MP4 | Video: h264, 1920x1080 | Audio: AAC, 44.1 KHz
Language: English | Size: 2.16 GB | Duration: 2h 33m
Master machine learning by doing it in practice, using an automated machine learning GUI that requires little/no coding.
What you'll learn
See an end-to-end, supervised machine learning process to tackle a regression problem, using Microsoft's Model Builder and ML .Net.
Understand the tasks and activities that take place behind the scenes. From data preparation all the way to model training and evaluation.
Understand data transformation, feature scaling, iterating through algorithms, evaluation metrics, overfitting, cross-validation and regularization.
Understanding the impact of evaluation metrics on model performance, and how to check for overfitting.
Understand the lasting fundamentals of machine learning that are independent of the tools or platforms one can use.
Gain a deep understanding of machine learning concepts by seeing them in action, during a practical machine learning demonstration.
Understand the importance of Exploratory Data Analysis (EDA) and the impact that the statistical distribution of the data has on model performance.
Learn how to set up Visual Studio and to configure it to enable Model Builder, the graphical tool that will be used to demonstrate the machine learning process.
Learn how to use Model Builder to train models without having to code.
Requirements
A basic understanding of supervised machine learning is required. The student would at the very least need to understand what regression is, what features are, and what it means for a model to be trained to fit a function to input features in order to predict labels.
The student needs to have a Windows machine with a few GB of free disk space to install Visual Studio, in order to replicate the machine learning process I will demonstrate. However, this is not essential.
A Windows machine is ideal, but a student with a Mac will still be able to follow along. The course content is visual enough to demonstrate the concepts, without the student having to physically do the machine learning exercise.
Description
In this course, you will get to understand the foundational concepts that underlie the supervised machine-learning process. You will get to understand complex topics such as:Exploratory Data Analysis, Data Transformation and Feature Scaling, Evaluation Metrics, Algorithms, trainers, and models,Underfitting and Overfitting, Cross-validation, Regularization, and much moreYou will see these concepts come alive by doing a practical machine-learning exercise, rather than by looking at presentations. We will be using a non-cloud-based machine-learning tool called Model Builder, inside of Visual Studio. There will be zero coding involved (except for the very last lesson). But even though there is little coding involved, you will still get a very detailed understanding of complex machine-learning concepts.This course requires you to have at least some theoretical exposure to the concepts of supervised and unsupervised machine learning. This course is designed to build on a basic, theoretical understanding of machine learning by doing a practical machine-learning exercise. The concepts taught in this course are foundational and will be relevant in the future, regardless of what machine learning platform or programming language you use. In the process, you will also get some exposure to Visual Studio, code projects, solutions, and the Microsoft Machine Learning ecosystem. But that is just a side benefit. This course focuses on machine learning itself, not the tools that are used.If you've already done any kind of machine learning or trained a model, this course might be too basic for you. This course may contain foundational knowledge that you may not have been taught before, but please be aware that this course is geared toward beginner and intermediate-level AI enthusiasts.
Overview
Section 1: Introduction
Lecture 1 Introduction, Prerequisites and Learning Outcomes
Lecture 2 Introducing Model Builder and the Approach for this Course
Section 2: Visual Studio and Model Builder
Lecture 3 Download, Install and Configure Visual Studio
Lecture 4 Launch Visual Studio and Start a Coding Project
Section 3: Model Builder and the Machine Learning Process
Lecture 5 Introducing Model Builder and the Machine Learning Process
Lecture 6 Model Builder Tasks
Lecture 7 Preparing Data for Machine Learning
Lecture 8 Machine Learning - Training a Model
Lecture 9 Evaluating the performance of a trained model
Section 4: Machine Learning Demo with Model Builder
Lecture 10 Machine Learning in Action Part 1: Getting training data
Lecture 11 Machine Learning in Action Part 2: Preparing the training data
Lecture 12 Demo Part 3
Lecture 13 Demo Part 4
Lecture 14 Understand and Interpret Model Performance
Lecture 15 Consuming a Model and Checking for Overfitting
Lecture 16 Course Summary
This course is for entry-level machine learning enthusiasts, who have had some kind of theoretical introduction to machine learning, but who wants to put the theory into practice.,Machine learning enthusiasts who do not have a background in Statistics, Data Science or programming, but who want to see the complexities of machine learning in practice.,Machine learning enthusiasts who want to learn about complex concepts by seeing them in action, rather than by seeing a presentation.,Technical beginners who want to learn solid machine learning fundamentals before progressing onto more advanced courses where a detailed knowledge of statistics, calculus and programming may be required.
