A-Z Machine Learning using Azure Machine Learning (AzureML)
Machine Learning is one of the hottest and top paying skills. It’s also one of the most interesting field to work on.
In this course of Machine Learning using Azure Machine Learning, we will make it even more exciting and fun to learn, create and deploy machine learning models. We will go through every concept in depth. This course not only teaches basic but also the advance techniques of Data processing, Feature Selection and Parameter Tuning which an experienced and seasoned Data Science expert typically deploys. Armed with these techniques, in a very short time, you will be able to match the results that an experienced data scientist can achieve.
This course will help you prepare for the entry to this hot career path of Machine Learning.
This course has more than 80 lectures and is over 11 hours in content. That simply means, we go through the details of Data Science and Machine Learning along with its implementation. Almost every topic has a hands-on lab that you can practice. I have dealt with almost all scenarios during my tenure with various governments across the world and Fortune 500 companies.
I am committed to and invested in your success. I have always provided answers to all the questions and not a single question remains unanswered for more than a few days. The course is also regularly updated with newer features.
Learning data science and then further deploying Machine Learning Models have been difficult in the past. To make it easier, I have explained the concepts using very simple and day-to-day examples. Azure ML is Microsoft’s way of democratizing Machine Learning. We will use this revolutionary tool to implement our models. Once learnt, you will be able to create and deploy machine learning models in less than an hour using Azure Machine Learning Studio.
Azure Machine Learning Studio is a great tool to learn to build advance models without writing a single line of code using simple drag and drop functionality. Azure Machine Learning (AzureML) is considered as a game changer in the domain of Data Science and Machine Learning.
This course has been designed keeping in mind entry level Data Scientists or no background in programming. This course will also help the data scientists to learn the AzureML tool. You can skip some of the initial lectures or run them at 2x speed, if you are already familiar with the concepts or basics of Machine Learning.
The course is very hands on and you will be able to develop your own advance models while learning,
Advance Data Processing methods
Statistical Analysis of the data using Azure Machine Learning Modules
MICE or Multiple Imputation By Chained Equation
SMOTE or Synthetic Minority Oversampling Technique
PCA; Principal Component Analysis
Two class and multiclass classifications
Support Vector Machine (SVM)
Understanding how to evaluate and score models
Detailed Explanation of input parameters to the models
How to choose the best model using Hyperparameter Tuning
Deploy your models as a webservice using Azure Machine Learning Studio
Feature selection using Filter-based as well as Fisher LDA of AzureML Studio
Recommendation system using one of the most powerful recommender of Azure Machine Learning
All the slides and reference material for offline reading
You will learn and master, all of the above even if you do not have any prior knowledge of programming.
This course is a complete Machine Learning course with basics covered. We will not only build the models but also explain various parameters of all those models and where we can apply them.
In this course, we will start with some basic terms which are used very frequently in machine learning.
I will also explain
What is Machine Learning and some real world examples.
Azure Machine Learning Introduction
Provide an overview of Azure Machine Learning Studio and high level architecture.
We would also look at
Steps for building an ML model.
Supervised and Unsupervised learning
Understanding the data and pre-processing
Different model types
The AzureML Cheat Sheet.
How to use Classification and Regression
What is clustering or cluster analysis
KDNuggets one of the leading forums on Data Science calls Azure Machine Learning as the next big thing in Machine Learning. It further goes on to say, “people without data science background can also build data models through drag-and-drop gestures and simple data flow diagrams.”
Azure Machine Learning’s library has many pre-built models that you can re-use as well as deploy them.
This course will also be a great help in preparing for the Microsoft 70-774 exam-Perform Data Science on Cloud using Azure Machine Learning. It covers almost all the topics of Azure Machine Learning.
So see you inside the course.
Basics of Machine Learning
This lecture provides an overview of the section of Basics of Machine Learning and what is covered in this section.
Why machine learning is the future? The Data explosion. We will also see some common examples of ML as well as discuss couple of case studies of Machine Learning.
In this lecture we cover,
- What is Machine Learning; definition and explanation
- How machines learn?
- Examples of Machine Learning
- Supervised, Unsupervised and Reinforcement Learning
In this lecture we will learn about reading and understanding the data
- Types of Variables
- Data Type and
- Category of the variables
We will learn various basic terms such as Mean, Mode, Median, Range and their importance along with what is probability and how to calculate it for some simple example.
In this lecture, we are going to cover four fundamental model types that you would build and related algorithms.
- Cluster Analysis
- Anomaly Detection
Getting Started with Azure ML
Provides the section overview of Getting started with AzureML.
Overview of AzureML and its high level architecture.
Step by step guide to create your first Free AzureML account.
Overview of AzureML studio and various components of it.
- Web services
- Trained Models
Workflow of Azure Machine Learning experiment.
- GET THE DATA
- PREPARE THE DATA
- FEATURE SELECTION
- CHOOSE AND APPLY LEARNING ALGORITHMS
- TRAIN AND EVALUATE THE MODEL
In this lecture we will cover the Azure ML Cheat Sheet for model selection.
In this lecture we will cover how to upload a dataset to the azure ML Workspace and also how to enter the data manually.
In this lecture we will cover how to convert the dataset format as well as how to unpack the zipped dataset.
In this lecture we cover how to import the data from external sources such as an HTTP link.
In this lecture, we are going to cover first part of the Data Transformation. We will cover some of the modules in Data Manipulation.
- Add Columns
- Add Rows
- Remove Duplicate Rows
- Select Coumns in a dataset
This lecture will cover data manipulation using
- Apply SQL Transformation
- Edit Metadata
- How to clean the missing values in a dataset
This lecture covers, two important modules of AzureML data processing; Partition and Sample and Split Data for
- How to partition the data to create train and test datasets
- Create different bins of data for cross validation of your results
- Create a random sample of observations or a more balanced dataset
You will learn the intuition behind the Logistic Regression and derive the mathematical formula for Logistic Regression.
Experiment to create the Logistic Regression model. We are going to build a model that will predict whether the loan application will get approved or not.
Every untrained model expects a set of parameters. They are also known as Hyperparameters. This lecture explains various parameters used for Logistic Regression.
Understanding the classification results and associated metrics such as
- Precision and recall
In this quick lecture, we are going to analyse the impact of certain parameters on the outcome of our model.
In this lecture, we will learn how to predict an outcome that can have multiple values. We are going to use the wine quality dataset and predict the quality of wine based on various characteristics or physiochemical properties of wine, that may affect its quality, such s the acidity, citric acid, residual sugar in it, density and so on.
We will learn what is a decision tree and how it gets constructed. We will build a decision tree on a small sample of data.
In this lecture we are going to cover what is known as Ensemble learning along with the two most popular techniques of Bagging and Boosting.
We are going to build a model based on Two Class Boosted decision tree using the data related with direct marketing campaigns of a banking institution.
In this lecture, we are going to cover one of the most interesting and very popular model called Decision Forest. Using Adult Census data, we will predict whether an individual earns more than $50K or not.
Multiclass Decision Forest using IRIS data which remains one of the most popular datasets on UCI.
Intuition of Support Vector Machine.
Adult census classification using Two Class SVM in AzureML.
How to use Tune Model Hyperparameter as well as how to train and configure the same.
Typical model deployment challenges and what is AzureML webservice.
Create and set up the web service for the machine learning model.
Consume the web service and use the end points using excel.