interface InstanceProductionVariantProps
Language | Type name |
---|---|
.NET | Amazon.CDK.AWS.Sagemaker.Alpha.InstanceProductionVariantProps |
Go | github.com/aws/aws-cdk-go/awscdksagemakeralpha/v2#InstanceProductionVariantProps |
Java | software.amazon.awscdk.services.sagemaker.alpha.InstanceProductionVariantProps |
Python | aws_cdk.aws_sagemaker_alpha.InstanceProductionVariantProps |
TypeScript (source) | @aws-cdk/aws-sagemaker-alpha ยป InstanceProductionVariantProps |
Construction properties for an instance production variant.
Example
// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import * as sagemaker_alpha from '@aws-cdk/aws-sagemaker-alpha';
declare const acceleratorType: sagemaker_alpha.AcceleratorType;
declare const instanceType: sagemaker_alpha.InstanceType;
declare const model: sagemaker_alpha.Model;
const instanceProductionVariantProps: sagemaker_alpha.InstanceProductionVariantProps = {
model: model,
variantName: 'variantName',
// the properties below are optional
acceleratorType: acceleratorType,
initialInstanceCount: 123,
initialVariantWeight: 123,
instanceType: instanceType,
};
Properties
Name | Type | Description |
---|---|---|
model | IModel | The model to host. |
variant | string | Name of the production variant. |
accelerator | Accelerator | The size of the Elastic Inference (EI) instance to use for the production variant. |
initial | number | Number of instances to launch initially. |
initial | number | Determines initial traffic distribution among all of the models that you specify in the endpoint configuration. |
instance | Instance | Instance type of the production variant. |
model
Type:
IModel
The model to host.
variantName
Type:
string
Name of the production variant.
acceleratorType?
Type:
Accelerator
(optional, default: none)
The size of the Elastic Inference (EI) instance to use for the production variant.
EI instances provide on-demand GPU computing for inference.
initialInstanceCount?
Type:
number
(optional, default: 1)
Number of instances to launch initially.
initialVariantWeight?
Type:
number
(optional, default: 1.0)
Determines initial traffic distribution among all of the models that you specify in the endpoint configuration.
The traffic to a production variant is determined by the ratio of the variant weight to the sum of all variant weight values across all production variants.
instanceType?
Type:
Instance
(optional, default: InstanceType.T2_MEDIUM)
Instance type of the production variant.