This website uses cookies to ensure you get the best experience.
Learn more.
A long training job can tolerate interruption if progress is saved. Which two features support a cost-conscious SageMaker training design? Select 2 answers.
Published
- Keep the only checkpoint on ephemeral local storage with no persistence.
- Assume Spot capacity cannot be interrupted.
- Configure supported checkpoints and verify resume behavior.
- Use Managed Spot Training for a compatible workload.
- Remove the maximum training duration so failed jobs consume resources forever.
Comments — 0
Sign in to comment