Skip to content
Your new Mentor for Data Science E-Learning.
Branch: master
Clone or download
Latest commit 8089001 Jul 15, 2019

README.md

Figure 1-1

Future Map

Figure 2-2

Towards the new Virgilio map

In the image above you can see the roadmap of the new version of Virgilio. For now just use the old organization of the project at the end of this page.

Contributing to Virgilio

Do you think that you can help in building the above future map? Check here the contribution guidelines and send and e-mail to giacomo.ciarlini@student.unife.it or text on Telegram to @Clone95.

What is Virgilio?

Studying and reading through the Internet means swimming in an infinite jungle of chaotic information, even more so in rapidly changing innovative fields.

Have you ever felt overwhelmed when trying to approach Data Science without a real “path” to follow? Are you tired of clicking “Run”, “Run”, “Run”.. on a Jupyter Notebook, with that false sense of confidence given by the comfort zone of the work of others?

Have you ever got confused because of the several and contradicting names for the same algorithm or approach, from different websites and fragmented tutorials?

I want to address these critical problems for free, for everyone. It’s not a joke.

Hi, I'm Virgilio.

Like I did with Dante, just some centuries ago, I'll be your mentor and reference point during your journey through the Internet, providing you complete and organic learning paths for several fields, tools, skills and more.

The vision of Virgilio is to give everyone the possibility to get into the incredible world of the Data Science and the business and creative possibilities that it offers, to get self-started as a practitioner, gain new cutting edge practical skills or just learn to discriminate good information from poor information.

We are doing this by providing only high-quality and coherent content, with clear step-by-step paths and a consistent naming system. Imagine Virgilio as an E-Mentor who will tell you what do to get the next step, the next skill, or to apply them in practice to create value.

But what does it mean in practice?

In other words, what's the target of the Virgilio project?

There are different scenarios:

  • A student from a different field who wants to explore the intersections and the possibilities offered by the Machine Learning and Statistical methods.
  • A curious individual who came in touch with one of the buzzwords related to these fields and wants to discriminate between reliable and unreliable information.
  • An experienced practitioner who wants to have a reference point for the latest techniques, papers and best practices.
  • A manager who wants to understand the possibilities of ML applied to their actual problems, like integration with production systems or new solutions from scratch.
  • A businessman who wants to understand if his data are suitable for an ML project, and what could be the real business value.
  • A salesman who needs to stay up-to-date with the latest technologies and Jargon.

Structure of Virgilio

As you can imagine it’s not easy to intercept all these different needs, so our solution ended up in a hierarchical structure which distinguish the content based on different levels of abstraction.

To do this, we got our inspiration from Dante’s amazing masterpiece “La Divina Commedia”, written in over 20 years of work and published for the first time in 1323 d.c.

Dante’s journey wouldn’t have been possible without his companion and guide, Virgilio, a roman famous poet (70 a.c.) who inspired generations of artists since the Roman hegemony in Europe.

In his journey, Dante travel across the different levels of the catholic conception of the divine world at that time, starting from the Inferno (the prison of the damned), passing through the Purgatorio and reaching eventually the Paradiso (you can call it Valhalla or Nirvana, as you prefer :-) ).

In your journey, you will start from scratch and eventually reach the theoretical knowledge and solid expendable skill.

The parallelism is natural:

Figure 2-2

Going from top to bottom increases the level of detail and decreases the level of abstraction.

In the Paradiso you won’t find a single line of code or a math formula, just plain English. Here’s the place for introductions, simple explanations, demystifications, and meta-guides (for example a guide about the best way to use Virgilio). It’s the best place for the not-techies, beginners and literally everyone who wants to get in touch with Data Science and Machine Learning without getting bored into technical details. Do you want to communicate these innovative fields? Pass from the Paradiso!

In the Purgatorio you can find technical guides for beginners (in the field or in general IT). For example, you’ll find guides about Python, maths, and statistics. You will find guides about study techniques, soft skills and you’ll learn how to develop an analytical mindset too. It’s an obliged step before the Hell. Depending on your starting skill, you’ll probably spend here most of the time, learning to code, understand math concepts and more! If you’re a complete beginner, follow the track we proposed, starting from the Fundamentals.

In the Inferno you won’t find gentle introductions or generic explanations, but a lot of different detailed guides, topics, hands-on tutorials and more! You’ll find an entire section dedicated to research and daily updates from the field! You’ll find guides like “how to train a massive neural network over hundreds of GPUs efficiently” or “How to deal with huge datasets”, or “how to fine-tune a preprocessing pipeline”. Think about Inferno as the place where you will pick up the sub-field you prefer and dive into that. It’s impossible to learn everything at once! One of the Virgilio’s most important learning strategies is “One enemy at time”, "Divide and conquer!". The three specializations that we provide are Computer Vision, Natural Language Processing, and Agent-based and Reinforcement Learning. These learning paths are the “final bosses” of the Virgilio’s experience: once you’ll have completed them you will hopefully be skilled enough to land an internship or tackle real business problems!

Above these you’ll find a plenty of other useful zones:

  • Business Skills: these are necessary to unlock the business value in the real world, and they are probably the most valuable asset for a “data guy” to have in his pocket.
  • Tools: here you find guides about useful tools for programming, or scientific computing in general.
  • Research: here you will learn how to find the right papers and digest them. In addition, you’ll discover which teams to follow for your interests.
  • Massive computation: here you’ll find hints and resources for computing on clusters, optimize your system and other advanced topics.
  • ML for Business applications: here we provide additional resources to select a sector and see what are the today’s available techniques for the problems of your interest.

Paradiso

  • Coming soon!

Purgatorio

  • Coming soon!

Inferno

  • Coming soon!

Old version of Virgilio

Virgilio was born as a side project just this April, and at the very beginning it was not meant to become what it is now! Because of that we're holding the old organization of Virgilio too (for a while).

New To Data Science?

Basic Python

Matrix Algebra

Advanced Python -- Coming Soon

Advanced Math -- Coming Soon

Python for Data Science

Math for Data Science -- Coming Soon

Complete Learning Paths

Machine Learning Study Path

Skills

Data Preprocessing

Data Collection -- Coming Soon

Data Visualization

Effective Communication -- Coming Soon

Impactful Presentations

Pragmatic Decision Making -- Coming Soon

Tools

Latex

Wolfram Alpha

GeoGebra

RegeX

MLDemos

A Lot More Coming Soon!

Topics

High Level Topics

1 - Demystification of the key concepts of Artificial Intelligence and Machine Learning

2 - Introduction to ML systems

3 - Introduction to Artificial Neural Networks

What do I need for ML?

Do you really need ML?

ML use cases

Learning how to learn

Low Level Topics

Chatbots - Build a complex and useful Virtual Assistant with DialogFlow and Flask

Introduction to Computer Vision using OpenCV and Python

Deep Learning in Cloud

Object Instance Segmentation using TensorFlow Framework and Cloud GPU Technology

Research

Zotero

State-of-Art Papers Explained

SUBSCRIBE TO THIS FACEBOOK GROUP WHERE WE POST NEWS AND UPDATES

If you like Virgilio, FILL THIS FORM and I'll contact you whenever a new free guide comes out.

You can’t perform that action at this time.