Microsoft's Machine Learning for Beginners
Written by Nikos Vaggalis   
Tuesday, 17 August 2021

A free, self-paced online course about Machine Learning is on offer from Microsoft's Azure Cloud Advocates. Its 24-lesson curriculum, expected to take 12-weeks to complete is targeted at those new to Machine Learning.


There are plenty of advantages in attending Machine Learning for Beginners. First you learn about the, currently trending, concepts of ML, acquiring skills highly sought after by employees. On top of that you learn them in conjunction with Python, the most popular and versatile language, also highly sought after.

This role of the middleman between Python and Machine Learning is played by none other than scikit-learn, which is chosen for good reason. As I explained in "Introduction to Machine Learning with Scikit-Learn".

Python certainly is the most popular language of doing ML, mainly due to the number of relevant libraries available. scikit-learn is one of the top Machine Learning libraries alongside PyTorch, NumPy, SciPy, TensorFlow and Theano. Additionally, scikit-learn is one of the easiest to learn as such perfect for beginning one's ML journey. That doesn't mean that it lacks functionality though; it is perfectly capable of pulling off many ML tasks such as classification, clustering, pre-processing, regression, etc.

It is also important to note that the authors have made a clear distinction between Machine Learning and AI. This course is about "classic machine learning" and does not concern itself with artificial intelligence, something that its sibling course AI for Beginners covers. This separation of topics means that ML for Beginners is not as complicated as it otherwise would be.

With that out of the way let's focus on the course itself.
As said, although self-paced it it spans 12-weeks and 3 months is the expected time required to complete it. Being self-paced rather than instructor-led doesn't lessen its value. On the contrary it is carefully planned and well structured. It includes quizzes, assignments, projects, group discussions and, of course, quality material.

The course is hosted on GitHub with links to You Tube videos intermixed with the actual textual lessons and is comprised of 24 lessons: 

  • Introduction to machine learning
    Learn the basic concepts behind machine learning
  • The History of machine learning
  • Fairness and machine learning
    What are the important philosophical issues around fairness that students should consider when building and applying ML models?
  • North American pumpkin prices I
    Visualize and clean data in preparation for ML
  • North American pumpkin prices II
    Build linear and polynomial regression models
  • North American pumpkin prices III
    Build a logistic regression model
  • A Web App 
    Build a web app to use your trained model
  • Introduction to classification
    Clean, prep, and visualize your data; introduction to classification
  • Delicious Asian and Indian cuisines 
    Introduction to classifiers
  • Delicious Asian and Indian cuisines 
    More classifiers
  • Delicious Asian and Indian cuisines 
    Build a recommender web app using your model
  • Introduction to clustering
    Clean, prep, and visualize your data; Introduction to clustering
  • Exploring Nigerian Musical Tastes 
    Explore the K-Means clustering method
  • Introduction to natural language processing 
    Learn the basics about NLP by building a simple bot
  • Common NLP Tasks 
    Deepen your NLP knowledge by understanding common tasks required when dealing with language structures
  • Translation and sentiment analysis 
    Translation and sentiment analysis with Jane Austen
  • Romantic hotels of Europe 
    Sentiment analysis with hotel reviews 1
  • Romantic hotels of Europe 
    Sentiment analysis with hotel reviews 2
  • Introduction to time series forecasting
  • World Power Usage 
    time series forecasting with ARIM
  • Introduction to reinforcement learning
    Introduction to reinforcement learning with Q-Learning
  • Help Peter avoid the wolf! 
    Reinforcement learning Gym
  • Extra lesson:Real-World ML scenarios and applications
    Interesting and revealing real-world applications of classical ML 

It is evident from that list that the course teaches a lot of practical applications of ML, mostly focused on its data science side, such as: 

  • To predict the likelihood of disease from a patient's medical history or reports.
  • To leverage weather data to predict weather events.
  • To understand the sentiment of a text.
  • To detect fake news to stop the spread of propaganda. 

However, if you're looking for the Neural Networks side of it then it's best to go with "AI for Beginners".

As prerequisites, it's recommended to have a basic understanding of Python while some JavaScript is also required when building the web app project. Tool wise you will need to have node and npm installed, as well as Visual Studio Code for both Python and JavaScript development. And of course a GitHub account. As far as Scikit goes,you're going to use it throughout so it's best to familiarize yourself with it.

After going through it, you'll be looking for the next steps. The instructors advise that you continue with the forthcoming Data Science for Beginners and, as already mentioned, AI for Beginners.

All in all, this course offers a first class opportunity to start your journey in Machine Learning!


More Information

Machine Learning for Beginners

Related Articles

Introduction to Machine Learning with Scikit-Learn

Learn Python with HackInScience

Fly Over the Moon With Microsoft And Python

Learn Python with Microsoft or the University of Michigan


To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.


.NET MAUI VS Code Extension Now Generally Available

Microsoft has announced that .NET MAUI VS Code extension is out of preview, and that it includes long-awaited new features – including XAML IntelliSense and Hot Reload.

IBM Launches The Granite Code LLM Series

IBM is introducing decoder-only code models for code generative tasks as part of its Granite collection. The models have been trained with code written in 116 programming languages and range in size f [ ... ]

More News

C book



or email your comment to:

Last Updated ( Tuesday, 17 August 2021 )