Prevent boto3.client('sagemaker').create_auto_ml_job() from deploying endpoint

0

When I invoke the .create_auto_ml_job() method both with and without the optional ModelDeployConfig kwarg, the autopilot job deploys an endpoint using the best model. Is there a way to prevent the .create_auto_ml_job() method from behaving this way? I do not wish to deploy the best model to an endpoint, and do not wish to have to delete this endpoint.

2 Answers
0

Thank you for reaching out to us.

I understand that you would like to prevent boto3.client('sagemaker').create_auto_ml_job() from deploying endpoint as you do not wish to deploy the best model to an endpoint, and do not wish to have to delete this endpoint, Please correct me If I am missing anything here.

I have followed the following example[1] to create the auto ml job using the boto3 create_auto_ml_job() API call with out using the modelDeployConfig, However noticed that the SageMaker creates a model, endpoint and endpoint config for an internal validation step by design and are deleted once the step is completed. It's just happening for the intermediate step and do not have any endpoint deployed which is up and running, you will have the artifacts and notebook which an be used later to deploy the model.

If you have any additional questions please feel free to open a case with sagemaker aws support for additional information along with your resource name, region and the sample code used for the example

Note: Due to security reason, this post is not suitable for sharing customer's resource.

Reference:

[1] https://sagemaker-examples.readthedocs.io/en/latest/autopilot/sagemaker_autopilot_direct_marketing.html

AWS
answered a year ago
0

According to this AWS documentation (see [1]), AutoML does launch a clarify job automatically:

Autopilot also helps explain how models make predictions using a feature attribution approach developed for Amazon SageMaker Clarify. Autopilot automatically generates a report that indicates the importance of each feature for the predictions made by the best candidate. This explainability functionality can make machine learning models more understandable to AWS customers. The model governance report generated can be used to inform risk and compliance teams and external regulators.

Currently there is no option to disable that, but you can generate job defenitions only for the AutoML job which will result in the job launching couple of processing jobs then producing couple of Jupyter Notebooks which container code to run AutoML job manually, from there oyu are free to ammend the code as required.

To do the above, please set the below parameter to True:

GenerateCandidateDefinitionsOnly=True

Please try the above and see how you go.

References:

[1] https://docs.aws.amazon.com/sagemaker/latest/dg/autopilot-automate-model-development.html : Automate model development with Amazon SageMaker Autopilot [2] https://boto3.amazonaws.com/v1/documentation/api/latest/reference/services/sagemaker.html#SageMaker.Client.create_auto_ml_job

Sam_E
answered a year ago

You are not logged in. Log in to post an answer.

A good answer clearly answers the question and provides constructive feedback and encourages professional growth in the question asker.

Guidelines for Answering Questions