Day2 - Machine Learning With Microsoft Azure

Hi there, i'm here again on my progress in the Introduction to machine learning on Azure with a Low-code Experience. Today was a little hectic as i'm currently enrolled in a deep learning nanodegree on Udacity and one of my projects is giving me a hard time nonetheless i still spared some time to study a handful of section on lesson 2(Introduction to machine learning). As yesterday was the study of lesson 1 or say introduction and high-level look into the course, today which is the day 2(Introduction to machine learning) i continued with lesson 2 and attempted three sections.

Here is a broader overview of what would be covered in lesson 2

• What machine learning is and why it's so important in today's world
• The historical context of machine learning
• The data science process
• The types of data that machine learning deals with
• The two main perspectives in ML: the statistical perspective and the computer science perspective
• The essential tools needed for designing and training machine learning models
• The basics of Azure ML
• The distinction between models and algorithms
• The basics of a linear regression model
• The distinction between parametric vs. non-parametric functions
• The distinction between classical machine learning vs. deep learning
• The main approaches to machine learning
• The trade-offs that come up when making decisions about how to design and training machine learning models In the process, students will also train their first machine learning model using Azure Machine Learning Studio.

What is Machine Learning ?

Machine learning is a data science technique used to extract patterns from data, allowing computers to identify related data, and forecast future outcomes, behaviors, and trends.

Machine Learning is a branch of Artificial Intelligence(AI) that deals with the extraction of patterns from data to perform a specific task rather than explicitly programming them. For instance, if we were to develop an android app to track the amount of infected COVID-19 patients all over the world with their cases and those recovered, as a developer, you might want to consume an API to get real-time statistics of each cases in a particular country moreover, you would also want to choose a language for programming such as JAVA or Kotlin and also some UI/UX design for the interface of the application. All of these rules are being explicitly stated one way or the other. In Traditional programming, the rules(i.e your algorithm coupled with the programming language of your choice) and data gives us the results or answers (COVID tracker application we aim to develop in this case), but in machine learning that's not the case. Instead we provide the data and answers we want and feed those to a machine learning model to provide certain rules on its own. The ML model tries to extract patterns and recognize those patterns from the data and maps them to answers provided to the model.

Applications Of ML

In today's world of BIG data, ML application is endless as more and more data are being generated daily also virtaully every industry is set to be revolutionalized with ML from healthcare to manufacturing, finance, entertainment, education and agriculture, the applications are extremely broad.

Applications

• Skin cancer
• Autonomous vehicles
• Games development and several more

Thats all for today guys, looking forward to tomorrow's lesson.