SageMaker Multi Model EndPoint and Inference Data Capture feature

Question

Does Data Capture feature used for model monitor and analytics work with the multi model endpoint (one container).. we ran into an error.  See error " An error occurred (ValidationException) when calling the CreateEndPointConfig operation: Data Capture Feature is not supported with MultiModel mode"
Theoretically, it should work because it is calling the DataCaptureConfig:

from sagemaker.model_monitor import DataCaptureConfig

endpoint_name = 'your-pred-model-monitor-' + strftime("%Y-%m-%d-%H-%M-%S", gmtime())
print("EndpointName={}".format(endpoint_name))

data_capture_config=DataCaptureConfig(
                        enable_capture = True,
                        sampling_percentage=100,
                        destination_s3_uri=s3_capture_upload_path)

Accepted Answer

SageMaker multi-model endpoints do not have support for SageMaker Model monitor as of writing this answer. So the error is pointing to exactly that.

However, if you are looking to implement data drift using sagemaker model monitor then you can do that my mimicking data capture config functionality by capturing inference input and prediction output and storing it in the format supported by Model Monitor. And then setup a customer monitoring container using the instructions listed [https://docs.aws.amazon.com/sagemaker/latest/dg/model-monitor-byoc-containers.html]()

SageMaker Multi Model EndPoint and Inference Data Capture feature

相關內容