RECENT TOPICS

Full Stack Python for AI: How open source Python enables each phase of the AI workflow By: Tripp Smith
Date: Oct. 12, 2017, 6 p.m.

Investments in AI are heating up, with the total market estimated as high as $126B by 2025. This talk will present case studies and code samples of how our clients are using Python today and how we expect this to evolve over the next few years as AI becomes increasingly ubiquitous. Python enables each phase of the AI pipeline: DevOps, Data Engineering, Model Development, Deep Learning, Cognitive User Interfaces, and Microservices. This talk will highlight how Python is a common glue across multiple disciplines that will allow cross functional teams work together to get real results from AI.

YouTube logo
Getting Off the Struggle Bus: Learning From Transit Data By: Spencer Chan
Date: Sept. 14, 2017, 6 p.m.

In an extended version of the lightning talk I gave for the spring ChiPy mentorship final presentations, I will go into more depth about how I collected and processed bus location data from the CTA's bus tracker API. I will also discuss interesting discoveries I made once I plotted the data, work I have done on the project since completing the mentorship (collecting data from 30 additional bus routes, converting visualizations from Bokeh/Python to D3.js, analyzing and visualizing bus bunching, etc), as well as future plans for the project.

YouTube logo
Anaconda: The Best of Everything in PyData By: Patrick Boland
Date: Sept. 14, 2017, 6 p.m.

A continued narrative of the tale of two snakes. In this talk, we will discuss some of the most impressive features of Anaconda, including built in binaries, command line interface, the history of the distribution, and why it is the right choice for just about every Python stack. This talk does not assume audience familiarity with the distribution. We will take advantage of the *better* batteries included nature of this distribution to step through beginner and intermediate concepts. I intend for the audience to feel comfortable and excited to give this a try on their own.

YouTube logo
Storm surge: hurricane flooding simulation using Python, Fortran, and GeoClaw By: Marc Kjerland
Date: Sept. 14, 2017, 6 p.m.

The 2017 hurricane season is proving to be one of the strongest in history, and predictive modeling plays an important role in evacuation and mitigation planning. Coastal communities in the path of hurricanes face several major hazards - strong winds, heavy rainfall, relentless waves, and storm surge. Storm surge is a type of transient sea level rise where water is forced towards the shore by winds, and the right conditions can produce very high levels - Hurricane Harvey raised Galveston Bay by upwards of ten feet, and in 2012 Hurricane Sandy produced 12-foot surge in Lower Manhattan. I'll discuss the current state of storm surge modeling with focus on an open-source package called GeoClaw, developed by academic researchers across the U.S. GeoClaw uses Python and Fortran to run a dynamic simulation of coastal flooding using storm and topography datasets, and thanks to some novel dimensionality reduction it can be run on a laptop.

YouTube logo
Fortune-Telling with Python: An Intro to Time Series Modeling By: Jonathan Balaban
Date: Aug. 10, 2017, 6 p.m.

Description: a pythonic tour of time series methodologies and packages, including ARIMA, seasonal models, and Markov approaches. Intermediate level with basic statistics and time data familiarity required. Bio: Jonathan Balaban is a senior data scientist, strategy consultant, and entrepreneur with ten years of private, public, and philanthropic experience. He currently teaches business professionals and leaders the art of impact-focused, practical data science at Metis.

YouTube logo
Selenium: Doing Magic with Websites Using Python By: Seth Weidman
Date: Aug. 10, 2017, 6 p.m.
Selenium is an excellent package that lets you dynamically interact with websites right from your Jupyter Notebook. At Metis, we teach Selenium early in our bootcamp. We have a great Selenium tutorial that involves lets you make a reservation on OpenTable using Python. In addition, several students have already completed excellent projects using Selenium. During this talk, Seth Weidman, a Senior Data Science Instructor at Metis, as well as the following two current Metis students [TO BE DETERMINED] will be presenting the projects that they just completed.
Tracking FCC bots with Python By: Chris Sinchok
Date: Aug. 10, 2017, 6 p.m.

I've been doing a bunch of analysis on the recent FCC public comments in Python (https://medium.com/@csinchok/an-analysis-of-the-anti-title-ii-bots-463f184829bc). Due to this work, I was quoted in Gizmodo, Ars Technica, and the BBC. I'd like to talk about how I approached this problem, how Python helped make sense of my findings, and what my conclusions are.

YouTube logo
A Gentle Introduction to Context Managers By: Aly Sivji
Date: June 8, 2017, 6 p.m.

We've all used context managers provided by the Python Standard Library to read from/write to a file. Have you ever wondered what was happening underneath the hood when you used a with statement? This talk will explore context managers, discuss various use cases, and show you how to implement a context manager to manage MongoDB connections.

YouTube logo
Python for Home-Ec By: Adam Forsyth
Date: June 8, 2017, 6 p.m.

Have you ever tried to make something with scrap wood, and wondered how to use it optimally? Do have a bunch of pickles and jams you made, and you want to eat them in an order that maximizes variety? These are real problems a co-worker of mine had, and we used Python to solve them. I'll show the data we started with, the solutions we came up with, and a bit of the computer science behind them. See some examples of how to think through problems and design your own algorithms to solve them.

YouTube logo
Python for mathematical visualization: a four-dimensional case study By: David Dumas
Date: May 11, 2017, 6 p.m.
This is a talk about creating pictures of a mathematical object---specifically, a 4-dimensional fractal "dust" that has been the subject of mathematical research in hyperbolic geometry since the 1980s. In the end this is accomplished using a little algebra, a little geometry, and a healthy dose of Python. That is, I will present a case study of using Python in several aspects of a mathematical visualization project, from the computation itself, to transforming and converting data, and finally for scripting the process of generating the images. Along the way I'll explain how Python's convenient idioms and containers (e.g. sets and set comprehensions) are a good fit for some of the algebraic and geometric questions that come up, how Scipy and Numpy enable fast numerical calculations, and how Python's strength as a language for scripting and automation allows easy orchestration of rendering of still images and frames of animations. The mathematical visualization project we describe is a collaboration with François Guéritaud (Université de Lille).