Since 2015, Data Science Africa (DSA) has held an annual summer school and workshop to train participants on machine learning and data science methods and provide an avenue for researchers to present work demonstrating the application of these techniques to problems relevant in the African context.

For 2018, there will be two events: DSA 2018 Nyeri in Kenya, May 31st to June 8th and DSA 2018 Abuja in Nigeria, 12th to 16th November, representing our goal to consolidate the regional community in East Africa while also expanding to other regions. DSA 2018 Nyeri will focus on training approximately 30 participants who will then be able to serve as trainers in their local contexts. DSA 2018 Nyeri will be a "Training of Trainers" event under the theme "end-to-end data science."

This year's speakers and instructors include:

Summer School on Machine Learning and Data Science
Dates: 31 May - 06 June 2018
Venue: Dedan Kimathi University of Technology, Nyeri, Kenya
 

We would like to expose data science trainers to the entire pipeline of data collection, analysis, and communication of results using relevant examples. We plan to have participants deploy sensor systems at the DeKUT conservancy and farm which will collect data that they will then analyse.

To benefit from this course participants must have some background in programming particularly programming with Python and machine learning. The registration process will include submission of worked examples in Jupyter notebooks. Registration is now OPEN.

School programme outline:

Draft Lecture Schedule

Configure a DSA Environment with Anaconda

Conda is an open source package management system and environment management system for installing multiple versions of software packages and their dependencies and switching easily between them. It works on Linux, OS X and Windows, and was created for Python programs but can package and distribute any software.

Overview

Using Anaconda consists of the following:

  1. Install miniconda on your computer
  2. Create a new conda environment using this project
  3. Each time you wish to work, activate your conda environment

Installation

Download the latest version of miniconda that matches your system.

  Linux Mac Windows
64-bit 64-bit (bash installer) 64-bit (bash installer) 64-bit (exe installer)
32-bit 32-bit (bash installer)   32-bit (exe installer)

Install miniconda on your machine. Detailed instructions:

  • Linux: http://conda.pydata.org/docs/install/quick.html#linux-miniconda-install
  • Mac: http://conda.pydata.org/docs/install/quick.html#os-x-miniconda-install
  • Windows: http://conda.pydata.org/docs/install/quick.html#windows-miniconda-install

Setup the dsa2018 environment.

git clone https://github.com/emwebaze/dsa2018materials.git
cd setup

Create dsa2018. Running this command will create a new conda environment that is provisioned with most of the libraries you need for this summerschool.

conda env create -f dsaenvironment.yml

Verify that the dsa2018 environment was created in your environments:

conda info --envs

Cleanup downloaded libraries (remove tarballs, zip files, etc):

conda clean -tp

Uninstalling

To uninstall the environment:

conda env remove -n dsa2018

Using Anaconda

Now that you have created an environment, in order to use it, you will need to activate the environment. This must be done each time you begin a new working session i.e. open a new terminal window.

Activate the dsa2018 environment:

OS X and Linux

$ source activate dsa2018

Windows

Depending on shell either:

$ source activate dsa2018

or

$ activate dsa2018

That’s it. Now you can fire up your Jupyter Notebook from this terminal and it will load all the necessary libraries.

To exit the environment when you have completed your work session, simply close the terminal window.

Troubleshooting and comments..

Use the comment section below to (a) ask questions that are not already answered (b) help your peers by providing answers to their questions, if you can.

Time

Activity

Material

08:00-09:00

Arrival and Registration

09:30-09:30

Welcome and Opening Remarks

09:30-10:30

Lecture 1: Introduction to Data Science
Dina Machuve - Nelson Mandela African Institute of Science and Technology

slides

10:30-11:00

Break

11:00-12:00

Lecture 2: Fundamentals of IoT
Jan Jongboom - ARM

slides

12:00-13:00

Practical Session: IoT

13:00-14:00

Lunch

14:00-15:30

Field work introduction, Prep IoT for FW1

15:30-17:30

Field Work 1: IoT Greenhouse deployment
Jan Jongboom (ARM), Ciira Maina (DeKUT)

Time

Activity

Material

08:30-09:30

Anomaly detection, density estimation tutorial
Tom Dietterich (Oregon State University)

slides

09:30-10:30

Python, Pandas and Jupyter tutorial
Claire Babirye (UTAMU) and Ben Akera (Makerere University)

ipynb

10:30-11:00

Break

11:00-12:30

TAHMO Weather station network and data control task
Tom Dietterich (Oregon State University)

slides ipynb

12:30-14:00

Lunch

14:00-15:30

Classification Tutorial & Research Software
Mike Smith (Sheffield University)

slides ipynb

15:30-17:30

Field Work 2: Camera Trap Deployment
Jan Jongboom (ARM), Ciira Maina (DeKUT)

Time

Activity

Material

08:30-09:30

Reinforcement Learning for ecosystem management
Tom Dietterich (Oregon State University)

slides

09:30-10:30

Computer Vision and Image Analysis
Earnest Mwebaze (Makerere, UN Pulse Lab)

ipynb

10:30-11:00

Break

11:00-12:30

Data Visualization Tutorial
Martin Mubangizi (Makerere, UN Pulse Lab)

ipynb

12:30-14:00

Lunch

14:00-15:30

Data Engineering and Infrastructure
Gen-Tao, Damon Civin (ARM)

slides ipynb

15:30-17:30

Field Work 3: Air Quality Sensor Deployment
Jan Jongboom (ARM), Ciira Maina (DeKUT)

Time

Activity

Material

08:30-09:30

Bayesian Methods
Neil Lawrence (Amazon, University of Sheffield)

slides ipynb

09:30-10:30

Deploying Models
Damon Civin (ARM)

slides ipynb

10:30-11:00

Break

11:00-12:30

Bayesian Methods Practical
Neil Lawrence (Amazon, University of Sheffield)

12:30-14:00

Lunch

14:00-15:00

Spatial Data Analysis
John Quinn (UN Pulse Lab, Makerere)

ipynb

15:00-15:30

Break

15:30-17:30

Spatial Data Analysis Practical
John Quinn (UN Pulse Lab, Makerere)

Time

Activity

Material

08:30-09:30

Engineering human vision : an introduction to convolutional neural networks
Sara Hooker (Google Brain)

slides

09:30-10:30

Introduction to Deep Learning
Max Welling (University of Amsterdam)

slides

10:30-11:00

Break

11:00-12:30

Practical Deep Learning
Max Welling (University of Amsterdam), Julius Adebayo (Google Brain)

13:30-14:30

Lunch

14:30-16:00

Deep Learning Practical: Tensorflow
Julius Abebayo (Google Brain)

ipynb

16:00-17:00

Interpretability, Bias in Models
Julius Adebayo (Google Brain)

Time

Activity

Material

08:30-09:30

Advanced Deep Learning
Max Welling (University of Amsterdam)

slides

09:30-10:30

Introduction to Natural Language Processing (NLP)
Mark Magumba (Makerere University)

10:30-11:00

Break

11:00-12:30

Deep Learning Practical II
Max Welling (University of Amsterdam), Julius Adebayo (Google Brain)

12:30-14:00

Lunch

14:00-15:00

Adversarial Methods
Moustapha Cisse (FAIR)

15:00-15:30

Break

15:30-16:30

Closing Panel
Ernest Mwebaze

Data Science in Africa Workshop
Dates: 07 June - 08 June 2018
Venue: Dedan Kimathi University of Technology, Nyeri, Kenya

Call for Registration

The workshop will be organized around paper presentations and interactive panel discussions. We invite participants interested in presenting work at the workshop to submit a short abstract describing the application of data science methods to problems relevant to Africa. The submission should be limited to a 3 page (max) submission following this LaTeX template. You will be notified whether your work will be presented orally or as a poster. We are interested in work addressing low resourced developing world challenges in the broad areas of:

  • Data Science for the Sustainable Development Goals
  • Healthcare
  • Agriculture
  • Wildlife conservation
  • Disaster response
  • Geospatial modelling
  • Telecommunications data modelling
  • Economic monitoring

During the panel discussions, we will unite a wide range of stakeholders, including data scientists, representatives from government, development practitioners and the private sector; this will provide a unique setting in which innovative solution driven ideas can thrive.

Participants will also develop a framework for attracting young African talent, mentors and researchers from academia, the public sector and the private sector in Africa to engage in activities geared towards harnessing big data and real-time analytics for the public good.

Workshop programme outline:

Draft Schedule

DSA 2018 Nyeri Workshop Day 1

Time

Presentation

08:00-09:00

Arrival and Registration

09:00-09:30

Workshop Opening
Vice Chancellor - DeKUT, General Chair, Program Chair

09:30-10:00

Keynote 1
Allela Roy - Intel

10:00-10:30

Keynote 2
Martin Mubangizi - Pulse Lab Kampala

10:30-11:00

Coffee Break

11:20-11:40

Talk 2 - Towards 2050: Big data driven decision support in integrated smallholder farming systems
Denis Mijibi (USOMI)

11:40-12:00

Talk 3 - How do models learn: understanding feature importance in image classification models
Sara Hooker (Google Brain)

slides

12:00-12:20

Talk 4 - Transfer learning for mobile phone credit scoring
Skyler Speakman - IBM Kenya

slides

12:20-13:00

Panel 1 - Data for public good
Moderator: Neil Lawrence. Panelists:[Martin Mubangizi, John Quinn, Mpho Mokoatle, Sam Hooker, Allela Roy,...]

13:00-14:00

Lunch Break & group photo

14:00-14:30

Keynote 3 - TBA
Kathleen Siminyu

14:30-14:50

Talk 5 - TBA
Joseph Orero (Strathmore University)

14:50-15:10

Talk 12 - Mapping exposure of cattle to rift valley fever virus along their migratory routes
Charles Mundia (DeKUT)

15:10-15:30

Talk 7 - Machine learning approach on reducing student dropout rate
Neema Mduma (NMAIST)

15:30-16:00

Tea Break

16:00-16:40

Panel 2 - Building local capacity - Africa Data Science Community
Moderator: John Quinn. Panelists:[Joseph Orero,Kathleen, Ciira, Dina,...]

16:40-17:00

Day 1 Wrap up
Billy Okal

DSA 2018 Nyeri Workshop Day 2

Time

Presentation

08:00-09:00

Arrival and Registration

09:00-09:30

Announcements & logistics
General Chair - Ciira Maina

09:30-10:00

Keynote 4
Timnit Gebru - Microsoft Research

10:00-10:30

Keynote 5
Tom Diettriech - Oregon State University

10:30-11:00

Coffee Break

11:00-11:20

Talk 8 - A real-time simplified smart agriculture system for small scale greenhouse farming
Denis Rubanga Pastory, Katsumori Hatanaka and Sawahiko Shimada

11:20-11:40

Talk 9 - Data-driven patient diagnosis with Dr. Elsa
Ally Jr Salim and Megan Allen

11:40-12:00

Talk 10 - Air quality monitoring
Mike Smith - Sheffield University

12:00-12:30

Panel 3 - Applications of data science in government
Moderator: Martin Mubangizi. Panelists:[Osbert Osuman, Andrew, Tom Diettriech, Timnit Gebru]

12:30-12:40

Talk 11 - DeKUT IoT and Data science projects
Jared Makario

12:40-13:00

Spotlight talks
Various

13:00-14:00

Lunch Break

14:00-14:30

Keynote 5 - TBA
Morris Agaba

14:30-14:50

Talk 12 - Converging blockchain and next-gen AI technologies for biomedical research
Iraneus Ogu (InSilico Medicine)

14:50-15:10

Talk 13 - Why Data Science is so important in Off Grid Solar
Alexandre Roussel (Fenix International)

15:10-15:30

Tea Break

16:00-16:20

Panel 4 - Lessons from DSA 2018
Moderator: Ernest Mwebaze.

16:20-16:30

Closing
Ciira Maina

Collaborators

Sponsorship

Gold Sponsors

Silver Sponsors

Sponsoring Data Science Africa 2018 Event is a great way to communicate your commitment to support the achievement of the sustainable development goals. To become a sponsor, please contact Data Science Africa at info@datascienceafrica.org