88
Lectures
13
Hours

Dive into the world of machine learning with Machine Learning with JS: Regression Tasks (Math + Code). This course offers a focused look at linear regression, blending theoretical knowledge with hands-on coding to teach you how to build and apply linear regression models using JavaScript.


What You Will Learn:

  • Core Principles of Linear Regression: Begin with the fundamentals of linear regression and expand into multiple regression techniques. Discover how these models can predict future outcomes based on past data.

  • Hands-On Coding: Engage directly with practical coding examples, utilizing JavaScript. You'll use Node.js for the computational aspects and React.js for dynamic data visualization.

  • Simplified Mathematics: We make the essential math behind the models accessible, focusing on concepts that allow you to understand and implement the algorithms effectively.

  • Project-Based Learning: Build a React application from scratch that not only plots data but also computes regression parameters and visualizes these computations in real-time. This hands-on approach will help solidify your learning through actual development experience.

  • Real-World Applications: Learn to forecast real-world outcomes using the models you build. Understand the importance of residuals and how to quantify model accuracy with statistical measures such as R-squared, Mean Absolute Error (MAE), and Mean Squared Error (MSE).

  • Advanced Topics in Depth: Go beyond basic regression with sessions on handling complex data types through multiple regression analysis, matrix operations, and model selection techniques.


Course Structure:

This course includes over 80 detailed video lectures that guide you through every step of learning machine learning with JavaScript:


  • Introduction and Setup: Start with an overview of the necessary tools and configurations. Understand the foundational terms and concepts in regression.

  • Interactive Exercises: Each new concept is paired with practical coding exercises that reinforce the material by putting theory into practice.

  • In-Depth Projects: Apply what you've learned in extensive, real-world projects. Predict salary ranges based on job data or estimate car prices with sophisticated regression models.


Why Choose This Course?

  • Targeted Learning: We focus on linear regression to provide a thorough understanding of one of the most common machine learning techniques.

  • Practical JavaScript Use: By using JavaScript, a language familiar to many developers, this course demystifies the process of integrating machine learning into web applications and backend services.

  • Project-Driven Approach: The projects are designed to reflect real industry problems, preparing you for technical challenges in your career.


Requirements

  • Basic knowledge of any programming language

Course curriculum

  • 2

    Linear Regression 101

  • 3

    Linear Regression Basics

  • 4

    Score Prediction

  • 5

    Model Evaluation

    • Residuals

    • Compute residuals in the code

    • R squared computation

    • Compute r2 in code

    • Mae computation

    • Compute MAE in code

    • MSE computation

    • Compute MSE in code

  • 6

    Prepare React JS Components

    • Create separate component for prediction

    • Model selection

    • Finish model selection

    • Formula with residual

  • 7

    Multiple Regression Basics

    • Multiple regression start

    • Multiple regression in App

    • Matrices explanation

    • Organize matrices in code

    • Matrix multiplication

    • Matrix multiplication in code

    • Another Multiplication

  • 8

    Multiple Regression Advanced

    • Calculate Determinant

    • Adjugate

    • Compute B coefficients

    • Compute coefficients in code

    • Store coefficients

    • Get coefficients on frontend

    • Display regression plane

  • 9

    Salaries Prediction Task

    • Data preparation

    • Parse data from CSV

    • Split data

    • Data seeding

    • Compute regression data

    • Explain stats

    • Store coefficients

    • Prepare data for r2

    • Compute r2

    • Store all data in JSON

    • Display data on the graph

    • Display regression plane on salaries

    • Predict salaries

  • 10

    Car Prices Prediction Task

    • Prepare car prediction

    • Format data to dictionary

    • Simplify car name

    • Fix typos in car names

    • Create category map

    • Process data to array

    • Debugging

    • One hot encode

    • Text to number parsing

    • Row categories

    • Data splitting

  • 11

    Model Training and Evaluation

    • Train the model

    • Compute r2 for car prices

    • Compute correlation array

    • Get correlated categories

    • Compute model with correlations

    • Include car names in model

    • Car prediction init in React

    • Export data

  • 12

    Data Visualization

    • Display all graphs

    • Improve model performance

    • Create inputs

    • Create selection for car names

    • Set car name value

    • Default values for inputs

    • Course end - Compute prediction

Instructor(s)

Software Engineer

Filip Jerga

Hey, I'm Filip Jerga! I've been deep in the software engineering world for over 10 years, building all sorts of cool stuff—from software for ultrasound devices with C++ to modern apps with React and Angular. I even grabbed a Master's in AI along the way. What really drives me now isn't just coding; it's making complex programming simple and fun for you. I've poured all my real-world experience and a passion for teaching into these courses. My goal is to help you think like a pro, tackle tough problems, and build practical skills you'll truly use to create amazing things.

FAQ

  • How to get help when I am stuck with the course?

    Every video contains a discussion where you can create a post describing an issue. The instructor usually responds within 1 business day.

  • What to do when I am unhappy with the course ?

    Within 30 days of the purchase, you can ask for a full refund. No questions asked. Your happiness is our priority.

  • Do I need to watch every lecture of really extensive course?

    Of course not! Every lecture is committed (explained in the introduction section). You can start watching at any lecture. Just download the correct version of the project attached to lecture resources.

Eincode Access Options

  • Single Course Access

    • Lifetime and unlimited Access to "Machine Learning Primer with JS: Regression (Math + Code)"

    €19,00

    Access to this course only.

    Get started now
  • Monthly Membership

    • Unlimited Access!
    • Access to any current course!
    • Access to future courses!

    €19,00 / month

    Cancel Anytime!

    Get started now
  • Annual Membership

    • Unlimited Access!
    • Access to any current course!
    • Access to future courses!

    €190,00 / year

    Cancel Anytime!

    Get started now