for Beautiful Analytics

Average rating: ***..
(3.18, 11 ratings)

As data science becomes more integrated with the day to day operations of business we see a bifurcation in the meaning of ‘analytics.’ To those of us doing data science we think of writing code, creating hundreds of figures and tables and looking through them quickly. To our friends in business development and senior management they think of reading comprehensive and well formatted reports.

We found that often ‘analysis work’ was requested on a very short timeline and was intended to be shared with many people. This creates problems, since doing any sort of science quickly is difficult. To make matters worse traditional research tools do not produce the most attractive graphs, charts or typesetting. This caused our analysis team to spend many late nights performing ad hoc analysis while simultaneously trying to get cogent reports out the door.

To help alleviate this problem we have designed a system that makes the transition from R-code-snippet to polished and typeset report a breeze. We will discuss choosing tools and implementing an architecture built around them.

We will talk about integrating PostgreSQL, LaTeX, R and Redis to create a powerful and flexible stack that is a great alternative to the commercial analytics and reporting platforms.

Photo of Noah Pepper

Noah Pepper

Lucky Sort

CEO of Lucky Sort, a Portland based startup building an intuitive and powerful platform for realtime text analysis using visual analytics and machine learning.

Homer Strong

Lucky Sort

Homer Strong is a data hacker in Portland. His background is in statistics, but he knows how to wrangle a unix terminal too.