From the course: Amazon Web Services Machine Learning Essential Training
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Understanding SageMaker - Amazon Web Services (AWS) Tutorial
From the course: Amazon Web Services Machine Learning Essential Training
Understanding SageMaker
- [Instructor] As we get started working with our next service, AWS SageMaker, let's take a look at a typical machine learning model workflow. You can see in this diagram, there are three high-level steps. To generate example data, train a model, and deploy the model, and then there are substeps of fetching, cleaning, and preparing the data, training and evaluating the model, and deploying and monitoring the deployed model. SageMaker has components that fit with each of these parts. SageMaker is a relatively new service at the time of this recording. It's only been in general availability since this last year's reinvent, so three months ago, and it has four general steps. The steps are a component that supports building, so it allows connection to other AWS services, and transforming data using something called a SageMaker notebook. Then training. You can use SageMaker's algorithms and frameworks, or you can use your own for distributed training. Distributed is really key here. The…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
(Locked)
Understanding ML platforms3m 53s
-
(Locked)
Understanding and using AWS Machine Learning9m 15s
-
(Locked)
Understanding SageMaker3m 54s
-
Create Jupyter notebooks with SageMaker6m 12s
-
(Locked)
Get data with SageMaker notebook6m 25s
-
(Locked)
Train model with SageMaker job3m 6s
-
(Locked)
Deploy and host model with SageMaker model2m 31s
-
(Locked)
Use model from SageMaker endpoint4m 7s
-
(Locked)
Selecting algorithm for model training5m 36s
-
(Locked)
Advanced use of SageMaker2m 37s
-
(Locked)
-
-
-