Creating data science environment
In the previous section, we introduced high-level Amazon SageMaker features that can often be used in isolation or together for end-to-end capabilities. In this section, we will focus on creating consistent and repeatable governed data science environments that can take advantage of the features discussed in the first section.
To build, train, and deploy models using Amazon SageMaker, ML builders need access to select AWS resources spanning the ML development life cycle. Because many different personas may be responsible for building ML models, the term ML builder refers to any individual tasked with model building. This could include data scientists, ML engineers, or data analysts.
Data science development environments provide ML builders with the AWS resources they need to build and train models. A data science environment could be as simple as an AWS account with access to Amazon SageMaker as well as AWS services commonly used with Amazon...
