Want a look into the Africa Deep Learning Indaba 2019?

Audacious Coder.
9 min readSep 6, 2019
DEEP LEARNING INDABA 2019

Hello fellow techies and tech enthusiasts. It's me Audacious obviously🤷‍♂️(trying to imitate youtube content makers😂). This blog is not part of the content I was doing previously but it is the same concept, deep learning.

So am about to bore you …….👀nuuuh, am about to blow your minds on the events that took place in The Annual Deep Learning Indaba 2019.

What is The Deep Learning Indaba?

This is an annual meeting of the African machine learning community that is held in Africa. The mission of the Deep Learning Indaba is to Strengthen African Machine Learning. In 2019, the Indaba aimed to see 700 members of Africa’s artificial intelligence community for a week-long event of teaching, research, exchange, and debate around the state of the art in machine learning and artificial intelligence. Yeah, a week-long hence am breaking this blog into three parts and most in which I will be heading straight to the point cause honestly, long literature bores me.

The event was so amazing and a humble shout out to Kenyatta University based in Kenya(just being patriotic)for hosting us. The event was scheduled to begin from the 25th- 30th of August 2019. In which it started out the exact day and time. I wonder how African timing did not affect this timing 😁but it eventually did🤣#tough love.

#SautiYetu: Raising Our Voice in AI

With each year, our self-confidence and self-ownership in the ongoing advances and impact of data science, machine learning and artificial intelligence (AI) become stronger — here in Kenya, in east Africa, and across our continent. Self-ownership is why we organize, why we build an interconnected African community and is what it means to Strengthen African Machine Learning. And so we meet together for the 3rd Indaba, proudly here in Nairobi, in Kenya. Our continent has faced many difficulties over the last year and with many political and humanitarian challenges, like the crises in Sudan and Ethiopia and Algeria, or the fierce debate over the rights of LGBT African citizens in Kenya and Botswana. And all this is happening while we are also struggling with the challenges of jobs for young people, and mitigating the risks of climate change.

Machine learning and artificial intelligence can play a role in addressing these problems. Our task is to be the people that help say when and how this might be possible. We must learn to be critical and to be able to see how the work we do can be used to responsibly address our continent’s and our world’s challenges.

We are up to the challenge. In the last year, we saw local leaders in AI rise up, organizing IndabaX days in 27 different countries. We saw the past winners of the Kambule and Maathai awards go on to have greater impact and spread their work wider. We saw our friends and partner groups, like Data Science Africa, Data Science Nigeria, AI Saturdays, WiMLDS and AIMS, all go on to build stronger communities and deeper expertise. At the last Indaba, our action was for Masakhane — We Build Together. Masakhane is what we have done: the responsibility for uplifting our communities is as strong as it has ever been. This year we say Sauti Yetu — Our Voice.

Our voice can be strong; our voice can create change; we have many different voices that must be heard. Use the Indaba as part of your voice. As we get older, some people attending the Indaba for the 3rd time, we must move to do more. As the Indaba gets closer, we ask that all of us together raise our voice, to be more confident, to take more ownership, and to build the technology that will help us responsibly build the next age of pan-African unity and prosperity. Our Voice for AI. Our Voice for Africa. Sauti Yetu Kwa AI ya Afrika.

DAY 1.

My day one started on Sunday as I attended 3 sessions of my choosing for there were many sessions and one picks the most favorable for their path.

The Sunday timetable,25th 2019.

DAY SESSION ON PYTHON REFRESHER COURSE

Why python?

The scientists' needs.

  • Get data (simulation, experiment control),
  • Manipulate and process data,
  • Visualize results, quickly to understand, but also with high-quality figures, for reports or publications.

PYTHON BENEFITS

  • Easy to learn Most scientists are not paid as programmers, neither have they been trained so. They need to be able to draw a curve, smooth a signal, do a Fourier transform in a few minutes.
  • Easy communication To keep code alive within a lab or a company it should be as readable as a book by collaborators, students, or maybe customers. Python syntax is simple, avoiding strange symbols or lengthy routine specifications that would divert the reader from a mathematical or scientific understanding of the code.
  • Efficient code Python numerical modules are computationally efficient. But needless to say that a very fast code becomes useless if too much time is spent writing it. Python aims for quick development times and quick execution times.
  • Universal Python is a language used for many different problems. Learning Python avoids learning new software for each new problem.

PYTHON PROS

  • Very rich scientific computing libraries
  • Well thought out language, allowing to write very readable and well-structured code: we “code what we think”.
  • Many libraries beyond scientific computing (web server, serial port access, etc.)
  • Free and open-source software, widely spread, with a vibrant community.
  • A variety of powerful environments to work in, such as IPython, Spyder, Jupyter notebooks, Pycharm, Visual Studio Code

Cons

Not all the algorithms that can be found in more specialized software or toolboxes

This link contains

  • Basic Python: Basic data types (Containers, Lists, Dictionaries, Sets, Tuples), Functions, Classes
  • Numpy: Arrays, Array indexing, Datatypes, Array math, Broadcasting
  • Matplotlib: Plotting, Subplots, Images
  • IPython: Creating notebooks, Typical workflows

Link to the practical content…

NOON SESSION: INTRODUCTION TO TENSORFLOW.

What is TensorFlow?

TensorFlow is an open-source machine learning framework for all developers. It is used for implementing machine learning and deep learning applications. To develop and research on fascinating ideas on artificial intelligence, Google team created TensorFlow. TensorFlow is designed in Python programming language, hence it is considered an easy to understand framework.

Importance/Benefits of the TensorFlow?

TensorFlow is well-documented and includes plenty of machine learning libraries. It offers a few important functionalities and methods for the same.TensorFlow is also called a “Google” product. It includes a variety of machine learning and deep learning algorithms. TensorFlow can train and run deep neural networks for handwritten digit classification, image recognition, word embedding and creation of various sequence models.

BUILD YOUR OWN TENSORFLOW.

Important Note: Please note that most of the Specialized sessions require pre-work in order for you to get the most out of them. This is for the more specialized people in the sector.

Build your own TensorFlow: This practical covers the basic idea behind Automatic Differentiation, a powerful software technique that allows us to quickly and easily compute gradients for all kinds of numerical programs. We will build a small Python framework that allows us to train our own simple neural networks, like Tensorflow does, but using only Numpy. NOTE: This practical is particularly long, so coming adequately prepared is very important.

  • Pre-work: Please read through all the background sections in the practical.
  • Background knowledge requirements:
  • Linear algebra (multiplying matrices, row vectors, column vectors, summation notation)
  • Calculus (derivatives and partial derivatives, Jacobian matrix)
  • Deep Learning (have used a framework like TensorFlow or PyTorch before)

//Day one ended with an onboarding party at some dope golfclub.

DAY TWO.

This day started quite lovely. So we sang the Kenya National Anthem led by Kathleen Simiyu. It was nice to share our heritage with other African cultures and have some of them wonder what the hell we were saying(anthem in Swahili)shout out to my Congolese friend😂🌚.Then followed by an official welcome from Vice-Chancellor of KU, Prof. P. Wainaina.

Dr. Aisha Walcott-Bryant

The second session was a keynote from Aisha Walcott-Bryant. Dr. Aisha Walcott-Bryant is a research scientist and manager at IBM Research Africa — Nairobi, Kenya. Her current research makes use of AI and Blockchain technologies to address the increasing incidence of chronic illnesses, such as many non-communicable diseases affecting the continent of Africa. Specifically, she is leading a team that is developing solutions for the management of chronic illnesses by providing clinical decision support as well as patient engagement services.

She joined the IBM Research Africa lab and leads the research efforts in mobility and transportation for developing cities. The aim was to provide significant and impactful value-added services that ease the movement of people, goods, and services in Africa. She and her colleagues developed innovative intelligent transportation systems data capture methods and analytical tools, to provide computational understanding about the local driving and infrastructure context.

Aisha has worked in Spain in the area of Smarter Cities at Barcelona Digital and Telefonica. She earned her Ph.D. in the Electrical Engineering and Computer Science Department at MIT in robotics, as a member of the Computer Science and Artificial Intelligent lab (CSAIL).

The pdf being very large, If you are interested in her keynote(very amazing about methods of curbing traffic put in place by using smartphones.)email me at : rose.delilahgesicho@gmail.com

Mondays timetable

DEEP FEEDFORWARD NETWORKS.

Deep feedforward networks, also often called feedforward neural networks, or multilayer perceptrons (MLPs), are the quintessential deep learning models. The goal of a feedforward network is to approximate some function f ∗ . For example, for a classifier, y = f ∗(x) maps an input x to a category y. A feedforward network defines a mapping y= f (x; θ) and learns the value of the parameters θ that result in the best function approximation. These models are called feedforward because information flows through the function being evaluated from x, through the intermediate computations used to define f, and finally to the output y. There are no feedback connections in which outputs of the model are fed back into itself. When feedforward neural networks are extended to include feedback connections, they are called recurrent neural networks.

IMPORTANCE OF DEEP FEEDFORWARD NETWORKS.

Feedforward networks are of extreme importance to machine learning practitioners. They form the basis of many important commercial applications. For example, the convolutional networks used for object recognition from photos are a specialized kind of feedforward network. Feedforward networks are a conceptual stepping stone on the path to recurrent networks, which power many natural language applications.

READY TO DIVE DEEP INTO THE PRACTICAL.

KEYNOTE 2: Leveraging AI to Accelerate Discovery in Life Science and Deliver Precision.

Prof. Abdoulaye Banire Diallo is a Professor (full professor at the age of 26) and Director of the Bioinformatics Lab at Université du Québec à Montréal (UQAM). He is also a co-founder and Chief Scientist of at My Intelligent Machines (MIMs). MIMs is among the top 10 startups to watch in Montreal according to Betakit. In Africa, he co-leads Acces Omic Senegal with the Institute of Recherche en Santé, Surveillance Épidémiologique et de Formation (IRESSEF) of Senegal. This initiative is building the foundation of precision in health in Senegal.

Currently, Abdoulaye’s research is at the intersection between bioinformatics, artificial intelligence, and genomics. He is developing technologies that empower life scientists in the health and agrigenomics sectors to address important questions by providing algorithms and machine learning methods to detect, study and monitor pathogens in epidemiological surveillance, to include genomic precision in livestock production models, and to model and monitor biodiversity in agriculture and forestry. He also provides Artificial Intelligent -powered technologies to help deliver to life science communities an easy-to-use and integrated genomic informed solutions.

The pdf is very large, If you are interested in his keynote to email me at rose.delilahgesicho@gmail.com

DATA SCIENCE

Competitive Data Science.

Platform other than Kaggle is Zindi.

This is a presentation from AlaEddineAyadi, Honestly, I loved this presentation, it was well done the humor blended well in the presentation and the speaker was very resourceful.

That's all I have for you guys from the first two days adding up to part one. Hope you will find this resourceful. I will be back with Part Two despite my laziness😊😁.

(too much youtube content got me like Clap and Follow)

Geek love,Audacious💕

--

--

Audacious Coder.

Hyper-active Data Scientist | STEMINIST | Neuroscience Enthusiast | Dancer |Writer of DS/ML articles😄