English | MP4 | AVC 1280×720 | AAC 44KHz 2ch | 2.5 Hours | 945 MB
A hands-on approach to Machine Learning using the easy drag-n-drop environment of Azure Machine Learning Studio
Machine Learning is the most in demand technical skill in today’s business environment. Most of the time though it is reserved for professionals that know how to code.
But Microsoft Azure Machine Learning Studio changed that. It brings a drag-n-drop easy to use environment to anyone’s fingertips. Microsoft is known for its easy-of-use tools and Azure ML Studio is no different.
However, as easy as Azure ML Studio is, if you don’t know Machine Learning, at least the basics, you won’t be able to do much with the tool. This is one of the goals of this course: To give you the foundational understanding about Machine Learning. You will get the base knowledge required to not only talk proficiently about ML, but also to put it into action and execute on business needs.
We will go through all the steps necessary to put together a Supervised Learning prediction model, whether you need Classification (for discrete values like “Approved” or “Nor Approved”) or Regression (for continuous values like “Salary” or “Price”).
The course will only require you to have basic knowledge of math including the basic operations and how to calculate average. Some exposure to Microsoft Excel would be good as during deployment of the live model, we will be using Excel to perform demonstrations.
This course has been designed keeping in mind technologists with no coding background as we use a “no-code approach”. It is very hands-on, and you will be able to develop your own models while learning. We will cover:
- Basics of the main three main types of Machine Learning Algorithms
- Supervised Learning in depth
- Classification by using the Titanic Dataset
- Understanding and selecting the features from the dataset
- Changing the metadata of features to work better with ML Algorithms
- Splitting the data
- Selecting the Algorithm
- Training, scoring, and evaluating the model
- Regression by using the Melbourne Real Estate Dataset
- Cleaning missing data
- Stratifying the data
- Tuning hyperparameters
- Deploying the models to a Excel
- Providing web service details to developers in case you want to integrate with external systems
- Azure ML Cheat Sheet
The course also includes 4 assignments with solutions that will give you an extra chance to practice your newly acquired Machine Learning skills.
In the end you will be able to use your own datasets to help your company with data prediction or, if you just want to impress the boss, you will be able to show the new tool you have just added to your toolbelt.
If you are not a coder and thought there would be no place for you to ride the Machine Learning wave, think again. You can not only be part of it, but you can master it and become a Machine Learning hero with Azure ML Studio.
What you’ll learn
- Examine the foundations of Supervised Machine Learning
- Use Azure ML Studio to create Predictive Models without code
- Evaluate different algorithms to find the one that works best
- Deploy models live to be used with new data
- Build a real estate model to predict house prices
- Experiment with the traditional Titanic Dataset to predict survival chances
Table of Contents
Welcome to the Course
1 Welcome
2 Compare Machine Learning Categories
3 Create a Free Azure Account
4 Define Azure ML Studio Features
Classification Using the Titanic Dataset
5 Introduction
6 Evaluate Model
7 Summary
8 Load the Dataset
9 Understand the Features
10 Select Features
11 Edit Metadata
12 Split the Data
13 Select the Algorithm
14 Train Model
15 Score Model
Refining The Classification Model
16 Introduction
17 Summary
18 Summarize The Data
19 Select More Features
20 Clean Missing Data
21 Stratify The Data
22 Tune The Hyperparameters
23 Evaluate Model in Depth
24 Compare Different Algorithms
25 Deploy The Model
Regression Using A Real Estate Dataset
26 Introduction
27 Explore The Data
28 Clean Missing Data
29 Edit Metadata
30 Test Model
31 Evaluate Model
32 Optional – MAE and RAE Explained
33 Summary
Refining The Regression Model
34 Introduction
35 Reassess Feature Selection
36 Hyperparameter Tuning
37 Compare Algorithms
38 Deploy The Model
39 Summary
Conclusion
40 What You Have Learned
41 Next Steps
42 Bonus Lecture
Resolve the captcha to access the links!