问题
I am trying to run my own algorithm container in amazon sagemaker,at the time of deployment time ,I am getting error like below.
predictor = tree.deploy(1, 'ml.m4.xlarge', serializer=csv_serializer)
ValueError: Error hosting endpoint decision-trees-sample-2018-03-01-09-59-06-832: Failed Reason: The primary container for production variant AllTraffic did not pass the ping health check.
then I run same line of code this time i am getting below error.
predictor = tree.deploy(1, 'ml.m4.xlarge', serializer=csv_serializer)
ClientError: An error occurred (ValidationException) when calling the CreateEndpoint operation: Cannot create already existing endpoint "arn:aws:sagemaker:us-east-1:69759707XXxXX:endpoint/decision-trees-sample-2018-03-01-09-59-06-832".
回答1:
Check out this issue: https://github.com/awslabs/amazon-sagemaker-examples/issues/210
@djarpin wrote:
The ping health check message is a general error that can be caused by several different issues. Typically the error message in the CloudWatch log group named /aws/sagemaker/Endpoints/ will provide a more detailed description of why the ping health check didn't pass.
Hope that helps!
来源：https://stackoverflow.com/questions/49047576/amazon-sagemake-throwing-error-building-your-own-algorithm-container-execution-t