Machine Learning School
Early-bird price (until March 31st): $300 ($385)
Many people know how to train Machine Learning models.
Unfortunately, this represents around 5% of the code of an end-to-end system.
This program will show you the other 95% and help you understand Machine Learning systems as a whole. My goal is to give you a framework to analyze and build a solution for any future problems you find.
The program centers around a 9-hour live cohort. We'll use AWS SageMaker to take a problem through the entire Machine Learning lifecycle. We will train, tune, deploy, and monitor the model.
You pay once to join the program and get lifetime access to every class, session, course, tutorial, and resource I bring to the community in the future. No recurrent payments. Ever.
What Do You Get From Joining?
When you join the program, you get access to the following:
- The "Machine Learning In Production" Course: A 9-hour, live cohort focused on building Machine Learning Systems.
- A group of curated resources that will help in your Machine Learning career.
- Guest sessions and future classes (I'm working with other creators to make this happen.)
Who Is This Program For?
This program is for anyone who wants to use Machine Learning to build solutions for real-world problems.
This is a technical program, not a business one. Consider a different option if you aren't comfortable reading and writing Python.
What Are The Prerequisites?
- The program is for people with experience writing software and a good grasp of Machine Learning terminology.
- You should be comfortable understanding Python, at least. Ideally, you should be proficient in writing Python to get most of the value from the program.
- You don't need experience building Machine Learning models, but you should be familiar with Machine Learning terminology. For example, you don't need to understand the architecture of a Deep Learning model, but you should understand what "training" a model means.
What Will Happen After You Finish The Program?
- You'll have a framework to use Machine Learning to solve real-world problems.
- You'll have a sound understanding of how to use AWS SageMaker to build an end-to-end system.
- You'll have a working example of training, tuning, deploying, and monitoring a Machine Learning system that will serve as a template for future projects.
- You'll build the skills you need to join a team, start a new project, build your startup, and accomplish something many companies worldwide don't have.
- You'll have the confidence to take a step up in your career. You'll learn something only a few people know, putting you ahead of the pack.
- You'll learn a few tricks from somebody who spent many nights trying to figure these things out.
- You'll get a certificate of completion.
What Shouldn't You Expect From This Program?
- This program is not an introduction to Machine Learning. We will focus on the practical aspects of Machine Learning systems but will not cover Machine Learning theory.
- This program is not an introduction to Python or software development. We will write a lot of code, and the program assumes you are familiar with the language and are competent in building software.
- This program is not a shortcut to changing your career. You will build skills many companies want, but a well-rounded engineer needs much more than you'll see here.
What Does The Program Look Like?
When you join the community, you'll get access to a 9-hour live cohort:
- Pre-recorded lectures.
- 6 live sessions of 90 minutes each (9 hours total).
- A recorded version of every session.
- Source code and resources.
Every month, the time of the cohort changes, so you can pick the schedule that better fits you. We will meet 3 times for 2 consecutive weeks over Google Meet. I will record every session, and you can access them anytime.
Here are the upcoming cohorts:
- April 17 - 28, 2023. 10 am EST. (Monday, Wednesday, and Friday)
- May 15 - 26, 2023. 2 pm EST. (Monday, Wednesday, and Friday)
- June's schedule is still TBD.
What's the Structure of the Program?
We will start the program by building a simple model and taking it through the entire Machine Learning lifecycle.
I packed every session with information and code. It will be intense. There will be many optional assignments.
Here is the structure of the course:
1. Getting Started
- An overview of SageMaker
- SageMaker's Python SDK
- Notebook instances
- Setting up the development environment
2. Training Models in SageMaker
- Building a simple model
- Preparing a training script
- Introduction to SageMaker Estimators
- Training our model using a Training Job
3. Channels and Hyperparameter Tuning
- Training a model from data stored in S3
- Hyperparameter Tuning in SageMaker
4. Deploying Models in SageMaker
- Overview of model deployment in SageMaker
- Deploying a model from a SageMaker Estimator
- Deploying a model trained outside SageMaker
5. Custom Endpoints
- Using custom pre/post-processing inference code
- Compiling and deploying a custom endpoint image
6. Monitoring and Scaling Models
- Setting up model monitoring
- Scaling endpoints automatically