From a068ee6ef8e690bbcdae678ae2663820226b1bf0 Mon Sep 17 00:00:00 2001 From: awstools Date: Thu, 13 Feb 2025 19:14:38 +0000 Subject: [PATCH] feat(client-sagemaker): Adds additional values to the InferenceAmiVersion parameter in the ProductionVariant data type. --- .../commands/CreateEndpointConfigCommand.ts | 4 +-- .../commands/DescribeEndpointConfigCommand.ts | 4 +-- .../client-sagemaker/src/models/models_0.ts | 15 ++++---- .../client-sagemaker/src/models/models_1.ts | 36 +++++++++++++++++-- codegen/sdk-codegen/aws-models/sagemaker.json | 18 ++++++++-- 5 files changed, 60 insertions(+), 17 deletions(-) diff --git a/clients/client-sagemaker/src/commands/CreateEndpointConfigCommand.ts b/clients/client-sagemaker/src/commands/CreateEndpointConfigCommand.ts index 9cba854a2fc2..5ab5dc7ea6c8 100644 --- a/clients/client-sagemaker/src/commands/CreateEndpointConfigCommand.ts +++ b/clients/client-sagemaker/src/commands/CreateEndpointConfigCommand.ts @@ -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 @@ -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", diff --git a/clients/client-sagemaker/src/commands/DescribeEndpointConfigCommand.ts b/clients/client-sagemaker/src/commands/DescribeEndpointConfigCommand.ts index e27084dba1bc..db86689d1137 100644 --- a/clients/client-sagemaker/src/commands/DescribeEndpointConfigCommand.ts +++ b/clients/client-sagemaker/src/commands/DescribeEndpointConfigCommand.ts @@ -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 @@ -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", diff --git a/clients/client-sagemaker/src/models/models_0.ts b/clients/client-sagemaker/src/models/models_0.ts index d5eb2e2bf5a6..464ab0bde081 100644 --- a/clients/client-sagemaker/src/models/models_0.ts +++ b/clients/client-sagemaker/src/models/models_0.ts @@ -1243,10 +1243,9 @@ export type TrainingInputMode = (typeof TrainingInputMode)[keyof typeof Training *

Specifies the training algorithm to use in a CreateTrainingJob request.

* *

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.

+ * 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.

* *

For more information about algorithms provided by SageMaker, see Algorithms. For * information about using your own algorithms, see Using Your Own Algorithms with @@ -1607,10 +1606,10 @@ export interface S3DataSource { *

If you choose ManifestFile, S3Uri identifies an object that * is a manifest file containing a list of object keys that you want SageMaker to use for model * training.

- *

If you choose AugmentedManifestFile, S3Uri 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. AugmentedManifestFile can only be used if the - * Channel's input mode is Pipe.

+ *

If you choose AugmentedManifestFile, S3Uri 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. AugmentedManifestFile can only be + * used if the Channel's input mode is Pipe.

* @public */ S3DataType: S3DataType | undefined; diff --git a/clients/client-sagemaker/src/models/models_1.ts b/clients/client-sagemaker/src/models/models_1.ts index d31e83b1028c..d530696cdd25 100644 --- a/clients/client-sagemaker/src/models/models_1.ts +++ b/clients/client-sagemaker/src/models/models_1.ts @@ -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; /** @@ -4610,11 +4612,41 @@ export interface ProductionVariant { *

NVIDIA driver version: 535.54.03

* *
  • + *

    CUDA version: 12.2

    + *
    • + * + * + *
    al2-ami-sagemaker-inference-gpu-2-1
    + *
    + *
      + *
    • + *

      Accelerator: GPU

      + *
      • + *
    • + *

      NVIDIA driver version: 535.54.03

      + *
      • + *
    • *

      CUDA driver version: 12.2

      *
      • *
    • - *

      Supported instance types: ml.g4dn.*, ml.g5.*, ml.g6.*, ml.p3.*, - * ml.p4d.*, ml.p4de.*, ml.p5.*

      + *

      CUDA Container Toolkit with disabled CUDA-compat mounting

      + *
      • + *
    + *
    + *
    al2-ami-sagemaker-inference-gpu-3-1
    + *
    + *
      + *
    • + *

      Accelerator: GPU

      + *
      • + *
    • + *

      NVIDIA driver version: 550.144.01

      + *
      • + *
    • + *

      CUDA version: 12.4

      + *
      • + *
    • + *

      Container Toolkit with disabled CUDA-compat mounting

      *
      • *
    *
    diff --git a/codegen/sdk-codegen/aws-models/sagemaker.json b/codegen/sdk-codegen/aws-models/sagemaker.json index f16dd316d111..b2e0efd79e5b 100644 --- a/codegen/sdk-codegen/aws-models/sagemaker.json +++ b/codegen/sdk-codegen/aws-models/sagemaker.json @@ -707,7 +707,7 @@ } }, "traits": { - "smithy.api#documentation": "

    Specifies the training algorithm to use in a CreateTrainingJob request.

    \n \n

    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.

    \n     \n

    For more information about algorithms provided by SageMaker, see Algorithms. For\n information about using your own algorithms, see Using Your Own Algorithms with\n Amazon SageMaker.

    " + "smithy.api#documentation": "

    Specifies the training algorithm to use in a CreateTrainingJob request.

    \n \n

    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.

    \n     \n

    For more information about algorithms provided by SageMaker, see Algorithms. For\n information about using your own algorithms, see Using Your Own Algorithms with\n Amazon SageMaker.

    " } }, "com.amazonaws.sagemaker#AlgorithmStatus": { @@ -55703,7 +55703,7 @@ "InferenceAmiVersion": { "target": "com.amazonaws.sagemaker#ProductionVariantInferenceAmiVersion", "traits": { - "smithy.api#documentation": "

    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.

    \n

    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.

    \n

    The AMI version names, and their configurations, are the following:

    \n
    \n
    al2-ami-sagemaker-inference-gpu-2
    \n
    \n
      \n
    • \n

      Accelerator: GPU

      \n
      • \n
    • \n

      NVIDIA driver version: 535.54.03

      \n
      • \n
    • \n

      CUDA driver version: 12.2

      \n
      • \n
    • \n

      Supported instance types: ml.g4dn.*, ml.g5.*, ml.g6.*, ml.p3.*,\n ml.p4d.*, ml.p4de.*, ml.p5.*

      \n
      • \n
    \n
    \n
    " + "smithy.api#documentation": "

    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.

    \n

    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.

    \n

    The AMI version names, and their configurations, are the following:

    \n
    \n
    al2-ami-sagemaker-inference-gpu-2
    \n
    \n
      \n
    • \n

      Accelerator: GPU

      \n
      • \n
    • \n

      NVIDIA driver version: 535.54.03

      \n
      • \n
    • \n

      CUDA version: 12.2

      \n
      • \n
    \n
    \n
    al2-ami-sagemaker-inference-gpu-2-1
    \n
    \n
      \n
    • \n

      Accelerator: GPU

      \n
      • \n
    • \n

      NVIDIA driver version: 535.54.03

      \n
      • \n
    • \n

      CUDA driver version: 12.2

      \n
      • \n
    • \n

      CUDA Container Toolkit with disabled CUDA-compat mounting

      \n
      • \n
    \n
    \n
    al2-ami-sagemaker-inference-gpu-3-1
    \n
    \n
      \n
    • \n

      Accelerator: GPU

      \n
      • \n
    • \n

      NVIDIA driver version: 550.144.01

      \n
      • \n
    • \n

      CUDA version: 12.4

      \n
      • \n
    • \n

      Container Toolkit with disabled CUDA-compat mounting

      \n
      • \n
    \n
    \n
    " } } }, @@ -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" + } } } }, @@ -60640,7 +60652,7 @@ "target": "com.amazonaws.sagemaker#S3DataType", "traits": { "smithy.api#clientOptional": {}, - "smithy.api#documentation": "

    If you choose S3Prefix, S3Uri identifies a key name prefix.\n SageMaker uses all objects that match the specified key name prefix for model training.

    \n

    If you choose ManifestFile, S3Uri 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.

    \n

    If you choose AugmentedManifestFile, S3Uri 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. AugmentedManifestFile can only be used if the\n Channel's input mode is Pipe.

    ", + "smithy.api#documentation": "

    If you choose S3Prefix, S3Uri identifies a key name prefix.\n SageMaker uses all objects that match the specified key name prefix for model training.

    \n

    If you choose ManifestFile, S3Uri 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.

    \n

    If you choose AugmentedManifestFile, S3Uri 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. AugmentedManifestFile can only be\n used if the Channel's input mode is Pipe.

    ", "smithy.api#required": {} } },