From the course: Machine Learning with SageMaker by Pearson

Unlock this course with a free trial

Join today to access over 25,200 courses taught by industry experts.

Building and automating ML pipelines in SageMaker

Building and automating ML pipelines in SageMaker - Amazon SageMaker Tutorial

From the course: Machine Learning with SageMaker by Pearson

Start my 1-month free trial Buy for my team

Building and automating ML pipelines in SageMaker

“

Thus far in this course, all of the work that we've done to create a machine learning model has been done in a manual fashion. However, in a production environment, you're most likely going to want to use some sort of CICD process, that's Continuous Integration, Continuous Deployment, CICD process in order were to go from new data to deployed machine learning model. And that's where ML pipelines come into play. These are end-to-end workflows that allow you to automate that process to create the model. So you could have something trigger the pipeline, such as new data is available or a new data set exists in S3. And then that could trigger the training, or rather the preparation of your data, so preprocessing of your data, the training, evaluation, and deployment of this new data set into a machine learning model. This allows you to reproduce and recreate your models very quickly in the event new data is received. So why should we use SageMaker pipelines instead of something else?…

Contents