0

I've re-trained a model and want to deployed to existing SageMaker endpoint, so that application don't need to make any change on the SageMaker Endpoint.

I use PyTorchModel from sagemaker.pytorch

When calling pytorch_model.deploy(instance_type='ml.m5d.large', initial_instance_count=1, endpoint_name="some-existing-endpoint")

Get an Error

botocore.exceptions.ClientError: An error occurred (ValidationException) when calling the CreateEndpointConfig operation: Cannot create already existing endpoint configuration "arn:aws:sagemaker:us-east-1:xxxxxxxx:endpoint-config/some-existing-endpoint".

I have seen a suggestion using update_endpoint=True,

But seems like doesn't have this options, with error:

TypeError: _create_sagemaker_model() got an unexpected keyword argument 'update_endpoint'

Anyone know how to deploy PyTorchModel to existing SageMaker endpoint?

Improve this question
2
  • update_endpoint=True seems to be deprecated, have you tried something like this - stackoverflow.com/questions/64169189/…? Commented Jul 15, 2024 at 16:14
  • that's the articles I referred for update_endpoint, as the other options, how to incorporate pytorchmodel with sklearner Commented Jul 16, 2024 at 9:30
Related questions
2
Not able to create an sagemaker endpoint with datacaptureconfig enabled using boto3 API
1
Sagemaker model deployment failing due to custom endpoint name
0
Sagemaker endpoint Invocation fails from external application
Load 6 more related questions Show fewer related questions

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.