AWS Certified Generative AI Developer - Professional
Get started today
Ultimate access to all questions.
Ultimate access to all questions.
No comments yet.
Which approach has the least operational overhead for implementing a model evaluation workflow that uses datasets stored in Amazon S3, publishes results to Amazon CloudWatch, and creates dashboards for accuracy metrics?
A
Configure CORS on the S3 bucket to allow cross-origin requests from Amazon Bedrock evaluation workflows
B
Store standardized evaluation inputs in Amazon S3 and grant the evaluation workflow appropriate IAM permissions to read the datasets
C
Use a scheduled AWS Lambda function to start evaluation jobs with the S3 dataset location, and another Lambda function to check job status and push results to CloudWatch
D
Use Amazon SageMaker AI notebook jobs with the fmv elos or ragas framework to run evaluations on datasets in S3
E
Run an Amazon SageMaker AI notebook job on a schedule by using the fmv elos or ragas framework to run evaluations that use the datasets in the S3 bucket. Write Python code in the notebook that makes direct InvokeModel API calls to the FMs and processes their responses for evaluation. Publish job status and results to Amazon CloudWatch Logs to measure the real world knowledge (RWK) score for text generation and toxicity for summarization as metrics for accuracy. Create a custom CloudWatch Logs Insights dashboard.