Full Stack Machine Learning in AWS (Batch #7)

Many AI and machine learning courses cover theoretical techniques of algorithms. In this course, we will train you to become a full stack machine learning engineer, capable of not just training models but also deploying and managing them in production.
You will learn machine learning primarly through building 8 production grade services, step by step.
You will learn how to build production machine learning services in AWS, how to integrate them with an application, and how to manage it through lifecycle.

Students who take this course will be able to:

  • Identify and frame business use cases that can be solved by AI and machine learning
  • Choose the right techniques, tools, frameworks to the business use cases
  • Build production machine learning services on AWS, and manage through their lifecycle.
  • Hands on implementation of end to end machine learning services
    • Session 1: July 27, 10:00am-11:30am PST (US Pacific Time)
    • Session 2: July 29, 10:00am-11:30am PST
    • Session 3: August 3, 10:00am-11:30am PST
    • Session 4: August 5, 10:00am-11:30am PST
    • Session 5: August 10, 10:00am-11:30am PST
    • Session 6: August 12, 10:00am-11:30am PST
    • Session 7: August 17, 10:00am-11:30am PST
    • Session 8: August 19, 10:00am-11:30am PST

    • 4 weeks / 8 sessions / 12 hours
    • 8 lectures / 8 hands-on projects
    • Capstone project, Github portfolio
    • Live Sessions, Real time interaction
    • Slack support after class and homeworks
    • Life time access to course materials

    Check the content tab for full course outlines.

    Developers, data scientists, students.

  • Basic familiarity with machine learning
  • AWS account (Free to sign up). Students can use their own AWS account to run the hands on labs and have all artifacts (models, datasets) in their account for further use
  • No coding prerequisite. We will provide self paced exercises where attendees, if interested, can code in Python, Java or Javascript.
    Est. time spend per week: 3 hours live class (required) + 3 hours homework (required) + 3 hours projects (bonus, optional).

    Full refund upon request before the first session ends (July 27th, 2020 12:00pm PT). 5% transaction fee is not refundable.

    If missed live sessions, you can watch recordings any time, along with interactive learning tools, slides, course notes
    Students have life time access to course materials

  • Earn Certificate of Completion
  • Scholarship is available, contact us for details
  • Batch #6: Jun 15 ~ Jul 8, 2020
  • Batch #5: May 11 ~ Jun 3, 2020
  • Batch #4: Mar 17 ~ Apr 9, 2020
  • Batch #3: Jan 31, 2020
  • Batch #2: Jan 28 ~ Feb 20, 2020
  • Batch #1: Oct 30 ~ Nov 22, 2019
  • This course includes 8 hands on workshops, and each workshop will cover:
  • A business use case and how machine learning maps to that use case
  • An AI/ML algorithm and AI/ML internals technical topic. For example, how and when to use Regression, Classification and how to map the right powerful AI algorithms for each data type.
  • How to build a production machine learning services (Feature Engineering, Model Training, Model Validation, Endpoints, Gateway/Lambda Integration, Application Integration).
  • One or more AWS tools. Across the 8 sessions we will cover AWS Sagemaker, BuiltIn Algorithms, Custom Docker Containers, Endpoint, AWS Lambda, AWS API Gateway, AWS Roles and Authentication, AWS Cloudwatch, AWS S3, Postman API testing, cURL, Production AI Best Practices. Microservice design pattern for AI deployment. We will also cover Navigator (by Pyxeda.ai), a life cycle overlay tool that makes AWS tools easier to configure and use.
  • You will build 7 working production ML services.

  • We will cover the following projects (business use cases):
  • Customer Churn: Detect whether your customers are about leave your business or your product. Predict which customers are at risk of churn.
  • Pricing Analysis: Understand what factors most affect price. Predict how price can change with features. Assess viability of price for future products
  • Customer approvals: Should you approve a loan for a particular customer? Predict whether a customer is likely to be delinquent on a bill?
  • Appointment planning: Is a customer likely to miss an appointment? Can you plan your appointment schedule more effectively?
  • Removing bias: Bias in your AI data can lead to poor outcomes, unhappy users or even legal problems. How to detect and remove sources of bias?
  • Sentiment Analysis: Are you customer’s happy with your product? What do their comments, tweets and other writings say about their feelings?
  • Recommendation system: What can you learn about your customers or users? Can you analyze their usage and see what else you can upsell to some users?

  • You will learn how to use cloud tools:
  • Configure and use S3 for your data
  • Feature engineer your data with Python code
  • Configure and use AWS Sagemaker. Deep dive on built in AWS Sagemaker algorithms KNN and XGBoost. How to hyper-parameter tune Sagemaker algorithms.
  • Bring custom code into AWS Sagemaker as a Docker container
  • Configuring and using a Sagemaker Endpoint.
  • Connecting a Sagemaker Endpoint to a public URL via AWS Gateway and Lambda.
  • Integrating REST microservices with applications. Using CURL/Postman for API testing.
  • Cloud AWS best practices. Cloudwatch for logs, managing endpoints.
  • Navigator (by Pyxeda.ai) for ease of use. How to use Navigator and AWS together.

  • You will learn machine learning algorithms, algorithm internals and technical concepts
  • How to select an algorithm for a use case, train, deploy and use it in production, and measure how well it is doing.
  • How to map each use case to machine learning - what type of machine learning can be used, for what data type. How to measure.
  • Powerful general purpose algorithms like XGBoost, LinearLearner, KNN, etc. and how they work internally and how to hyper-parameter tune them for best performance.
  • Metrics and practices for algorithmic evaluation and how to map them to use case.
  • AI Trust, Fairness and Bias. Managing AI related risks
  • Machine learning explainability
  • Advanced aspects of production ML - live monitoring and diagnosis, model versioning, retraining and others.
  • Module 1: Build and run a production ML in the cloud - Customer Churn
    In this first session, we will show the steps needed to build and run a production ML in the cloud. We will illustrate these steps with our first business use case, Customer Churn.
    What students will learn:
    • Overview of the production ML lifecycle and all of its steps. How to go from data to running production ML
    • Description of cloud services that can be used for each stage
    • Overview of a customer churn problem and how to build a ML service for it with AWS with Binary Classification Algorithms
    • Lab 1: Build a ML service for Customer Churn and test it.

    Module 2: Evaluating the effectiveness of a production ML - Pricing Analysis
    This session session builds upon the first. We examine the use case of Pricing Analysis and how a production ML based on Regression techniques can address this use case. We delve into more depth on how to evaluate the effectiveness of the production ML.
    What students will learn:
    • How to build and evaluate production ML that use Regression algorithms
    • Overview of a business Pricing Problem and how to build a ML service for it using four powerful Regression Algorithms (Linear Learner, XGBoost, KNN and Factorization Machines)
    • How to bring in data from other sources (like Excel).
    • Common approaches to Feature Engineering - One Hot Encoding and Missing Value handling.
    • Lab 2: Build a ML service for Pricing Analysis and test it

    Module 3: Building and Hyper-Parameter Tuning - Credit Approvals
    In this session, we show how to hyper parameter tune ML algorithms trained in the cloud, and how to retrain and optimize them.
    What stuents will learn:
    • How to build and hyper-parameter tune KNN for Binary Classification in the cloud
    • Overview of the Credit Approvals business use case, and how to build a production ML to solve this problem
    • How hyper-parameters affect solution quality and performance
    • How to retrain a production ML with new information and how to iteratively deploy increasingly accurate models
    • How to bring multiple datasets to bear for the same production ML, optimizing over time
    • Lab 3: Build a production ML service for Credit Approvals using publicly available datasets. also iteratively tune and upgrade the service in production

    Module 4: Test and Evaluate the Quality of a Production ML - Appointment Planning
    In this session, we will focus on methods to test and evaluate the quality of production ML. In particular, we delve into evaluation metrics beyond accuracy and cover both ML metrics (such as Confusion Matrix) and production service metrics (such as latency and scale) that are important for production.
    What students will learn:
    • How to evaluate a production ML with both ML metrics (like Accuracy, Confusion Matrix, True/False Positive/Negative etc.) and code metrics (throughput, accuracy, etc.)
    • How to test their ML service using CURL and Postman
    • How to integrate their ML service into Python, Java or Javascript applications
    • Lab 4: Build a ML services for Appointment planning, run the test experiments and integrate it with an application code module in the language of their choice

    Module 5: Bias in AI/machine learning
    In this session, we focus on the critical issue of AI Bias. We cover the common ways that Bias can enter a production ML implementation and how to detect and remove Bias. As a case study, we use the publicly available COMPAS dataset.
    What students will learn:
    • What is AI Bias and why should it be avoided?
    • How can AI Bias enter? How can it be detected and removed?
    • How to test for Bias and apply Feature Engineering to reduce it.
    • Lab 5: Build a ML service with an open source dataset and test for Bias.

    Module 6: Building a production ML for Text Data - Sentiment Analysis
    This session focuses on text data. We introduce the Sentiment Analysis use case and describe how ML can be used for this purpose. We dive deep into algorithms for Text Classification, particularly the Bag of Words approach, its operation, advantages and limitations.
    What students will learn:
    • How to build and evaluate production ML that use Text Classification algorithms
    • Overview of a business Sentiment Analysis problem and how to build an ML for it using Text Classification in either Binary or Multiclass form
    • How to bring in data from other sources (like Excel) and analyze social media feeds.
    • Lab 6: Build a ML service on AWS for Sentiment Analysis with Twitter feeds

    Module 7: Mapping ML problems to Techniques- Recommendation System
    In this session we combine what we have learned so far into a methodology for mapping problems to ML methods for numerical, categorical and free form text data. We cover in depth a second powerful general purpose AI algorithm - XGBoost, and its hyper parameter tuning.
    What students will learn:
    • A methodology for mapping use cases (with different data types) into ML algorithms and lifecycle steps
    • Internals and hyper parameter tuning for production XGBoost in the cloud
    • Introduction to recommendation system, and sample data sets.
    • Lab 7: Build recommendation system

    Module 8: Advanced topics in production ML, Complete and present your projects
    We will provide an overview of advanced topics. In the code lab, students will complete their custom project and create a github and project repository.
    What students will learn:
    • Best practices for production ML (model versioning, model integrity, retraining cycles, microservice API management and versioning). How to manage application changes and upgrades.
    • Best practices (debugging, instances, your cloud bill :-)).
    • Lab 8: Finish custom project and projects presentation

    Nisha Talagala
    founder of Pyxeda AI. Previously, Nisha co-founded ParallelM which pioneered the MLOps practice of managing machine learning in production. Nisha is a recognized leader in the operational machine learning space.
    Sindhu Ghanta
    Senior Machine Learning Engineer at Pyxeda AI
    • Start Date: ended
    • Venue: Online (zoom)
    • Fee:
      $299 $299 USD
    • Students enrolled:40
    • Status: course ended
    • Course Preview:

    • Any questions? Contact Us