Designing a Machine Learning Model

By Janani Ravi
This course covers the important differences between various canonical problems in machine learning, as well as the considerations in choosing the right solution techniques, based on the specifics of the problem you are trying to solve and the data that you have available.
Play course overview
Course info
Level
Intermediate
Updated
Aug 13, 2019
Duration
3h 25m
Table of contents
Course Overview
Course Overview 2m
Exploring Approaches to Machine Learning
Module Overview 1m Prerequisites and Course Outline 2m Sentiment Analysis as a Binary Classification Problem 2m Representation Machine Learning Systems 2m Traditional ML vs. Deep Learning 3m Traditional Machine Learning Systems 4m Deep Learning and Neural Networks 5m A Case Study: Sentiment Analysis 7m Rule Based vs. ML Based Analysis 7m Module Summary 1m Traditional ML Algorithms and Neural Network Design 5m
Choosing the Right Machine Learning Problem
Module Overview 1m Module Summary 2m Recommendation Systems 4m Supervised and Unsupervised Learning 8m Choosing the Right Machine Learning Problem 7m Reinforcement Learning 7m
Choosing the Right Machine Learning Solution
Module Overview 2m Regression Models 2m Evaluating Classifiers 4m The Curse of Dimensionality 5m Clustering Models 5m Evaluating Regression Models 5m Choosing Classification Algorithms 3m Module Summary 1m Dimensionality Reduction Techniques 3m Choosing Regression Algorithms 4m Types of Classification 4m
Designing Machine Learning Workflows
Module Overview 1m Averaging and Boosting, Voting and Stacking 2m The Machine Learning Workflow 4m Module Summary 1m Custom Neural Networks: Their Characteristics and Applications 4m Case Study: PyTorch on the Cloud 6m Ensemble Learning 6m
Building Simple Machine Learning Solutions
Module Overview 1m Exploring the Regression Dataset 3m Install and Set Up 2m Multiple Regression Using Analytical and Machine Learning Techniques 2m Classification Using Logistic Regression 4m Simple Regression Using Analytical and Machine Learning Techniques 5m Classification Using Decision Trees 3m Module Summary 1m Dimensionality Reduction Using Principal Component Analysis 4m Exploring the Classification Dataset 3m Clustering Using K-means 7m Dimensionality Reduction Using Manifold Learning 5m
Building Ensemble Solutions and Neural Network Solutions
Module Overview 1m Summary and Further Study 2m Exploring and Preprocessing the Regression Dataset 3m Classification Using Hard Voting and Soft Voting 5m Regression Using Gradient Boosting 4m Regression Using Neural Networks 8m Regression Using Bagging and Pasting 5m
Description
Course info
Level
Intermediate
Updated
Aug 13, 2019
Duration
3h 25m
Description

As Machine Learning explodes in popularity, it is becoming ever more important to know precisely how to frame a machine learning model in a manner appropriate to the problem we are trying to solve, and the data that we have available. In this course, Designing a Machine Learning Model you will gain the ability to appropriately frame your use-case and then choose the right solution technique to model it. First, you will learn how rule-based systems and ML systems differ and how traditional and deep learning models work. Next, you will discover how supervised, unsupervised, and reinforcement learning techniques differ from each other. You will learn how classic supervised learning techniques such as regression and classification complement classic unsupervised techniques such as clustering and dimensionality reduction. You will then understand the assumptions and outcomes of these four classes of techniques and how solutions can be evaluated. Finally, you will round out your knowledge by designing end-to-end ML workflows for canonical ML problems, ensemble learning, and neural networks. When you’re finished with this course, you will have the skills and knowledge to identify the correct machine learning problem setup, and the appropriate solution technique for your use-case.

About the author
Janani Ravi
About the author

A problem solver at heart, Janani has a Masters degree from Stanford and worked for 7+ years at Google. She was one of the original engineers on Google Docs and holds 4 patents for its real-time collaborative editing framework.

More from the author
Building Features from Image Data
Advanced
2h 10m
Aug 13, 2019
Building Features from Nominal Data
Intermediate
2h 40m
Aug 12, 2019
More courses by Janani Ravi
Section Introduction Transcripts
Section Introduction Transcripts

Course Overview
[Autogenerated] Hi. My name is generally Ravi and welcome to the school's on designing a machine learning Marty a little about myself. I have a master's degree in electrical engineering from Stanford and have offered companies such as Microsoft, Google and Flip Cards at Google was one of the first engineers working on real time collaborative editing in Google Dogs and I hold four patterns for its underlying technology's. I currently work on my own Start up loony Con, a studio for high quality video content. As machine learning explodes in popularity, it is becoming ever more important to know precisely how to frame a machine learning model in a manner appropriate to the problem we're trying to solve on the data that we have available in this course, you will gain the ability to appropriately frame your use case and then choose the right solution technique to model it. First, you will learn how rule based systems and ML systems differ on how traditional and deep learning more _____ work. Next, you will discover how supervise, unsupervised and reinforcement learning techniques differ from each other. You will learn how classic supervised learning techniques such as digression and classifications compliment classic unsupervised techniques such as clustering and dimensionality reduction. You will then understand the assumptions and outcomes off these four classes of techniques on how solutions can be evaluated. Finally, you will round out your knowledge by designing end to an MLB clothes for canonical ML problems. Ensemble learning as the less neutral networks. When you're finished with the schools, you will have the skills and knowledge to identify the correct machine learning problem set up under appropriate solution technique for your use case.

You're the smartest person in the room

PROVE IT

Your team won't just close the skills gap

THEY'LL HURDLE IT