KDnuggets Home » News » 2017 » Dec » Opinions, Interviews » DeepSchool.io: Deep Learning Learning ( 17:n49 )

DeepSchool.io: Deep Learning Learning


What I truly envision for deep school is that this will build a whole lot of Meetup nodes across the world where people will learn, mentor and network around sharing AI knowledge.



By Sachin Abeywardana, Founder of DeepSchool.io

DeepSchool.io is an open-source, community based project to teach the A-Z of Deep Learning (DL). All lessons are interactive and (hopefully) funny, occasionally taking jabs at Mr. Trump (check out this notebook on Trump Tweets). This project came out of a weekly class that I did at Arbor Networks where I work as a Data Scientist.

Personally I come from a background where I did a PhD in Machine Learning. However, with the development of tools such as Keras, DL has become a lot more accessible to the general community. It is no longer just an academic curiosity.

Even with these available tools teaching Deep Learning can be quite difficult. The first lesson I did was a complete train wreck. I had forgotten where I started and jumped straight into a multi layered Deep Net. I took for granted that people would understand what a loss function is, and what regression vs logisitic regression is.

Conversely I did not want to spend too much time on the mathematics either. I wanted to create something that would get people tackling DL problems fast instead of diving too deep into the theory. I spent 6 months or so on Andrew Ng’s DL course that did go through the theory. This unfortunately did not equip me with the tools necessary towards actually being comfortable with using DL in any meaningful way. The goal is to focus on the bigger picture of what you can do with DL.

Goals

  1. Make Deep Learning easier (minimal code).
  2. Minimise required mathematics.
  3. Make it practical (runs on laptops).
  4. Open Source Deep Learning Learning.
  5. Grow a collaborating practical community around DL.
  6. Memes: No seriously. Make DL fun and interactive, this means more Trump tweets.

The assumed knowledge is that you are able to code in Python. I make all code available in Jupyter Notebooks for the sole reason being that you can interact with it. Running on a single python script decreases this interactivity.

How is it different to fast.ai?

A few differences which include shorter videos and that the Jupyter notebooks are mostly self sufficient. But I’m hoping that the main difference is community. What I truly envision for deep school is that this will build a whole lot of Meetup nodes across the world where people will learn, mentor and network around sharing AI knowledge.

 
Bio: Sachin Abeywardana is a PhD in Machine Learning and Founder of DeepSchool.io.

Original. Reposted with permission.

Related:


Sign Up