SATRDAY
The R community and some of South Africa's most forward thinking companies have come together to bring satRday back for its third edition. This conference provides an opportunity to hear from and network with top Researchers, Data Scientists and Developers from the R community in South Africa and beyond.
Some pictures from Joburg's first satRday in April 2019.
Speakers
Keynote Speakers
Omayma Said
Data Scientist
Freelancer
Omayma is a data scientist with a background in electronics engineering and holds a master's degree in Systems Engineering and Engineering Management.
Prior to becoming a freelancer, Omayma worked on developing algorithms and analyzing users data at a tech company which providies an online marketplace for jobs. She enjoys working in data-informed environments where data analysis can drive well-informed decisions, support building products, create value and help in improving the overall quality of research.
She shares her passion for all things data within the R-ladies community, where she has instructed hands-on workshops, as a DataCamp instructor as well as in her blog suitably titled "Once Upon Data".
Danielle Navarro
Computational Cognitive Scientist
University of New South Wales
Danielle leads the Computational Cognitive Science lab, a research group that uses a mix of experimental methods and computational modelling techniques to explore how the human mind works. Her research focuses on human concept learning, reasoning and decision making.
Dani's self-titled blog focuses on both her work as well as some of her other interests including; language and cultural evolution, cognitive development and statistical methods in the behavioural sciences.
She is also one of the co-founders of R-Ladies Sydney.
2019 Speakers
Cesaire Tobias
Consultant
Fractal Value Advisors
Megan Beckett
Data Scientist
Exegetic Analytics
Amieroh Abrahams
Masters Student
University of the Western Cape
Marinus Louw
Junior data Scientist
Exegetic Analytics
Luis de Sousa
Principal Consultant
Syeop
Natasha Sing
Senior Data Analyst
Derivco
Oscar Baruffa
Strategic & Business Planner
PETCO
Bianca Peterson
Director
Conquest Analytics & Training
Jeff Fletcher
Machine Learning Field Engineer
Cloudera
Andrew Collier
Data Scientist
Exegetic Analytics
Francois Van Heerden
Senior Analytics Researcher
Kantar
Helge Reikeras
Data Scientist
OfferZen
Gerard Walsh
Junior data Scientist
Exegetic Analytics
Vishalin Pillay
Data Analyst
Derivco
Ekow Duker
co Founder of Zindi
Ixio Analytics
Peter Kamerman
Physiologist
University of the Witwatersrand
Michael Johnson
BI Architect
SQLSA
Robert Bennetto
CTO
Chalcid
Hanjo Odendaal
Data Scientist
Stellenbosch, BER, Eighty20
Quintin De Kok
Solutions Architect
Netstar
Eckhard Briedenhann
Backend specialist
Chalcid
Workshops
satRday Johannesburg will kick off on the 5th of April 2019 with a day of workshops instructed by our Keynote speakers. Note that these will be full day workshops and take place in parallel, so you have to choose between them - decisions decisions!
(*) Workshops available to training pass holders only.
Building web applications with Shiny
Omayma Said
Shiny provides an easy way to create interactive web applications using R to share analysis, build dashboards and more. This workshop will focus on building Shiny applications from scratch.
By the end of the tutorial, you will:
- Understand the structure of Shiny applications (ui and server functions)
- Have learnt the basics of reactivity and how to use it to build more efficient applications
- Be able to customize the appearance of Shiny applications
- Know the different ways to share a Shiny application
Prerequisites: Basic knowledge of R
Data visualisation in R, from ggplot2 to gganimate
This hands on workshop will provide an introduction to data visualisation in R. The core of the workshop will focus on creating clean data visualisations with ggplot2, and turning them into animations with gganimate. The workshop is aimed at a beginner to intermediate level. Some prior familiarity with tidyverse (dplyr, ggplot2) would be helpful but not assumed, as the workshop will include a refresher / quick tutorial for beginners.
Topics covered will include:
- Building and customising plots with ggplot2
- Understanding how ggplots are structured
- Building animated plots with gganimate
- Links between human vision and data visualisation
- Interactive plots with ggplotly
- Under the hood: graphics devices, grid graphics, etc
Data Carpentry Workshop for R novices
Data Carpentry workshops teach introductory computational skills needed for data management and analysis in all domains of research. For tickets and more information on the syllabus please visit the
workshop webpage.
Registration for this workshop is separate from satRday.
Programme
Workshops Programme
| Start | End | Friday 5 April 2019 |
|---|
| 8:30 |
9:00 |
Registration |
| 9:00 |
10:30 |
First Session |
| 10:30 |
11:00 |
Tea / Coffee
|
| 11:00 |
12:30 |
Second Session
|
| 12:30 |
13:30 |
Lunch
|
| 13:30 |
15:00 |
Third Session |
| 15:00 |
15:30 |
Tea / Coffee
|
| 15:30 |
17:00 |
Fourth Session
|
Conference Programme
Standard talks are 20 minutes and lightning () talks are a mere 5 minutes.
Click on the title for any talk to view the details.
| Start | End | Saturday 6 April 2019 |
|---|
| 8:00 |
8:30 |
Registration |
| 8:30 |
8:35 |
Welcome |
| 8:35 |
9:45 |
First Session
Katrin Tirok
- Understanding the state of R in South Africa: 2018 in review (Francois Van Heerden)
Most of us make use of CRAN daily. It is a central part of the R experience and it also turns out to be a rich source of data! By analysing the downloads from CRAN, we can get a deeper understanding of how R is being used within SA. Are you curious about package trends within SA?
- What I learned from running Shiny in production (Luis de Sousa)
In this talk I share my experience of the do's and don'ts of running Shiny in production. We'll also cover some of the guidance Joe Cheng (original creator of Shiny) gave earlier this year at rstudio::conf 2019.
- One-liners: 3 x packages with single-line calls to boost your productivity (Oscar Baruffa)
1 fun and 2 seriously-fun packages. Who couldn't do with boosting productivity? And in only one line of code! Get a friendly sound notification in your scripts, automate the slog of early EDA, and make sure your dplyr calls aren't silently failing you!
- Geo-spatial data visualization in R (Natasha Sing)
Guess what's the buzzing “V” word these days?
Veganism! My talk investigates the growing acceptance of veganism around the world using geospatial visualization of data in R. Leveraging Google's Map and Geocoding API keys in R we can create maps, plot data points and explore the world with fun.
- The Dota 2 Landscape in R (Vishalin Pillay)
Dota 2 is a massively popular multiplayer online battle arena (MOBA) video game, comparable to strategy games like Chess or Go, but with infinitely more degrees of freedom. Can we use the data wrangling and graphical powers of R to uncover the ever-changing landscape of the professional scene?
- Why tidy data matters (Gerard Walsh)
Having your data in a form that is easily interpreted by humans is helpful for machine processing too. Tidy data, where a column of your data is a variable and each row is an observation, allows us to easily understand the data and utilize off-the-shelf tools to manipulate/visualize the data.
|
| 9:45 |
10:30 |
Keynote: Omayma Said
Gemma Dawson
Interpreting Machine Learning Models: Why and How!
With the increasing adoption of machine learning-based solutions in different domains, the impact on humans is becoming more significant. Systems that use black-box algorithms in crucial areas like health care, employment, politics are getting used more often with the promise of providing higher accuracy. However, this accuracy comes at the cost of interpretability . This introduces a barrier against wider adoption of such algorithms in sensitive areas and raises the skepticism of the impacted individuals. This talk will focus on interpretable machine learning, why it is crucial from technical and ethical perspectives and how to integrate it in our workflows. We will go over some of the R packages that provide interpretations such as {lime} and {iml}, while giving real examples and playing a game!
|
| 10:30 |
11:00 |
Coffee/Tea
|
| 11:00 |
12:15 |
Second Session
Vebashini Naidoo & Emma Collier
- Petabytes of data: How to use R at scale (Jeff Fletcher)
You have great R programming skills and a laptop that's ready to go, but you're suddenly faced with having to process a trillion rows or a petabyte of JSON. This talk will run through the tools used to process very large data sets for R programmers, whether for data science or machine learning.
- How the cloud changed the way I do analytics and research (Hanjo Odendaal)
When we get into analytics our first real hit of power is estimating a Random Forest using the Titanic dataset. We quickly progress from this in search of more cores and RAM, quickly hitting a wall. Moving your analytics to the cloud is analogous to moving from coffee to cocaine... just WAY cheaper!
- A Fiery approach to R web services (Quintin de Kok)
The final step from modeling to production is almost always the most difficult, specially in the realtime analytics arena. This talk is designed to ease the progression from modeling to realtime prediction/fit using ProjectTemplate and Fiery to build a Docker based prediction web service.
- Zindi Competition Platform (Ekow Duker)
With over 1,600 registered users, Zindi is the largest online Pan- African community of data scientists solving the continent’s most pressing challenges using artificial intelligence and machine learning.
- Detecting marine heatwaves (MHWs) by using R-programming (Amieroh Abrahams)
Anthropologically mediated warming has negatively affected marine and terrestrial ecosystems. This has resulted in consequences for humanity and ecological functioning. Here I will demonstrate the various R packages produced and utilized by South African scientists to detect these extreme events.
|
| 12:15 |
13:15 |
Lunch
|
| 13:15 |
13:45 |
Panel Discussion
David Clark
- Ethics in Data Science
Panel:
- Danielle Navarro
- Vebashini Naidoo
- Retha Langa
- Peter Kamerman
- Robert Bennetto
|
| 13:45 |
14:45 |
Third Session
Bianca Peterson
- Exploring the South African School Directory with R (Megan Beckett)
South Africa’s education system is incredibly expansive and diverse. Let’s see what this actually means by taking some seemingly forgotten Excel files from the Department of Education website, improve the school geocoding then create an interactive, exploratory, enlightening map of our schools.
- Reference classes: An antidote to complexity (Eckhard Briedenhann & Robert Bennetto)
R is a fantastic language for performing vectorised calculations, but implementing complex design patterns can turn into a nightmare when lists and vectors are all you have to work with. We'll provide some templates of how to manage complexity through the S3/S4 class system and showcase the usefulness with some Directed Acyclic Graph (DAG) examples.
- YOLO in R (Vongani Maluleke)
Object detection is a continuously advancing field with new, better and more complex models being introduced every year, making it hard for an ordinary R-user to keep up. YOLO is one of the best object detection models that an ordinary R-user can implement in 2 lines of R code with no GPU usage.
- Three indispensable REGEX hacks (Marinus Louw)
Having regular expressions (or REGEX for short) in your tool kit is an essential data science skill which allows you to do complicated text manipulation with minimal effort.
|
| 15:00 |
15:30 |
Coffee/Tea
|
| 15:30 |
17:10 |
Fourth Session
Amieroh Abrahams
- How to write a report using R Markdown (Bianca Peterson)
Learn how to write elegant, fully reproducible reports using R Markdown, with dozens of output formats to choose from, e.g. Word, PDF, HTML. Reports can include both narrative text, tables and figures, with the option of also displaying code chunks. Writing reports have never been this easy!
- EXCELsioR! (Cesaire Tobias)
The ubiquity of spreadsheets as a source of data should be celebrated rather than loathed. The freedom of the gridded blank canvas allows users to create almost unhindered. While this can cause pain for the data-munger it creates opportunities to work with data we otherwise wouldn't have had.
- How to Get Ahead in Advertising (Helge Reikerås)
John Wanamaker said, “half the money I spend on advertising is wasted; the trouble is I don't know which half". The task of data scientists in marketing is finding which half. In this talk we'll discuss models and methods for the so-called multi-channel attribution problem, with implementation in R.
- Extending Power BI with R (Michael Johnson)
Power BI is a great tool for self-service BI but lacks some of the advanced analytics capabilities available in other tools. By incorporating R support into the Power BI application Microsoft enables a true hybrid analytics experience.
- Deploying a Plumber API with Docker (Andrew Collier)
Plumber makes it possible to easily transform your R functions into an API. Developing and testing the API locally is simple, but deploying it can be more of a challenge. Docker provides just what you need to ensure that the API is launched in a well defined, stable and isolated environment. These are some of the handiest tools to have in your toolbox!
This short talk will show how to:
- whip up a quick API by merely decorating your existing R code with comments, then
- write a simple Dockerfile to provide a runtime environment and finally
- deploy the API in the cloud.
All of this in the timeframe of a lightning talk. Booya!
- Become an animaniac with gganimate (Peter Kamerman)
Who doesn't enjoy a good GIF. I'll give a brief example of how you can bring data to life using gganimate. For my example, I'll illustrate changes in major causes of disability adjusted life years over the past 20 years (using open data).
|
| 17:10 |
17:15 |
RStudio Student Lightning Talk Prize & Closing
|
satRday R-Ladies lunch meetup
When: April 6, 12:15 - 13:15
Both satRday keynote speakers, Omayma Said and Danielle Navarro, are involved in R-Ladies. This is an international organisation that strives to promote gender diversity in the R community.
This event, although primarily aimed at women, welcomes all who are interested as it will provide a great opportunity for all current and prospective R users to meet and chat with like-minded people.
If you are interested in attending similar events in the future please get in contact with us via our page or our .
Please note this meetup is only open to those attending satRday.
Venue
This year satRday will be hosted at the prestigious Discovery Head Office in Sandton. Situated in the hub of Johannesburg the venue boasts a 5-star Green Star SA rating by the GBCSA.
Registration
Workshop Pass
A workshop pass gives you access to one of the
workshops held by our Keynote Speakers. These full day workshops will be run in parallel on the 5th April
2019.
|
Workshop Pass
(Early Bird R 1000.00) |
| 1 Day Workshop |
|
| Lunch |
|
| Networking breaks with refreshments provided |
|
Conference Pass
The Conference Pass gives you access to South Africa's third sat
Rday. Come and join us on the 6th April 2019 to meet with and hear from both local and international R enthusiasts!
|
Conference Pass
(Early Bird R 200.00) |
| All Conference Talks |
|
| Lunch |
|
| Networking breaks with refreshments provided |
|
Important Dates
Working up to the conference on 6 April 2019, these are the most important dates on your calendar:
Call for Papers
The deadline for talk submissions is 11 March 2019.
What are the benefits of giving a presentation?
- Speakers attend the conference for free!
- Prompt feedback on your proposed presentation.
Feel free to submit one or more proposal(s). We will be considering the following presentation formats:
- Tutorial [60 min]
- Lightning Talk [5 min]
- Standard Talk [20 min]
If you run into any problems while trying to submit your proposal please send us an
email.
Code of Conduct
satRday is dedicated to providing a harassment-free and inclusive conference experience for all in attendance regardless of, but not limited to, gender, gender expression, sexual orientation, employment, disabilities, physical attributes, age, ethnicity, social standing, religion or political affiliation.
Any form of harassment of participants involving anyone involved with satRday will not be tolerated. Sexual innuendos and imagery are not appropriate for any conference venue, including presentations.
Anyone violating these rules may be given a warning or expelled from the conference (without a refund) at the discretion of the conference organisers.
Our full code of conduct/anti-harassment policy can be found here.
