O'Reilly logo
live online training icon Live Online training

Managed Machine Learning Systems and Internet of Things

Learn the building blocks of Managed ML and Hardware AI with Python

Noah Gift

The next evolution of AI and ML is cloud-native, managed platforms and custom hardware AI, inclusive of things like custom AI chips, Internet of Things devices, and more. AWS has Sagemaker, which allows for a fully-managed build, train and deployment lifecycle, including automatic hyperparameter tuning. Microsoft Azure has Azure ML Studio which includes high-level tools allow for drag and drop workspace machine learning workflows. Google has AutoML, which allows developers with limited machine learning expertise to train high quality models, by automatically inferring the correct hyperparameters and model to use. In this training, learn to how to use these managed platforms to create solutions in a fraction of the time as a “roll your own" ML solution. Additionally, learn to compare how each cloud managed solution compares and be able to pick the right solution for the task at hand.

Learn to use these hardware devices to create end-to-end ML solutions by integrating both IoT, and on device ML models. Two examples covered are Apple which has developed a dedicated AI chip used in the Core ML2 framework, and AWS DeepLens, which can run Sagemaker trained models. Both Hardware AI and Managed ML services are both tightly coupled and rapidly innovating. In the very near future it is possible all production Machine Learning will be using Hardware AI components of some variety: (IoT, GPU, TPUs) and Managed and Auto ML solutions.

What you'll learn-and how you can apply it

  • Utilize Hardware AI to build products including: TPUs, GPUs, FPGAs, A11 Bionic
  • Compare, choose, and implement, Auto and Managed Machine Learning Systems including: Google Auto-ML, AWS Sagemaker and Azure ML Studio
  • Perform IoT programming fundamentals with Python, AWS Greengrass, AWS Deep Lens and Raspberry Pi
  • Build production Machine Learning models with AWS Sagemaker
  • Create iOS Core ML2 Applications with Swift Playgrounds
  • Train a computer to identify numbers in handwriting and identify cars in pictures

This training course is for you because...

  • You’re a technical leader who wants to know what is next for Machine Learning and AI
  • You’re a currently involved in Data Science, Analytics or Machine Learning training and are looking for additional material to supplement your learning.
  • You’re a software developer who wants to understand how to get more deeply involved in the Data Science movement.
  • You’re a currently involved in Data Science, Analytics or Machine Learning training and are looking for additional material to supplement your learning.
  • You are Business and analytics professional with some SQL experience and are looking to move to the next generation of Data Science.
  • You are a Junior Data Scientist who is looking to expand into more advanced ML topics


  • Some previous programming experiences in any language is helpful: should understand how to execute statements and write functions. Python experience preferred, but not required.
  • Basic understanding of statistics, probability and machine learning terminology
  • Explore Topics in Pragmatic AI Book with special attention to chapters on cloud and machine learning

Course Set-up:

  • Jupyter notebook either Google Colaboratory or local. Google Colab preferred
  • Python 3.6 (if you cannot use Google Colab Notebooks)
  • Optional AWS, GCP and Azure Account
  • Optional XCode 9.4 or Higher

Recommended Preparation:

Recommended Follow-up:

About your instructor

  • Noah Gift is lecturer and consultant at both UC Davis Graduate School of Management MSBA program and the Graduate Data Science program, MSDS, at Northwestern. He is teaching and designing graduate machine learning, AI, Data Science courses and consulting on Machine Learning and Cloud Architecture for students and faculty. These responsibilities including leading a multi-cloud certification initiative for students. He has published close to 100 technical publications including two books on subjects ranging from Cloud Machine Learning to DevOps. Gift received an MBA from UC Davis, a M.S. in Computer Information Systems from Cal State Los Angeles, and a B.S. in Nutritional Science from Cal Poly San Luis Obispo.

    Professionally, Noah has approximately 20 years’ experience programming in Python. He is a Python Software Foundation Fellow, AWS Subject Matter Expert (SME) on Machine Learning, AWS Certified Solutions Architect and AWS Academy Accredited Instructor, Google Certified Professional Cloud Architect, Microsoft MTA on Python. He has worked in roles ranging from CTO, General Manager, Consulting CTO and Cloud Architect. This experience has been with a wide variety of companies including ABC, Caltech, Sony Imageworks, Disney Feature Animation, Weta Digital, AT&T, Turner Studios and Linden Lab. In the last ten years, he has been responsible for shipping many new products at multiple companies that generated millions of dollars of revenue and had global scale. Currently he is consulting startups and other companies.


The timeframes are only estimates and may vary according to how the class is progressing

Segment 1 Hardware AI and IoT Fundamentals (180 minutes)

Part One: 1.5 Hours

  • Learn Introductory Concepts in Hardware AI and IoT
  • Explore current state of Hardware AI chips and IoT technology
  • Learn what TPUs are, and classify handwriting samples using Google Cloud Platform (GCP) and TensorFlow
  • Learn how GPUs accelerate Machine Learning workloads on both AWS (Amazon Web Services) and GCP
  • Learn about FPGAs (Field Programmable Gate Arrays) on AWS
  • Learn about dedicated neural network hardware in Apple’s A11 chip for the iPhone
  • Learn AWS IoT programming fundamentals
  • Explore the AWS Greengrass IoT Framework
  • Explore AWS DeepLens SDK including custom Sagemaker model deployment
  • Explore Raspberry Pi and AWS Greengrass integration
  • Example project using Python and AWS Lambda on IoT devices
  • Q/A: 15 Minutes
  • Break: 15 Minutes

Part Two: 45 Minutes

  • Learn iOS Core ML Fundamentals with Swift Playgrounds
  • Explore getting, converting and integrating Core ML Models
  • Learn to classify Images with Vision and Core ML
  • Learn to analyze natural language text with Core ML
  • Q/A: 15 Minutes

Segment 2 Managed and Auto Machine Learning Fundamentals (180 minutes)

Part One: 1.5 Hours

  • Begin building production ML models with AWS Sagemaker
  • Learn to Use AWS Sagemaker Notebooks to do data science and machine learning
  • Learn to how to Train AWS Sagemaker custom Model and use automatic hyperparameter auto-tuning
  • Learn how to train and host Scikit-Learn models by building Scikit Docker container
  • Learn how to scale up AWS Sagemaker training and hosting
  • Learn to operationalize ML Models with AWS Sagemaker
  • Learn to create hosting endpoints for AWS Sagemaker models
  • Learn how to use Chalice/AWS Lambda for Production Deployments
  • Explore how to perform model A/B testing for AWS Sagemaker models
  • Learn to create automl solutions with AWS Machine Learning service
  • Learn to train a model predict responses to a Marketing offer
  • Learn to use the ML Model to generate predictions
  • Explore automl features of AWS Machine Learning services: Feature Transformations with Data Recipes, hyperparameter tuning and more.
  • Q/A: 15 Minutes
  • Break: 15 Minutes

Part Two: 45 Minutes

  • Learn to use Google (GCP) Managed and Auto-ML Machine Learning Solutions
  • Learn to perform Machine Learning with Big Query datasets
  • Explore GCP AutoML service
  • Learn to use GCP Machine Learning Engine with scikit-learn and XGBoost
  • Learn to classify handwriting with TPUs and the Tensorflow framework using Datalab and MINST dataset.
  • Learn to use Microsoft Azure Machine Learning Studio
  • Explore drag and drop visual workspaces with Azure Machine Learning Studio
  • Learn to train an ML model using Automobile Price History
  • Create predictions using Azure Machine Learning Studio model
  • Q/A: 15 Minutes