PyData Carolinas 2016
2016-09-14..2016-09-16
Durham, United States
| title | author | length |
|---|---|---|
| The modern research skill set: Using Vagrant, Ansible, and Python to support rese | Bret Davidson | 38m |
NCSU Libraries builds reproducible classroom environments using Vagrant + Ansible + Python and stores the provisioning as text on GitHub so installs are… | ||
| Scikit build: A build | Jean Christophe Fillion Robin, Matt McCormick, Max Smolens, Omar Padron | 37m |
scikit-build claims to be glue between setuptools and CMake that actually finds dependencies, supports extra compilers and cross compilation, and generates… | ||
| Let it shine telling your data story | Sarah Bird | 39m |
Shows concrete practices not platitudes: use real data because dummy data often breaks designs with overplotting or no trend and forces last minute rewrites… | ||
| Making Sense Out of Flight Test Data with Python | Luke Starnes | 42m |
Aggregates 20 GB per sorty into a messy iRIG Chapter 10 container, decodes hundreds of messages with a machine‑readable XML “decoder ring,” and still wrestles… | ||
| Dynamics in Graph Analysis: Adding Time as a Structure for Visual and Statisti | Benjamin Bengfort | 38m |
Claims that embedding time as a structural element improves graph analytics and visualization, implemented with NetworkX and Graph Tool and tooling called… | ||
| Improving delivery of safe water to African communities with | Ginny Ghezzo, Vladimir Stemkovski | 39m |
Data analysis claims falling water delivery is driven not by lowered per‑user consumption but by losing users: 181 issued cards drop to 77 inactive and 45… | ||
| Open Data, Networks and the Law | Iain Carmichael, Michael Kim | 40m |
Court Listener supplies about 3 million opinions and ~30 million citations; a SCOTUS subnetwork of 30,000 SCOTUS cases and 250,000 citations drives the… | ||
| A pythonista’s pipeline for large scale geospatial analytics | Alice Broadhead | 28m |
An automated pipeline maps billions of device locations daily to millions of POIs using HDFS storage, geospatial joins and a Pythonic stack. It favors… | ||
| Integrating Scala Java with your Python code | Marius van Niekerk | 33m |
Claims an easy interop technique and a new library spylon that lets Python talk to the JVM without full rewrites, using py4j and standard JVM tooling… | ||
| Testing is Fun in Python! | Andrew Knight | 41m |
Claims that testing in Python is practical and enjoyable and surveys five frameworks: unittest, doctest, nose, pytest, and Avocado. Shows concrete examples… | ||
| Introduction to Zeppelin Notebooks and PySpark 2 0 | Kevin Prybol | 20m |
Zeppelin is shown running with an integrated PySpark 2.0 interpreter and a reproducible repo that builds in ~2 hours via Vagrant/VirtualBox. Live demo passes… | ||
| Sharing Your Side Projects Online and Making Your Github the Best Résumé It Ca | Timothy Hopper | 39m |
Practical checklist for turning side projects into readable, runnable repos: make a clear README-driven development front page, add a license, declare… | ||
| More Then Words : Business Applications of Recurrent Networks | Michael Alcorn | 27m |
Red Hat uses recurrent neural networks to predict churn and prioritize escalations, training on 18 months of behavior to forecast 6 months ahead and… | ||
| Scalable Patient Records De duplication using machine learning | Jaafar Ben Abdallah | 38m |
An incremental, updatable model sits on a Python ML de‑duplication library and uses learned field weights, active learning and predicate blocking to avoid… | ||
| Reach More People: SMS Data Collection with RapidPro | Erin Mullaney, Rebecca Muraya | 34m |
Claim: SMS reaches broad audiences — more than 90% of the world's population over 6 will have a cell phone by 2020 and 350 billion text messages are sent every… | ||
| Transforming Data to Unlock Its Latent Value | Tony Ojeda | 38m |
A repeatable EDA framework claims to turn messy tables into actionable views by a prep phase (identify entity hierarchy, create derived categories, bin… | ||
| Snakes on a Cloud: "Data science, meet DevOps" | Kyle Snavely | 36m |
Engineers use Forecast a Python analysis engine built with Pandas and NumPy to pull data from Splunk and Graphite (about 1 TB of logs and 12 GB of metrics per… | ||
| Keynote: Data, Decision Making, and Being Human | Carol Willing | 36m |
Data-driven claims and concrete artifacts, not platitudes: Jupyter notebooks are shown as the tool that helped Filipino health teams coordinate tuberculosis… | ||
| ITK in Biomedical Research and Commercial Applications | Matt McCormik | 01h 26m |
PyData Carolinas 2016
The Insight Segmentation and Registration Toolkit (www.itk.org) has become a standard in academia and industry for scientific… | ||
| Interactive Data Visualization Tools in R | Ashton Drew | 01h 21m |
Overview of concrete results not promises: uses R Shiny plus leaflet, plotly, DT to turn the Quakes dataset (≈1,000 events since 1964) into a working app with… | ||
| Data Science Driven Business Insights with Python and PySp | Chunhui Higgins, Bo Zhang, Chul Sung | 02h 00m |
Python on Bluemix plus PySpark and DashDB run a production pipeline that cleans web and offline logs, trains over 80 model variations, and deploys scoring and… | ||
| Introduction to Julia for Pythonistas | John Pearson | 01h 57m |
Claims Julia solves the two‑language problem by giving near‑C speed with high‑level syntax, demonstrates REPL workflows, arrays, broadcasting, and that slicing… | ||
| Introduction to Pandas | Daniel Chen | 01h 29m |
PyData Carolinas 2016
This tutorial aims to expose future data analysis with the Python Pandas library. The tutorial is meant for absolute beginners to Pandas… | ||
| Connected: A Social Network Analysis Tutorial with NetworkX | Rob Chew, Peter Baumgartner | 01h 32m |
Hands-on tutorial that codes with NetworkX in Jupyter notebooks on a real dataset: romantic ties among Grey's Anatomy characters. It builds graphs from edge… | ||
| Turning Jupyter Notebooks into Data Applications | Peter Parente, Gino Bustelo, Justin Tyberg | 01h 29m |
Shows a concrete workflow: use Jupyter dashboard plus declarative widgets to turn a notebook into a standalone web app that loads Durham health‑inspection… | ||
| Getting Started With Bokeh | Sarah Bird | 01h 46m |
PyData Carolinas 2016
Bokeh is a python interactive visualization library that uses web browsers for its presentation. Bokeh supports a wide variety of… | ||
| Deep Language Modeling for Question Answering using Keras | Ben Bolte | 01h 51m |
Demonstrates how to modify Keras to reproduce the published QA models (Feng et. al. 2015 and Tan et. 2016), showing exact layer choices, loss and optimizer… | ||
| Scalable Data Science with Spark and R | Zeydy Ortiz, Rob Montalvo | 01h 18m |
Demonstrates using SparkR DataFrame API on Apache Spark 2.0 inside a DataBricks hosted cluster: create and persist Spark data frames from R, explain lazy… | ||
| Datascience on the web | Francois Dion, Don Jennings | 01h 19m |
Demo shows how a Jupyter workflow becomes a Flask web app: start with hello.py and inherit.py, serve on port 5000, render Jinja2 templates and handle GET/POST… | ||
| Keynote: How to Engage with the South Big Data Innovation Hub | Lea Shanley | 22m |
Describes concrete programs, funding and software: four regional hubs funded after the 2012 White House initiative with NSF seed grants (about $6M in 2015 and… | ||
| Keynote: Data Science Community and You | Anthony Scopatz | 48m |
Presents concrete metrics and policy tests: diversity maps to Rényi entropy and diversity order q, gamma = alpha × beta, and choosing q changes whether… | ||
| Keynote: The NumFOCUS Ecosystem | Leah Silen | 11m |
NumFOCUS is a 501(c)3 that fiscally sponsors open‑source projects (IPython first in 2014), handling accounts, invoicing, grants and hackathon logistics; it… | ||
| Principles of Reporting Systems: It's About Time we Talked About Bitemporality | James Powell | 39m |
Argues that bitemporal/full temporality makes reports reproducible and provable: capture both natural time and observation time so reruns explain value… | ||
| Building self evolving video game AI in Python | Marshall Wang | 39m |
Builds an imitation AI that uses two hours of gameplay and a hierarchical random forest to predict human actions on the Pygame title LaserCat (handles… | ||
| Matplotlib 2 0 or "One does not simply change all the defaults" | Thomas Caswell | 35m |
Matplotlib 2.0 actually changes defaults and documents what changed: a new Vega 10 color cycle with C0–C9 color notation, new perceptually linear colormaps… | ||
| How to get public data from public servants | Hope E Paasch | 41m |
Explains how to get specific datasets: write a dated FOIA/FOIAA letter, ask for data dictionary and table layout, and request delivery as delimited files (FTP… | ||
| JupyterHub: A "things explainer overview" | Carol Willing | 19m |
JupyterHub spawns and manages one Jupyter notebook server per user, with a Hub that handles authentication, a proxy that routes traffic, and pluggable spawners… | ||
| Dynamic Object Gaze Tracking with OpenCV | Shariq Iqbal, Jeff MacInnes | 33m |
Pipeline uses OpenCV SIFT+FLANN+RANSAC to find the static artwork inside each eye‑camera frame, computes a homography and applies perspective transform so gaze… | ||
| Scraping Lyrics with Celery | Rebecca Conley, Mark Lavin | 26m |
This talk will demonstrate an example application of using Celery to extract all of the lyrics of the inspiring and influential Taylor Swift from the Internet… | ||
| HistomicsTK: An open source python toolkit for web based analysis of dig | Deepak Roy Chittajallu | 28m |
HistomicsTK claims an open-source Python toolkit that performs web-based analysis of whole-slide histopathology images (50,000 px tiles, ~7.5 GB, ~1M nuclei… | ||
| Online customer targeting as a classification problem | Ulric Wong | 38m |
Predicting conversions from retargeting data is framed as a supervised classification problem with heavy class skew and censoring and Maxpoint supplies a… | ||
| Balancing scale and interpretability in analytical applications with sklearn and e | Lanhui Wang | 40m |
PyData Carolinas 2016 describes an implemented stacking framework that partitions a large feature space into analyst-specific subsets, fits scikit-learn… | ||
| Just Bring Glue Leveraging Multiple Libraries To Quickly Build Powerful New Tools | Rob Agle | 42m |
Two small projects: snakecam finds tests and runs them by parsing test names, caching timestamps and spawning a child process (about eight imports, ~231 lines)… | ||
| Identifying Racial Bias in Policing Practices: Open Data Policing | Colin Copeland | 39m |
A web app called OpenDataPolicingNC.com aggregates ~20,000,000 North Carolina traffic stops, exposes agency and officer dashboards via an API architecture, and… | ||
| Julia for Modern Data Analysis | Josh Day | 30m |
Julia claims high performance through JIT compilation and multiple dispatch and shows concrete tooling: interoperates with Python via PyCall and R via RCall… | ||
| Python, C, C++, and Fortran Relationship Status: It’s Not That Complicated | Philip Semanchuk | 36m |
Clear actionable claims: three practical paths — wrap / extension / embed — with concrete pros and cons; tools that actually work include ctypes (no C++), cffi… | ||
| PyData in Production: Lesson Learned from Various PyData Deployment Strategies | Josh Howes | 39m |
Describes concrete deployments of the PyData stack: Pandas and sklearn models run behind a Java service layer using Apache Thrift as an RPC shim, interactive… | ||
| Stemgraphic - A Stem and Leaf Plot for the Age of Big Data | Francois Dion | 37m |
A Python module that revives the stem-and-leaf for large data claims to let analysts read individual values while scaling to production sizes it provides a CLI… | ||
| Lightning Talks | Peter bomgardner | 44m |
A set of five-minute demos shows concrete results: a Twitterbot that generates city accolades using NLTK and WordNet and runs on Heroku with a scheduler; a… | ||
| nothing matches | ||