A company's machine learning (ML) specialist is designing a scalable data storage solution for Amazon SageMaker. The company has an existing TensorFlow-based model that uses a train.py script. The…

Question

Options

• APut the TFRecord data into an Amazon S3 bucket. Use AWS Glue or AWS Lambda to reformat
• BRewrite the train.py script to add a section that converts TFRecord data to protobuf format. Point
• CUse SageMaker script mode, and use train.py unchanged. Point the SageMaker training
• DUse SageMaker script mode, and use train.py unchanged. Put the TFRecord data into an

• A
  4% (1)
• B
  4% (1)
• C
  17% (4)
• D
  75% (18)

Explanation

SageMaker script mode lets you bring your own training script (train.py) and framework without modification. TFRecord is TensorFlow's native binary format, which train.py already knows how to read. By placing the TFRecord files in Amazon S3 and pointing the SageMaker training job's data channel to that S3 path, the data is streamed directly into the container with zero code changes. Option A requires reformatting the data (unnecessary overhead). Option B requires rewriting train.py to add a conversion step (defeats the purpose). Option C is similar to D but implies File mode instead of S3 URI pointing, which is less straightforward - the key distinction in D is that the TFRecord data stays in S3 and train.py is used unchanged.

No community discussion yet for this question.