Many beginners intrigued by Data Science/ML/AI behold it in the awe and fear reserved for a hairy monster. Most of it originates in the fear of math, not to mention the overwhelming variety of buzzwords flying past in all directions.

It’s not all dark and gloomy though. If you develop a strong intuition for certain fundamental concepts, you won’t just find your way in this forest, but will also feel confident and equipped to create your own paths.

The instructors aim to aid building that intuition through an immersive, daylong workshop. You will learn machine learning concepts and tools from first principles!

Why Attend?:

Today it certainly is imperative for tech professionals of all kind to be, at the very least, initiated on concepts of machine learning and data science. While there is no dearth of material available on the internet to educate yourself, it certainly is too overwhelming for someone who’s just made up the mind to plunge into this seemingly vast, bottomless sea. Needless to say, however, that the journey is extremely satisfying as well as rewar ding, if you are equipped to take it.

Numpy, Scipy, Pandas, scikit-learn etc are emerging as some of the most handy tools for the modern sailors and intuition for concepts like probability theory, functions, differential calculus, gradient descent algorithm are the planks that form the very rafts on which one sails. Let’s build our own and enjoy the first dip !

Maths Knowledge Requirements

Part 1 of the workshop will cover all the maths well. So you should know how to do addition, subtraction, multiplication and division. Plus, you need pen, paper and some grit to keep writing as the existential Xs stare back at you asking “Who am I? Why am I here?”

Programming Knowledge Requirements

You should know basic programming (reading and writing from/to files, flow controls (if-else), looping constructs like for loop, while, variable assignments). In other words, you should have programmed few hundred lines in any mainstream programming language.

What will you learn?

We will cover the following in this two part workshop,

Part 1 : Mathematics of data science

A lot of really interested, good programmers and tech professionals seem to maintain distance from data science because they are just plain scared of the math. The workshop will be a refresher of the basic concepts and does not assume any prior knowledge greater than addition, subtraction, multiplication and division.

1. Indices and logarithms
2. Functions as transformations.
3. Combinatorics
4. Idea of probability as long term frequency of events.
5. Limits, integration and derivatives.
6. Application of Derivative for finding minima and maxima. Gradient Descent / Ascent to solving optimization problems.

Part 2 : Tools and Concepts of Data Science

We will cover a high level overview of field of machine learning and introduction to the Python data ecosystem in machine learning. We strongly believe that the best way to learn machine learning is by building few algorithms from scratch. So we will build a supervised ML application from scratch in Python. Since ML is a very vast field, I will spend some time on study guidelines and how to approach the field.

1. Big picture of machine learning : supervised Vs. unsupervised, generative Vs. discriminative models. Mostly plain English content, covering big picture (~ 20 minutes)
2. Introduction to Python data ecosystem: few hands on exercises on numpy and pandas to serve as warm-up (~ 30 minutes)
3. Building a regression/classifier from scratch in Python (~ 45 - 50 minutes), using all the concepts learnt in the first part of the workshop. We complete a full circle about how the concepts from first part of workshop tie up with applications.
4. Solving a more involved problem by using scikit-learn APIs directly (~ 30 minutes). Introduction to concepts of bias Vs. variance, testing models and feature engineering.
5. Next steps, how to study and which resources can be used (~ 20 minutes)
6. Summarizing what we learnt, question-answers (~ 10 minutes)

Computing​ ​Requirements

• Pen/pencils and lots of blank papers.
• Laptop (operating system of your choice), charged battery + charger.
• Python installed on the laptop + IDE of your choice/terminal.
  • No hard choice between python 2 Vs python 3.
• Following libraries MUST​ be installed.
  • Numpy
  • Scipy
  • Pandas
  • scikit-learn
• It won’t be possible to provide installation support at the time of workshop. So all requirements should be pre-installed. Without the installations, you won’t get anything out of the workshop.