What should the Machine Learning team do to address the requirements with the least amount of code and fewest steps?
Implement an AWS Lambda function to log Amazon SageMaker API calls to Amazon S3. Add code to push a custom metric to Amazon CloudWatch. Create an alarm in CloudWatch with Amazon SNS to receive a notification when the model is overfitting.
Use AWS CloudTrail to log Amazon SageMaker API calls to Amazon S3. Add code to push a custom metric to Amazon CloudWatch. Create an alarm in CloudWatch with Amazon SNS to receive a notification when the model is overfitting.
Implement an AWS Lambda function to log Amazon SageMaker API calls to AWS CloudTrail. Add code to push a custom metric to Amazon CloudWatch. Create an alarm in CloudWatch with Amazon SNS to receive a notification when the model is overfitting.
Use AWS CloudTrail to log Amazon SageMaker API calls to Amazon S3. Set up Amazon SNS to receive a notification when the model is overfitting
Explanations:
While implementing an AWS Lambda function to log SageMaker API calls and pushing a custom metric to CloudWatch is feasible, it introduces more complexity and extra steps than necessary. Lambda is not required when CloudTrail can log API calls natively.
AWS CloudTrail automatically logs Amazon SageMaker API calls to S3. Pushing a custom metric to CloudWatch is a simple task during model training, and creating a CloudWatch alarm with SNS for overfitting notifications is a valid and efficient solution.
AWS Lambda does not need to be involved in logging SageMaker API calls, as AWS CloudTrail can handle this directly. The addition of Lambda makes the solution unnecessarily complicated.
While CloudTrail logs SageMaker API calls to S3, CloudTrail does not have built-in functionality to monitor model overfitting. The approach lacks the necessary step of setting up a CloudWatch alarm for overfitting.