You want your own TensorFlow code with SageMaker AI managing the infrastructure. What do you use?
Why B is the answer
Script mode with a supported framework
Script mode runs your own framework code, while SageMaker AI manages the training infrastructure.
In detail
- Script mode runs your own code on a supported framework.
- There is no need to build a container.
Use case
An illustrative scenario, not a real incident.
A team keeps its own TensorFlow code, and SageMaker AI runs it on managed infrastructure through script mode.
See it in different ways
mindmap
root((SageMaker AI))
Data
Data Wrangler
Ground Truth
Clarify
Train
JumpStart
Automatic model tuning
Debugger
Deploy
Pipelines
Model Registry
Operate
Model Monitorflowchart LR S0["Write training script"] --> S1["Choose framework"] --> S2["SageMaker AI runs it"]
Key point: Script mode runs your code on managed infrastructure.
Why the other options are wrong
- A. SageMaker Canvas
- Canvas is no-code.
- C. A notebook that you run by hand on an EC2 instance
- That leaves you managing the infrastructure.
- D. A built-in XGBoost algorithm
- A built-in algorithm is not your own code.