Skip to content

Commit

Permalink
feat(client-sagemaker): Adds additional values to the InferenceAmiVer…
Browse files Browse the repository at this point in the history 
…sion parameter in the ProductionVariant data type.
  • Loading branch information
awstools committed Feb 13, 2025
1 parent aeef69f commit a068ee6
Show file tree
Hide file tree
Showing 5 changed files with 60 additions and 17 deletions.
4 changes: 2 additions & 2 deletions clients/client-sagemaker/src/commands/CreateEndpointConfigCommand.ts
View file
Original file line number Diff line number Diff line change
Expand Up @@ -96,7 +96,7 @@ export interface CreateEndpointConfigCommandOutput extends CreateEndpointConfigO
* RoutingConfig: { // ProductionVariantRoutingConfig
* RoutingStrategy: "LEAST_OUTSTANDING_REQUESTS" || "RANDOM", // required
* },
* InferenceAmiVersion: "al2-ami-sagemaker-inference-gpu-2",
* InferenceAmiVersion: "al2-ami-sagemaker-inference-gpu-2" || "al2-ami-sagemaker-inference-gpu-2-1" || "al2-ami-sagemaker-inference-gpu-3-1",
* },
* ],
* DataCaptureConfig: { // DataCaptureConfig
Expand Down Expand Up @@ -209,7 +209,7 @@ export interface CreateEndpointConfigCommandOutput extends CreateEndpointConfigO
* RoutingConfig: {
* RoutingStrategy: "LEAST_OUTSTANDING_REQUESTS" || "RANDOM", // required
* },
* InferenceAmiVersion: "al2-ami-sagemaker-inference-gpu-2",
* InferenceAmiVersion: "al2-ami-sagemaker-inference-gpu-2" || "al2-ami-sagemaker-inference-gpu-2-1" || "al2-ami-sagemaker-inference-gpu-3-1",
* },
* ],
* ExecutionRoleArn: "STRING_VALUE",
Expand Down
4 changes: 2 additions & 2 deletions clients/client-sagemaker/src/commands/DescribeEndpointConfigCommand.ts
View file
Original file line number Diff line number Diff line change
Expand Up @@ -73,7 +73,7 @@ export interface DescribeEndpointConfigCommandOutput extends DescribeEndpointCon
* // RoutingConfig: { // ProductionVariantRoutingConfig
* // RoutingStrategy: "LEAST_OUTSTANDING_REQUESTS" || "RANDOM", // required
* // },
* // InferenceAmiVersion: "al2-ami-sagemaker-inference-gpu-2",
* // InferenceAmiVersion: "al2-ami-sagemaker-inference-gpu-2" || "al2-ami-sagemaker-inference-gpu-2-1" || "al2-ami-sagemaker-inference-gpu-3-1",
* // },
* // ],
* // DataCaptureConfig: { // DataCaptureConfig
Expand Down Expand Up @@ -181,7 +181,7 @@ export interface DescribeEndpointConfigCommandOutput extends DescribeEndpointCon
* // RoutingConfig: {
* // RoutingStrategy: "LEAST_OUTSTANDING_REQUESTS" || "RANDOM", // required
* // },
* // InferenceAmiVersion: "al2-ami-sagemaker-inference-gpu-2",
* // InferenceAmiVersion: "al2-ami-sagemaker-inference-gpu-2" || "al2-ami-sagemaker-inference-gpu-2-1" || "al2-ami-sagemaker-inference-gpu-3-1",
* // },
* // ],
* // ExecutionRoleArn: "STRING_VALUE",
Expand Down
15 changes: 7 additions & 8 deletions clients/client-sagemaker/src/models/models_0.ts
View file
Original file line number Diff line number Diff line change
Expand Up @@ -1243,10 +1243,9 @@ export type TrainingInputMode = (typeof TrainingInputMode)[keyof typeof Training
* <p>Specifies the training algorithm to use in a <a href="https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingJob.html">CreateTrainingJob</a> request.</p>
* <important>
* <p>SageMaker uses its own SageMaker account credentials to pull and access built-in algorithms
* so built-in algorithms are universally accessible across all Amazon Web Services accounts. As a
* result, built-in algorithms have standard, unrestricted access. You cannot restrict
* built-in algorithms using IAM roles. Use custom algorithms if you require specific
* access controls.</p>
* so built-in algorithms are universally accessible across all Amazon Web Services accounts. As a result, built-in algorithms have standard,
* unrestricted access. You cannot restrict built-in algorithms using IAM roles. Use
* custom algorithms if you require specific access controls.</p>
* </important>
* <p>For more information about algorithms provided by SageMaker, see <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html">Algorithms</a>. For
* information about using your own algorithms, see <a href="https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms.html">Using Your Own Algorithms with
Expand Down Expand Up @@ -1607,10 +1606,10 @@ export interface S3DataSource {
* <p>If you choose <code>ManifestFile</code>, <code>S3Uri</code> identifies an object that
* is a manifest file containing a list of object keys that you want SageMaker to use for model
* training. </p>
* <p>If you choose <code>AugmentedManifestFile</code>, <code>S3Uri</code> identifies an object that is
* an augmented manifest file in JSON lines format. This file contains the data you want to
* use for model training. <code>AugmentedManifestFile</code> can only be used if the
* Channel's input mode is <code>Pipe</code>.</p>
* <p>If you choose <code>AugmentedManifestFile</code>, <code>S3Uri</code> identifies an
* object that is an augmented manifest file in JSON lines format. This file contains the
* data you want to use for model training. <code>AugmentedManifestFile</code> can only be
* used if the Channel's input mode is <code>Pipe</code>.</p>
* @public
*/
S3DataType: S3DataType | undefined;
Expand Down
36 changes: 34 additions & 2 deletions clients/client-sagemaker/src/models/models_1.ts
View file
Original file line number Diff line number Diff line change
Expand Up @@ -4358,6 +4358,8 @@ export interface ProductionVariantCoreDumpConfig {
*/
export const ProductionVariantInferenceAmiVersion = {
AL2_GPU_2: "al2-ami-sagemaker-inference-gpu-2",
AL2_GPU_2_1: "al2-ami-sagemaker-inference-gpu-2-1",
AL2_GPU_3_1: "al2-ami-sagemaker-inference-gpu-3-1",
} as const;

/**
Expand Down Expand Up @@ -4610,11 +4612,41 @@ export interface ProductionVariant {
* <p>NVIDIA driver version: 535.54.03</p>
* </li>
* <li>
* <p>CUDA version: 12.2</p>
* </li>
* </ul>
* </dd>
* <dt>al2-ami-sagemaker-inference-gpu-2-1</dt>
* <dd>
* <ul>
* <li>
* <p>Accelerator: GPU</p>
* </li>
* <li>
* <p>NVIDIA driver version: 535.54.03</p>
* </li>
* <li>
* <p>CUDA driver version: 12.2</p>
* </li>
* <li>
* <p>Supported instance types: ml.g4dn.*, ml.g5.*, ml.g6.*, ml.p3.*,
* ml.p4d.*, ml.p4de.*, ml.p5.*</p>
* <p>CUDA Container Toolkit with disabled CUDA-compat mounting</p>
* </li>
* </ul>
* </dd>
* <dt>al2-ami-sagemaker-inference-gpu-3-1</dt>
* <dd>
* <ul>
* <li>
* <p>Accelerator: GPU</p>
* </li>
* <li>
* <p>NVIDIA driver version: 550.144.01</p>
* </li>
* <li>
* <p>CUDA version: 12.4</p>
* </li>
* <li>
* <p>Container Toolkit with disabled CUDA-compat mounting</p>
* </li>
* </ul>
* </dd>
Expand Down
18 changes: 15 additions & 3 deletions codegen/sdk-codegen/aws-models/sagemaker.json
View file
Original file line number Diff line number Diff line change
Expand Up @@ -707,7 +707,7 @@
}
},
"traits": {
"smithy.api#documentation": "<p>Specifies the training algorithm to use in a <a href=\"https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingJob.html\">CreateTrainingJob</a> request.</p>\n <important>\n <p>SageMaker uses its own SageMaker account credentials to pull and access built-in algorithms\n so built-in algorithms are universally accessible across all Amazon Web Services accounts. As a\n result, built-in algorithms have standard, unrestricted access. You cannot restrict\n built-in algorithms using IAM roles. Use custom algorithms if you require specific\n access controls.</p>\n </important>\n <p>For more information about algorithms provided by SageMaker, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html\">Algorithms</a>. For\n information about using your own algorithms, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms.html\">Using Your Own Algorithms with\n Amazon SageMaker</a>. </p>"
"smithy.api#documentation": "<p>Specifies the training algorithm to use in a <a href=\"https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateTrainingJob.html\">CreateTrainingJob</a> request.</p>\n <important>\n <p>SageMaker uses its own SageMaker account credentials to pull and access built-in algorithms\n so built-in algorithms are universally accessible across all Amazon Web Services accounts. As a result, built-in algorithms have standard,\n unrestricted access. You cannot restrict built-in algorithms using IAM roles. Use\n custom algorithms if you require specific access controls.</p>\n </important>\n <p>For more information about algorithms provided by SageMaker, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/algos.html\">Algorithms</a>. For\n information about using your own algorithms, see <a href=\"https://docs.aws.amazon.com/sagemaker/latest/dg/your-algorithms.html\">Using Your Own Algorithms with\n Amazon SageMaker</a>. </p>"
}
},
"com.amazonaws.sagemaker#AlgorithmStatus": {
Expand Down Expand Up @@ -55703,7 +55703,7 @@
"InferenceAmiVersion": {
"target": "com.amazonaws.sagemaker#ProductionVariantInferenceAmiVersion",
"traits": {
"smithy.api#documentation": "<p>Specifies an option from a collection of preconfigured Amazon Machine Image (AMI)\n images. Each image is configured by Amazon Web Services with a set of software and driver\n versions. Amazon Web Services optimizes these configurations for different machine\n learning workloads.</p>\n <p>By selecting an AMI version, you can ensure that your inference environment is\n compatible with specific software requirements, such as CUDA driver versions, Linux\n kernel versions, or Amazon Web Services Neuron driver versions.</p>\n <p>The AMI version names, and their configurations, are the following:</p>\n <dl>\n <dt>al2-ami-sagemaker-inference-gpu-2</dt>\n <dd>\n <ul>\n <li>\n <p>Accelerator: GPU</p>\n </li>\n <li>\n <p>NVIDIA driver version: 535.54.03</p>\n </li>\n <li>\n <p>CUDA driver version: 12.2</p>\n </li>\n <li>\n <p>Supported instance types: ml.g4dn.*, ml.g5.*, ml.g6.*, ml.p3.*,\n ml.p4d.*, ml.p4de.*, ml.p5.*</p>\n </li>\n </ul>\n </dd>\n </dl>"
"smithy.api#documentation": "<p>Specifies an option from a collection of preconfigured Amazon Machine Image (AMI)\n images. Each image is configured by Amazon Web Services with a set of software and driver\n versions. Amazon Web Services optimizes these configurations for different machine\n learning workloads.</p>\n <p>By selecting an AMI version, you can ensure that your inference environment is\n compatible with specific software requirements, such as CUDA driver versions, Linux\n kernel versions, or Amazon Web Services Neuron driver versions.</p>\n <p>The AMI version names, and their configurations, are the following:</p>\n <dl>\n <dt>al2-ami-sagemaker-inference-gpu-2</dt>\n <dd>\n <ul>\n <li>\n <p>Accelerator: GPU</p>\n </li>\n <li>\n <p>NVIDIA driver version: 535.54.03</p>\n </li>\n <li>\n <p>CUDA version: 12.2</p>\n </li>\n </ul>\n </dd>\n <dt>al2-ami-sagemaker-inference-gpu-2-1</dt>\n <dd>\n <ul>\n <li>\n <p>Accelerator: GPU</p>\n </li>\n <li>\n <p>NVIDIA driver version: 535.54.03</p>\n </li>\n <li>\n <p>CUDA driver version: 12.2</p>\n </li>\n <li>\n <p>CUDA Container Toolkit with disabled CUDA-compat mounting</p>\n </li>\n </ul>\n </dd>\n <dt>al2-ami-sagemaker-inference-gpu-3-1</dt>\n <dd>\n <ul>\n <li>\n <p>Accelerator: GPU</p>\n </li>\n <li>\n <p>NVIDIA driver version: 550.144.01</p>\n </li>\n <li>\n <p>CUDA version: 12.4</p>\n </li>\n <li>\n <p>Container Toolkit with disabled CUDA-compat mounting</p>\n </li>\n </ul>\n </dd>\n </dl>"
}
}
},
Expand Down Expand Up @@ -55791,6 +55791,18 @@
"traits": {
"smithy.api#enumValue": "al2-ami-sagemaker-inference-gpu-2"
}
},
"AL2_GPU_2_1": {
"target": "smithy.api#Unit",
"traits": {
"smithy.api#enumValue": "al2-ami-sagemaker-inference-gpu-2-1"
}
},
"AL2_GPU_3_1": {
"target": "smithy.api#Unit",
"traits": {
"smithy.api#enumValue": "al2-ami-sagemaker-inference-gpu-3-1"
}
}
}
},
Expand Down Expand Up @@ -60640,7 +60652,7 @@
"target": "com.amazonaws.sagemaker#S3DataType",
"traits": {
"smithy.api#clientOptional": {},
"smithy.api#documentation": "<p>If you choose <code>S3Prefix</code>, <code>S3Uri</code> identifies a key name prefix.\n SageMaker uses all objects that match the specified key name prefix for model training. </p>\n <p>If you choose <code>ManifestFile</code>, <code>S3Uri</code> identifies an object that\n is a manifest file containing a list of object keys that you want SageMaker to use for model\n training. </p>\n <p>If you choose <code>AugmentedManifestFile</code>, <code>S3Uri</code> identifies an object that is\n an augmented manifest file in JSON lines format. This file contains the data you want to\n use for model training. <code>AugmentedManifestFile</code> can only be used if the\n Channel's input mode is <code>Pipe</code>.</p>",
"smithy.api#documentation": "<p>If you choose <code>S3Prefix</code>, <code>S3Uri</code> identifies a key name prefix.\n SageMaker uses all objects that match the specified key name prefix for model training. </p>\n <p>If you choose <code>ManifestFile</code>, <code>S3Uri</code> identifies an object that\n is a manifest file containing a list of object keys that you want SageMaker to use for model\n training. </p>\n <p>If you choose <code>AugmentedManifestFile</code>, <code>S3Uri</code> identifies an\n object that is an augmented manifest file in JSON lines format. This file contains the\n data you want to use for model training. <code>AugmentedManifestFile</code> can only be\n used if the Channel's input mode is <code>Pipe</code>.</p>",
"smithy.api#required": {}
}
},
Expand Down

0 comments on commit a068ee6

Please sign in to comment.