Chicago Python Data Special Interest Group (SIG) and Jellyvision are teaming up to present Lightning Talks! This month's meeting is graciously hosted and sponsored by Jellyvision!
5:30 - Doors open
6:00 - Talks Start
8:30 - See you all next time!
Analysis of Divvy Rides using PageRank
by Paul Russum
This project is performing analysis on the Chicago bike sharing company Divvy and uses the PageRank graph algorithm to determine which stations rank highest under multiple scenarios, such as workday commutes and Chicago events. The ride information lends itself nicely to a graph data model; each station acts as a node, and each ride is an edge. The data is available directly from Divvy and contains from_station, to_station, timestamps and rider demographics. The analysis is over 3 million individual rides occurring from April to December 2018.
how to beat a computer at math
by James rodovich
Computers are supposed to be good at math, but sometimes they're not. In this talk I'll help the audience understand numerical stability and how they can use it to outsmart a calculator or spreadsheet, which could be useful in the event of a robot apocalypse, or just to brag.
Tips for Career Transition
by Jie Liang
Career transition tips from my journey back to the workforce as a data scientist: passion found through self discovery and market research and job landed by iterative data driven decisions.
people shape software: what I learned comparing Python and R APIs
by James Lamb
I've spent the last year working on major improvements to a large IIoT platform and on a variety of open source projects. Through these activities, I've encountered this problem many times: "we have/need Python and R packages that do the same thing".
The question: "how could two communities of developers given the same functional requirements arrive at such different APIs?". In this talk, I'll explain some social forces that shape software, and argue that you can use CI to control their impact and make code better!
Putting Multi-arm Bandits in Context
by MJ Berends
Makefiles for Data Projects
by Ray Buhr
What is `make` and why should I use it? What are some examples of how `make` can help improve my workflow on data projects?
by Tanya Schlusser
Dan Chak's 'Enterprise Rails' has a ton of insight that's applicable to all web applications, not just Rails. His 4th through 8th chapters are relevant for the Data SIG: how should we (re)structure a website's database to enable analytics, moving it from "startup grade" to "enterprise grade"? This talk attempts to distill the most important points from those chapters into 10 minutes.
Productionizing ML Models on a Budget
by Katie Simmons
Katie will provide a brief case study from her time as a data engineer at a small startup helping data scientists to get trained models out of Jupyter notebooks and into production using flask and cookiecutter.
Optimizing Lineups For Fantasy Hockey
by Anthony Shook
In an attempt to gamble my way to a life of luxury and leisure, I built a tool to calculate the optimal lineups for DraftKings Daily Fantasy Hockey slates. Using Python and Julia, I scraped websites, cleaned and shaped data, and ran some greedy integer math, all in search of fame and glory. Or, at least a little pocket money.