Past Meetup

November 2017 Lightening Talk Night

This Meetup is past

123 people went

Location image of event venue

Details

Agenda:
6:30 - Pizza and Networking:
7:00 - Announcements
7:10 - Bohdan Khomtchouk, Ph.D. (Stanford University): Outrageously fast interactive heatmap graphics using R, JavaScript, and Plotly
7:25 - John Sheehan: Taming the Shiny DT::datatable Beast
7:40 - Mike Gahan: Labeling Images with the WanderingEye Package
7:55 - Stoney Vintson: Multi-layer maps with flexdashboard
8:10 - Pete Mohanty: From Data to Document: A Practical Guide to Parameterizing Reports
8:25 - Joseph Rickert: Working with TensorFlow from R

#-------------------------

Bohdan Khomtchouk, Ph.D. (Stanford University)

Outrageously fast interactive heatmap graphics using R, JavaScript, and Plotly

We will discuss the world's fastest interactive heatmap software built with a combination of R, JavaScript, and Plotly. This talk is based on open source peer-reviewed published work: http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0176334

#------------------------

John Sheehan

Taming the Shiny DT::datatable Beast

John will describe a function that helps create a datatable from a database view or table with lots of control over formatting (shown below), using a simplified spec in a csv file. The csv file has a row for each column, specifying the database column name, the name for the datatable column, the data type (text, bigtext, float, int, date, percent, …), and optional # digits for float, and finally optional column width. "Bigtext' truncates the text, showing the full text on hover. A typical invocation is

output$mytable <-DT::renderDataTable(getViewToDT("myview")

In addition to making it easier to code, it allows changes to the underlying table and how it is displayed with no change to the Shiny code.

#--------------------

Mike Gahan

Labeling Images with the WanderingEye Package

WanderingEye is an R package that allows the user to capture image label data from the following APIs:

AWS Rekognition ( https://aws.amazon.com/rekognition/ )
Google Cloud Vision ( https://cloud.google.com/vision/ )
Microsoft Computer Vision ( https://azure.microsoft.com/en-us/services/cognitive-services/computer-vision/ )
IBM Watson Visual Recognition ( https://www.ibm.com/watson/services/visual-recognition/ )
ClarifAI ( https://www.clarifai.com/ )

The package was initially created at the San Diego Zoo Hackathon in order to identifier animal trap image data. Each API has their strengths and weaknesses but when all the label data is used as an ensemble, the prediction power is much greater. I would like to talk about how to use the package and some of its main features. https://github.com/mgahan/WanderingEye

#----------------

Stoney Vintson

Multi-layer maps with flexdashboard

Examples will include CDC data mapped in flexdashboard using tmap to leaflet and NYTimes geo visualization recreation with ggplot2

#---------------

Pete Mohanty

From Data to Document: A Practical Guide to Parameterizing Reports

Using the example of survey data, this talk walks through automating reports with R and RMarkdown using RStudio using knitr. This is great for portions of the document that don't change (e.g., "the survey shows substantial partisan polarization"). The motivation is really twofold: efficiency (maximize the reusability of code, minimize copy and pasting errors) and reproducibility (maximize the number of people and computers that can reproduce findings). I provide additional tips for collaborating as well as highlights of demo code, which can be found here: https://github.com/rdrr1990/datascience101/blob/master/automating/mohanty_automation_guide.md

#------------------

Joseph Rickert

Working with TensorFlow from R