Parameter validation failed - "MemorySizeInMb", must be one of: MemorySizeInMB, MaxConcurrency, ProvisionedConcurrency

[robmcd]

Describe the bug
getting an error when trying to create model endpoint config.

To reproduce

from sagemaker_core.resources import EndpointConfig, Endpoint
from sagemaker_core.main.shapes import ProductionVariant, ProductionVariantServerlessConfig

endpoint_config = EndpointConfig.create(
    endpoint_config_name='endpoint-config',
    production_variants=[
        ProductionVariant(
            variant_name='variantA', 
            model_name='MyModel',
            serverless_config=ProductionVariantServerlessConfig(
                memory_size_in_mb=1024,
                max_concurrency=2
            )
        )
    ]
)

Getting error:

ParamValidationError: Parameter validation failed:
Missing required parameter in ProductionVariants[0].ServerlessConfig: "MemorySizeInMB"
Unknown parameter in ProductionVariants[0].ServerlessConfig: "MemorySizeInMb", must be one of: MemorySizeInMB, 
MaxConcurrency, ProvisionedConcurrency

Expected behavior
No error raised.

Screenshots or logs
If applicable, add screenshots or logs to help explain your problem.

Bug information
A description of your system. Please provide:

• SageMaker Core version: 1.0.16
• Python version: 3.11

Additional context
I'm guessing that there's some case sensitive issue happening here...?

[robmcd]

Similar error:
Unknown parameter in InferenceSpecification: "SupportedResponseMimeTypes", must be one of: Containers, SupportedTransformInstanceTypes, SupportedRealtimeInferenceInstanceTypes, SupportedContentTypes, SupportedResponseMIMETypes

[wm3]

I also have the same issue. I can't create any serverless inference endpoints using sagemaker-core because of this.

[MarcusKoolaard]

Also having this issue.

Potential PR:
Add
"memory_size_in_mb": "MemorySizeInMB", "supported_response_mime_types": "SupportedResponseMIMETypes",
to SPECIAL_SNAKE_TO_PASCAL_MAPPINGS in
src/sagemaker_core/main/utils.py.

@robmcd is this something I can contribute to?

[robmcd]

Also having this issue.

Potential PR: Add "memory_size_in_mb": "MemorySizeInMB", "supported_response_mime_types": "SupportedResponseMIMETypes", to SPECIAL_SNAKE_TO_PASCAL_MAPPINGS in src/sagemaker_core/main/utils.py.

@robmcd is this something I can contribute to?

@MarcusKoolaard , I'm just a messenger here and not afiliated with AWS. Whilst I dont have tim eto contribute/review I would very much appreciate a fix so am very much in support of your contribution!