本文属于机器翻译版本。若本译文内容与英语原文存在差异，则一律以英文原文为准。

# 编译模型 (AWS Command Line Interface)
<a name="neo-job-compilation-cli"></a>

本节介绍如何使用 AWS Command Line Interface (CLI) 管理机器学习模型的 Amazon SageMaker Neo 编译任务。您可以创建、描述、停止和列出编译作业。

1. 创建编译作业

   [CreateCompilationJob](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_CreateCompilationJob.html)通过 API 操作，您可以指定数据输入格式、用于存储模型的 S3 存储桶、用于写入已编译模型的 S3 存储桶以及目标硬件设备或平台。

   下表演示了如何基于您的目标是设备还是平台来配置 `CreateCompilationJob` API。

------
#### [ Device Example ]

   ```
   {
       "CompilationJobName": "neo-compilation-job-demo",
       "RoleArn": "arn:aws:iam::<your-account>:role/service-role/AmazonSageMaker-ExecutionRole-yyyymmddThhmmss",
       "InputConfig": {
           "S3Uri": "s3://<your-bucket>/sagemaker/neo-compilation-job-demo-data/train",
           "DataInputConfig":  "{'data': [1,3,1024,1024]}",
           "Framework": "MXNET"
       },
       "OutputConfig": {
           "S3OutputLocation": "s3://<your-bucket>/sagemaker/neo-compilation-job-demo-data/compile",
           # A target device specification example for a ml_c5 instance family
           "TargetDevice": "ml_c5"
       },
       "StoppingCondition": {
           "MaxRuntimeInSeconds": 300
       }
   }
   ```

   如果您使用框架训练模型并且目标设备是目标，则可以选择指定用于该[https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_InputConfig.html#sagemaker-Type-InputConfig-FrameworkVersion](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_InputConfig.html#sagemaker-Type-InputConfig-FrameworkVersion)字段的 PyTorch `ml_* `框架版本。

   ```
   {
       "CompilationJobName": "neo-compilation-job-demo",
       "RoleArn": "arn:aws:iam::<your-account>:role/service-role/AmazonSageMaker-ExecutionRole-yyyymmddThhmmss",
       "InputConfig": {
           "S3Uri": "s3://<your-bucket>/sagemaker/neo-compilation-job-demo-data/train",
           "DataInputConfig":  "{'data': [1,3,1024,1024]}",
           "Framework": "PYTORCH",
           "FrameworkVersion": "1.6"
       },
       "OutputConfig": {
           "S3OutputLocation": "s3://<your-bucket>/sagemaker/neo-compilation-job-demo-data/compile",
           # A target device specification example for a ml_c5 instance family
           "TargetDevice": "ml_c5",
           # When compiling for ml_* instances using PyTorch framework, use the "CompilerOptions" field in 
           # OutputConfig to provide the correct data type ("dtype") of the model’s input. Default assumed is "float32"
           "CompilerOptions": "{'dtype': 'long'}"
       },
       "StoppingCondition": {
           "MaxRuntimeInSeconds": 300
       }
   }
   ```

**备注：**  
如果您使用 PyTorch 版本 2.0 或更高版本保存模型，则该`DataInputConfig`字段为可选字段。 SageMaker AI Neo 从您创建的模型定义文件中获取输入配置 PyTorch。有关如何创建定义文件的更多信息，请参阅为 * SageMaker AI Neo 保存模型*下的[PyTorch](neo-compilation-preparing-model.md#how-to-save-pytorch)部分。
仅支持此 API 字段 PyTorch。

------
#### [ Platform Example ]

   ```
   {
       "CompilationJobName": "neo-test-compilation-job",
       "RoleArn": "arn:aws:iam::<your-account>:role/service-role/AmazonSageMaker-ExecutionRole-yyyymmddThhmmss",
       "InputConfig": {
           "S3Uri": "s3://<your-bucket>/sagemaker/neo-compilation-job-demo-data/train",
           "DataInputConfig":  "{'data': [1,3,1024,1024]}",
           "Framework": "MXNET"
       },
       "OutputConfig": {
           "S3OutputLocation": "s3://<your-bucket>/sagemaker/neo-compilation-job-demo-data/compile",
           # A target platform configuration example for a p3.2xlarge instance
           "TargetPlatform": {
               "Os": "LINUX",
               "Arch": "X86_64",
               "Accelerator": "NVIDIA"
           },
           "CompilerOptions": "{'cuda-ver': '10.0', 'trt-ver': '6.0.1', 'gpu-code': 'sm_70'}"
       },
       "StoppingCondition": {
           "MaxRuntimeInSeconds": 300
       }
   }
   ```

------
**注意**  
对于 `OutputConfig` API 操作，`TargetDevice` 和 `TargetPlatform` API 操作是互相排斥的。您必须从两个选项中选择一个。

   要根据框架查找 `DataInputConfig` 的 JSON 字符串示例，请参阅 [Neo 期望的输入数据形状](https://docs.aws.amazon.com/sagemaker/latest/dg/neo-troubleshooting-compilation.html#neo-troubleshooting-errors-preventing)。

   有关设置配置的更多信息，请参阅 API 参考中的[InputConfig[OutputConfig](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_OutputConfig.html)](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_InputConfig.html)、和 [TargetPlatform](https://docs.aws.amazon.com/sagemaker/latest/APIReference/API_TargetPlatform.html) SageMaker API 操作。

1. 配置 JSON 文件后，运行以下命令来创建编译作业：

   ```
   aws sagemaker create-compilation-job \
   --cli-input-json file://job.json \
   --region us-west-2 
   
   # You should get CompilationJobArn
   ```

1. 通过运行以下命令描述编译作业：

   ```
   aws sagemaker describe-compilation-job \
   --compilation-job-name $JOB_NM \
   --region us-west-2
   ```

1. 通过运行以下命令停止编译作业：

   ```
   aws sagemaker stop-compilation-job \
   --compilation-job-name $JOB_NM \
   --region us-west-2
   
   # There is no output for compilation-job operation
   ```

1. 通过运行以下命令列出编译作业：

   ```
   aws sagemaker list-compilation-jobs \
   --region us-west-2
   ```