Intelligent Analytics and Automation with ML & AI


Course Overview

Machine learning products are being used daily, perhaps without realizing it. The future of machine learning is already here, so gearing yourself up with machine learning skills is a good move now. You and your teams can use Artificial Intelligence to deliver value to your customers. You too can build and deploy machine learning products to meet business opportunities.

You’ll learn what Artificial Intelligence is, and the fundamentals of Machine Learning and Deep Learning. The knowledge will be applied to different technologies, in different programming languages, and on different Cloud platforms. The tools will always be evolving, but the methodology remains. In this 5-day, hands-on course, you will learn how you can leverage Artificial Intelligence in many business scenarios.


Key Learning Areas

Attendees will acquire first-hand experience in using Machine Learning and the different tools around it:

  • How you, your team, and your company should think the Machine Learning Lifecycle
  • How to choose the right algorithm for the right problem
  • How to use the right metrics to monitor your models
  • How to use libraries in R and Python to predict values
  • How to build a Neural Network
  • How to use Cognitive Services and how industry takes advantage of them
  • How to simplify the Machine Learning process with Azure Machine Learning Studio and Amazon SageMaker
  • How to identify and mitigate the risks of Machine Learning

Each course module consists of a lecture followed by a hands-on lab designed to reinforce the principles presented during lecture.


Course Outline

Introduction to Machine Learning

  • The Machine Learning Process
  • Supervised Learning
  • Unsupervised Learning
  • Azure ML Studio
  • Lab: Machine Learning with Azure Machine Learning Studio

Visualization and Statistics

  • Data Visualization
  • Histogram
  • Scatter plot
  • Just enough statistics
  • Variables
  • Lab: Data Visualization
  • Lab: Compare two models

Improving and Deploying Machine Learning Models

  • Ensemble models
  • Bagging
  • Boosting
  • Deploying your model
  • Lab: Principal Component Analysis
  • Lab: Deploying your Model
  • Lab: Using your Model

Machine Learning with R – R Basics

  • Why R
  • R Overview
  • R Introduction
  • R Interface
  • R Workspace
  • Help
  • R Packages
  • Input/output
  • Reusing Results
  • Lab: Iris Dataset

Machine Learning with R – Data Input

  • Data Types
  • Importing Data
  • Missing Data
  • Date Values
  • Lab: Iris Data Set – Visualize Dataset

Machine Learning with R – Data Manipulation

  • Creating New Variable
  • Operators
  • Control Structures
  • Built-in functions
  • Sorting Data
  • Merging Data
  • Data Type Conversions
  • Lab: Iris Dataset – Evaluate and Predict

Machine Learning with Python – Data Handling

  • Introduction to NumPy
  • Arrays
  • Aggregations
  • Introduction to Pandas
  • Handling missing Data
  • Combining datasets
  • Lab: Predicting Bicycle Traffic
  • Lab: K-Means on digit

Machine Learning with Python – Visualization and Modeling

  • Introducing Matplotlib
  • Simple Line Plots
  • Simple histogram
  • Introducing Scikit-Learn
  • Model Validation
  • Feature Engineering
  • Linear Regression
  • Principal Component Analysis
  • Decision Trees and Random Forests
  • Lab: Random Forest for Classifying Digits
  • Lab: Eigen faces

Robotic Process Automation (RPA)

  • Introduction to RPA
  • What are the benefits of RPA?
  • What are the pitfalls of RPA?
  • Comparing the different major RPA systems available
  • Business applications for RPA
  • Lab: creating a basic robot using UIPath

Machine Learning with Internet of Things (IoT)

  • What is IoT?
  • How is it used?
  • Characteristics of an IoT solution
  • Why Azure IoT?
  • Lab: Create an Azure IoT Central application

Deep Learning

  • What is Deep Learning?
  • Why is it useful?
  • Activation functions
  • Regularization Methods
  • Basic architectures
  • Azure Cognitive Services
  • Lab: Handwritten Digit Recognition using Convolutional Neural Networks

Azure Cognitive Services: Audio-Based Machine Learning

  • What services are offered for Audio
  • What services are offered for Speech
  • What are their purposes?
  • How are these services being utilized today?
  • LAB: Setting up recognized speakers with Speaker Recognition
  • LAB: Invoking audio-based services.
  • Lab: Having your say… Make your Desktop app obey only recognized voices

Azure Cognitive Services: Computer Vision/Image-Based Services

  • What is Computer Vision?
  • What services are offered for Computer Vision?
  • Other Image-based services
  • Examples of Image-based services utilized today
  • Lab: Recognizing a face from an image.
  • Lab: Step by Step: Invoking image-based services
  • Lab: Building a Vision Accessibility feature which describes the objects found in the image taken for the blind

Azure Cognitive Services: Text/Language-Based Services

  • What are the Text-based services?
  • What is their purpose?
  • What are the Language-based services?
  • What is natural language processing?
  • Examples of text and language services
  • LAB: Analyzing Intent from text messages
  • LAB: Building a LUIS model for Hair and Nail Salon conversations
  • LAB: invoking natural language services
  • Lab: Building a Bot which determines bad intent
  • Lab: Building a Bot which automates the scheduling of a Hair appointment

Introduction to Machine Learning with AWS

  • What is Amazon SageMaker?
  • Machine Learning with Amazon SageMaker
  • Explore, Analyze and Process Data
  • How it is used in the industry
  • Lab: Build a recommendation system

Building Machine Learning Models with AWS

  • Train Models with SageMaker
  • Deploy Models with SageMaker
  • Lab: Build a recommendation system – Part 2

Security Risks of Machine Learning Models

  • Data Poisoning
  • Evasion Attacks
  • Inversion by Surrogate Model
  • Adversarial Attacks
  • Impersonation

Mitigating Risks of Machine Learning Models

  • Model Monitoring with Azure Machine Learning
  • Data Drift Monitoring with Azure Machine Learning
  • Models Explainers with Azure Machine Learning

Who Benefits

This course is designed for software developers and for business consultants who want to learn the landscape of Artificial Intelligence. The hands-on labs involve a broad range of technologies with some light coding. Coding is done in Python, R, C#. Some labs will be done with visual (non-coding) programming. The course is valuable for non-developers as many business scenarios are presented.

In order to work along with the labs, attendees must have the following:

  • A modern web browser
  • An Azure subscription (Azure Pass, MSDN subscription, free trial, company subscription, etc.)
  • An AWS Subscription
  • Local administrative permissions to install additional tools that may be required during the course of the labs


Basic programming experience is a plus

Want this course for your team?

Atmosera can provide this course virtually or on-site. Please reach out to discuss your requirements.