- Mais recentes
- Mais votos
- Mais comentários
Hello hugoflores,
You can use SageMaker APIs - https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_DeleteEndpoint.html to delete the endpoint and https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateEndpoint.html. to create an endpoint. This an be automated either using SageMaker Pipelines or a Lambda function.
Here are a few resources towards that:
https://awsfeed.com/whats-new/machine-learning/build-a-ci-cd-pipeline-for-deploying-custom-machine-learning-models-using-aws-services
https://github.com/aws-samples/aws-lambda-layer-create-script
https://github.com/aws/sagemaker-python-sdk/issues/1200
https://aws.amazon.com/blogs/machine-learning/call-an-amazon-sagemaker-model-endpoint-using-amazon-api-gateway-and-aws-lambda/
https://www.sagemakerworkshop.com/step/deploymodel/
HTH,
Chaitanya
Thanks Chaitanya,
I was able to create the endpoint using the "create_endpoint" method and following one of the links provided.
def create_endpoint(endpoint_name, config_name): """ Create SageMaker endpoint with input endpoint configuration. Args: endpoint_name (string): Name of endpoint to create. config_name (string): Name of endpoint configuration to create endpoint with. Returns: (None) """ try: sagemaker.create_endpoint( EndpointName=endpoint_name, EndpointConfigName=config_name ) except Exception as e: print(e) print('Unable to create endpoint.') raise(e) name = 'name-of-the-endpoint' config = 'name-of-the-endpoint-config' #this one has to exist on the Endpoint configuration list on sagemaker create_endpoint(name, config)
Regards
Edited by: hugoflores on Aug 17, 2021 9:36 AM
Conteúdo relevante
- AWS OFICIALAtualizada há 3 anos
- AWS OFICIALAtualizada há 2 anos
- AWS OFICIALAtualizada há um ano