Amazon SageMaker Service Update: This release adds support for cross account access for SageMaker Model Cards through AWS RAM.

AWS · AWS · commit 79807829f615 · 2023-08-09T18:09:47.000Z
diff --git a/.changes/next-release/feature-AmazonSageMakerService-eb94af5.json b/.changes/next-release/feature-AmazonSageMakerService-eb94af5.json
@@ -0,0 +1,6 @@
+{
+    "type": "feature",
+    "category": "Amazon SageMaker Service",
+    "contributor": "",
+    "description": "This release adds support for cross account access for SageMaker Model Cards through AWS RAM."
+}
diff --git a/services/sagemaker/src/main/resources/codegen-resources/service-2.json b/services/sagemaker/src/main/resources/codegen-resources/service-2.json
@@ -8439,8 +8439,8 @@
       ],
       "members":{
         "ModelCardName":{
-          "shape":"EntityName",
-          "documentation":"<p>The name of the model card to export.</p>"
+          "shape":"ModelCardNameOrArn",
+          "documentation":"<p>The name or Amazon Resource Name (ARN) of the model card to export.</p>"
         },
         "ModelCardVersion":{
           "shape":"Integer",
@@ -13236,7 +13236,7 @@
         },
         "ModelCardName":{
           "shape":"EntityName",
-          "documentation":"<p>The name of the model card that the model export job exports.</p>"
+          "documentation":"<p>The name or Amazon Resource Name (ARN) of the model card that the model export job exports.</p>"
         },
         "ModelCardVersion":{
           "shape":"Integer",
@@ -13269,8 +13269,8 @@
       "required":["ModelCardName"],
       "members":{
         "ModelCardName":{
-          "shape":"EntityName",
-          "documentation":"<p>The name of the model card to describe.</p>"
+          "shape":"ModelCardNameOrArn",
+          "documentation":"<p>The name or Amazon Resource Name (ARN) of the model card to describe.</p>"
         },
         "ModelCardVersion":{
           "shape":"Integer",
@@ -22020,8 +22020,8 @@
           "documentation":"<p>The maximum number of model card versions to list.</p>"
         },
         "ModelCardName":{
-          "shape":"EntityName",
-          "documentation":"<p>List model card versions for the model card with the specified name.</p>"
+          "shape":"ModelCardNameOrArn",
+          "documentation":"<p>List model card versions for the model card with the specified name or Amazon Resource Name (ARN).</p>"
         },
         "ModelCardStatus":{
           "shape":"ModelCardStatus",
@@ -24252,6 +24252,12 @@
       },
       "documentation":"<p>Configure the export output details for an Amazon SageMaker Model Card.</p>"
     },
+    "ModelCardNameOrArn":{
+      "type":"string",
+      "max":256,
+      "min":1,
+      "pattern":"(arn:aws[a-z\\-]*:sagemaker:[a-z0-9\\-]*:[0-9]{12}:model-card/.*)?([a-zA-Z0-9](-*[a-zA-Z0-9]){0,62})"
+    },
     "ModelCardProcessingStatus":{
       "type":"string",
       "enum":[
@@ -26842,7 +26848,7 @@
         },
         "CompilerOptions":{
           "shape":"CompilerOptions",
-          "documentation":"<p>Specifies additional parameters for compiler options in JSON format. The compiler options are <code>TargetPlatform</code> specific. It is required for NVIDIA accelerators and highly recommended for CPU compilations. For any other cases, it is optional to specify <code>CompilerOptions.</code> </p> <ul> <li> <p> <code>DTYPE</code>: Specifies the data type for the input. When compiling for <code>ml_*</code> (except for <code>ml_inf</code>) instances using PyTorch framework, provide the data type (dtype) of the model's input. <code>\"float32\"</code> is used if <code>\"DTYPE\"</code> is not specified. Options for data type are:</p> <ul> <li> <p>float32: Use either <code>\"float\"</code> or <code>\"float32\"</code>.</p> </li> <li> <p>int64: Use either <code>\"int64\"</code> or <code>\"long\"</code>.</p> </li> </ul> <p> For example, <code>{\"dtype\" : \"float32\"}</code>.</p> </li> <li> <p> <code>CPU</code>: Compilation for CPU supports the following compiler options.</p> <ul> <li> <p> <code>mcpu</code>: CPU micro-architecture. For example, <code>{'mcpu': 'skylake-avx512'}</code> </p> </li> <li> <p> <code>mattr</code>: CPU flags. For example, <code>{'mattr': ['+neon', '+vfpv4']}</code> </p> </li> </ul> </li> <li> <p> <code>ARM</code>: Details of ARM CPU compilations.</p> <ul> <li> <p> <code>NEON</code>: NEON is an implementation of the Advanced SIMD extension used in ARMv7 processors.</p> <p>For example, add <code>{'mattr': ['+neon']}</code> to the compiler options if compiling for ARM 32-bit platform with the NEON support.</p> </li> </ul> </li> <li> <p> <code>NVIDIA</code>: Compilation for NVIDIA GPU supports the following compiler options.</p> <ul> <li> <p> <code>gpu_code</code>: Specifies the targeted architecture.</p> </li> <li> <p> <code>trt-ver</code>: Specifies the TensorRT versions in x.y.z. format.</p> </li> <li> <p> <code>cuda-ver</code>: Specifies the CUDA version in x.y format.</p> </li> </ul> <p>For example, <code>{'gpu-code': 'sm_72', 'trt-ver': '6.0.1', 'cuda-ver': '10.1'}</code> </p> </li> <li> <p> <code>ANDROID</code>: Compilation for the Android OS supports the following compiler options:</p> <ul> <li> <p> <code>ANDROID_PLATFORM</code>: Specifies the Android API levels. Available levels range from 21 to 29. For example, <code>{'ANDROID_PLATFORM': 28}</code>.</p> </li> <li> <p> <code>mattr</code>: Add <code>{'mattr': ['+neon']}</code> to compiler options if compiling for ARM 32-bit platform with NEON support.</p> </li> </ul> </li> <li> <p> <code>INFERENTIA</code>: Compilation for target ml_inf1 uses compiler options passed in as a JSON string. For example, <code>\"CompilerOptions\": \"\\\"--verbose 1 --num-neuroncores 2 -O2\\\"\"</code>. </p> <p>For information about supported compiler options, see <a href=\"https://github.com/aws/aws-neuron-sdk/blob/master/docs/neuron-cc/command-line-reference.md\"> Neuron Compiler CLI</a>. </p> </li> <li> <p> <code>CoreML</code>: Compilation for the CoreML <a href=\"https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_OutputConfig.html\">OutputConfig</a> <code>TargetDevice</code> supports the following compiler options:</p> <ul> <li> <p> <code>class_labels</code>: Specifies the classification labels file name inside input tar.gz file. For example, <code>{\"class_labels\": \"imagenet_labels_1000.txt\"}</code>. Labels inside the txt file should be separated by newlines.</p> </li> </ul> </li> <li> <p> <code>EIA</code>: Compilation for the Elastic Inference Accelerator supports the following compiler options:</p> <ul> <li> <p> <code>precision_mode</code>: Specifies the precision of compiled artifacts. Supported values are <code>\"FP16\"</code> and <code>\"FP32\"</code>. Default is <code>\"FP32\"</code>.</p> </li> <li> <p> <code>signature_def_key</code>: Specifies the signature to use for models in SavedModel format. Defaults is TensorFlow's default signature def key.</p> </li> <li> <p> <code>output_names</code>: Specifies a list of output tensor names for models in FrozenGraph format. Set at most one API field, either: <code>signature_def_key</code> or <code>output_names</code>.</p> </li> </ul> <p>For example: <code>{\"precision_mode\": \"FP32\", \"output_names\": [\"output:0\"]}</code> </p> </li> </ul>"
+          "documentation":"<p>Specifies additional parameters for compiler options in JSON format. The compiler options are <code>TargetPlatform</code> specific. It is required for NVIDIA accelerators and highly recommended for CPU compilations. For any other cases, it is optional to specify <code>CompilerOptions.</code> </p> <ul> <li> <p> <code>DTYPE</code>: Specifies the data type for the input. When compiling for <code>ml_*</code> (except for <code>ml_inf</code>) instances using PyTorch framework, provide the data type (dtype) of the model's input. <code>\"float32\"</code> is used if <code>\"DTYPE\"</code> is not specified. Options for data type are:</p> <ul> <li> <p>float32: Use either <code>\"float\"</code> or <code>\"float32\"</code>.</p> </li> <li> <p>int64: Use either <code>\"int64\"</code> or <code>\"long\"</code>.</p> </li> </ul> <p> For example, <code>{\"dtype\" : \"float32\"}</code>.</p> </li> <li> <p> <code>CPU</code>: Compilation for CPU supports the following compiler options.</p> <ul> <li> <p> <code>mcpu</code>: CPU micro-architecture. For example, <code>{'mcpu': 'skylake-avx512'}</code> </p> </li> <li> <p> <code>mattr</code>: CPU flags. For example, <code>{'mattr': ['+neon', '+vfpv4']}</code> </p> </li> </ul> </li> <li> <p> <code>ARM</code>: Details of ARM CPU compilations.</p> <ul> <li> <p> <code>NEON</code>: NEON is an implementation of the Advanced SIMD extension used in ARMv7 processors.</p> <p>For example, add <code>{'mattr': ['+neon']}</code> to the compiler options if compiling for ARM 32-bit platform with the NEON support.</p> </li> </ul> </li> <li> <p> <code>NVIDIA</code>: Compilation for NVIDIA GPU supports the following compiler options.</p> <ul> <li> <p> <code>gpu_code</code>: Specifies the targeted architecture.</p> </li> <li> <p> <code>trt-ver</code>: Specifies the TensorRT versions in x.y.z. format.</p> </li> <li> <p> <code>cuda-ver</code>: Specifies the CUDA version in x.y format.</p> </li> </ul> <p>For example, <code>{'gpu-code': 'sm_72', 'trt-ver': '6.0.1', 'cuda-ver': '10.1'}</code> </p> </li> <li> <p> <code>ANDROID</code>: Compilation for the Android OS supports the following compiler options:</p> <ul> <li> <p> <code>ANDROID_PLATFORM</code>: Specifies the Android API levels. Available levels range from 21 to 29. For example, <code>{'ANDROID_PLATFORM': 28}</code>.</p> </li> <li> <p> <code>mattr</code>: Add <code>{'mattr': ['+neon']}</code> to compiler options if compiling for ARM 32-bit platform with NEON support.</p> </li> </ul> </li> <li> <p> <code>INFERENTIA</code>: Compilation for target ml_inf1 uses compiler options passed in as a JSON string. For example, <code>\"CompilerOptions\": \"\\\"--verbose 1 --num-neuroncores 2 -O2\\\"\"</code>. </p> <p>For information about supported compiler options, see <a href=\"https://awsdocs-neuron.readthedocs-hosted.com/en/latest/compiler/neuronx-cc/api-reference-guide/neuron-compiler-cli-reference-guide.html\"> Neuron Compiler CLI Reference Guide</a>. </p> </li> <li> <p> <code>CoreML</code>: Compilation for the CoreML <a href=\"https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_OutputConfig.html\">OutputConfig</a> <code>TargetDevice</code> supports the following compiler options:</p> <ul> <li> <p> <code>class_labels</code>: Specifies the classification labels file name inside input tar.gz file. For example, <code>{\"class_labels\": \"imagenet_labels_1000.txt\"}</code>. Labels inside the txt file should be separated by newlines.</p> </li> </ul> </li> <li> <p> <code>EIA</code>: Compilation for the Elastic Inference Accelerator supports the following compiler options:</p> <ul> <li> <p> <code>precision_mode</code>: Specifies the precision of compiled artifacts. Supported values are <code>\"FP16\"</code> and <code>\"FP32\"</code>. Default is <code>\"FP32\"</code>.</p> </li> <li> <p> <code>signature_def_key</code>: Specifies the signature to use for models in SavedModel format. Defaults is TensorFlow's default signature def key.</p> </li> <li> <p> <code>output_names</code>: Specifies a list of output tensor names for models in FrozenGraph format. Set at most one API field, either: <code>signature_def_key</code> or <code>output_names</code>.</p> </li> </ul> <p>For example: <code>{\"precision_mode\": \"FP32\", \"output_names\": [\"output:0\"]}</code> </p> </li> </ul>"
         },
         "KmsKeyId":{
           "shape":"KmsKeyId",
@@ -33891,8 +33897,8 @@
       "required":["ModelCardName"],
       "members":{
         "ModelCardName":{
-          "shape":"EntityName",
-          "documentation":"<p>The name of the model card to update.</p>"
+          "shape":"ModelCardNameOrArn",
+          "documentation":"<p>The name or Amazon Resource Name (ARN) of the model card to update.</p>"
         },
         "Content":{
           "shape":"ModelCardContent",
