MLS-C01 · Question #44
A Machine Learning Specialist is developing a custom video recommendation model for an application. The dataset used to train this model is very large with millions of data points and is hosted in…
The correct answer is A. Load a smaller subset of the data into the SageMaker notebook and train locally. Confirm that the. The standard AWS best practice is a two-phase approach: (1) Load a small representative subset of the data into the SageMaker notebook to prototype, test data pipelines, and confirm the algorithm and preprocessing logic work correctly. (2) Once validated locally, launch a full…
Question
A Machine Learning Specialist is developing a custom video recommendation model for an application. The dataset used to train this model is very large with millions of data points and is hosted in an Amazon S3 bucket. The Specialist wants to avoid loading all of this data onto an Amazon SageMaker notebook instance because it would take hours to move and will exceed the attached 5 GB Amazon EBS volume on the notebook instance. Which approach allows the Specialist to use all the data to train the model?
Options
- ALoad a smaller subset of the data into the SageMaker notebook and train locally. Confirm that the
- BLaunch an Amazon EC2 instance with an AWS Deep Learning AMI and attach the S3 bucket to
- CUse AWS Glue to train a model using a small subset of the data to confirm that the data will be
- DLoad a smaller subset of the data into the SageMaker notebook and train locally. Confirm that the
How the community answered
(45 responses)- A80% (36)
- B4% (2)
- C2% (1)
- D13% (6)
Explanation
The standard AWS best practice is a two-phase approach: (1) Load a small representative subset of the data into the SageMaker notebook to prototype, test data pipelines, and confirm the algorithm and preprocessing logic work correctly. (2) Once validated locally, launch a full SageMaker managed training job that reads data directly from S3 at scale - the training job runs on a separate, appropriately sized compute instance and is not constrained by the notebook's 5 GB EBS volume. This avoids the need to move terabytes of data onto the notebook while still letting the Specialist develop and validate their approach interactively. Options B, C, and D all involve workarounds (attaching S3 to EC2, using Glue, or never scaling up) that are either incorrect or inefficient.
Topics
Community Discussion
No community discussion yet for this question.