Azure Machine Learning (Data Preparation & Feature Engineering)

Preparing Data for Feature Engineering and Machine Learning in Microsoft Azure

AML Course on Pluralsight

I am very pleased to announce that my next course, ‘Preparing Data for Feature Engineering and Machine Learning in Microsoft Azure‘ is being released on Pluralsight.

This course is designed in partnership with Microsoft and will be part of 2 different data paths:

The paths are now LIVE.

Full Disclosure: Some of the links in this post are affiliate links. I receive a commission when you purchase through them.

The course covers the following modules:

Getting Started with Azure Machine Learning

Get an introduction to Azure Machine Learning and different services Azure provides to run machine learning tasks

Differentiating Data, Features, Targets, and Models

Going back to absolute basics, and understand the basic terms like data, features, targets, and models

Preparing Input Data for Machine Learning Models

Run Exploratory Data Analysis (EDA). In this module, we will also do various data preparation tasks like Sampling, binning, dealing with outliers, record sampling, attribute sampling, discretizing data, etc.,

Handling Missing Data

We will discuss different strategies to handle missing data and their advantages and disadvantages.

Role of Feature Engineering in Machine Learning

This module explores the role of feature engineering in machine learning. Also, I have discussed the feature engineering techniques for numeric and non-numeric data.

Split a Data Set into Training and Testing Subsets

Strategies to split the dataset into training and test subsets, in a way to avoid the issue of data leakage

Identify Data-level Issues in Machine Learning Models

Some machine learning models have specific data requirements. This module explores all the common data level requirements of different machine learning models.

So, What’s Next?

Here are some clips from the course uploaded on YouTube.

Cross-validation in Machine Learning

Cross validation in Azure Machine Learning

How One-hot Encoding Works (and Demo)

How One-hot Encoding Works

And, view the trailer of the course here. If you are convinced, take the entire course with a free Pluralsight subscription.

With the course, you also get all the code material I used in the course in the Discussion files section. You have a discussion forum, where you can post your doubts. I will make sure to reply to you as soon as possible.

In addition, you also get MCQs to test your understanding

Screenshots from the ‘Data Preparation and Feature Engineering in Microsoft Azure Machine Learning’ Course

Entropy-based Discretization
How Entropy-based Discretization Works
High-dimensional dataset problem
Dealing with a high-dimensional dataset
How Feature Engineering affects Model Complexity Azure Machine Learning
Understanding the role of model complexity in feature engineering

To view the list of all my courses on Pluralsight, check out this page. I would love to know your feedback. Please comment on how you like the course. Check out the resources I used to produce this course.

Follow Me to Receive Updates on AML

Want to be notified as soon as I post? Subscribe to the RSS feed / leave your email address in the subscribe section. Share the article to your social networks with the below links so it can benefit others.

Share the AML Course in Your Network

You may also like