训练 AutoML 分类或回归模型

本页面介绍了如何使用 Cloud 控制台或 Vertex AI API 根据表格数据集训练 AutoML 分类或回归模型。

表格 AutoML 分类或回归模型的优化目标

当您使用表格数据集训练 AutoML 模型时,Vertex AI 会根据模型类型和用于目标列的数据类型选择默认优化目标。

分类模型最适合的情况:

优化目标 API 值 在什么情况下使用该目标
AUC ROC maximize-au-roc 最大化接收者操作特征 (ROC) 曲线下的面积。区分不同的类别。二元分类的默认值。
对数损失 minimize-log-loss 使预测概率尽可能准确。仅限于支持的多类别分类目标。
AUC PR maximize-au-prc 最大化精确率/召回率曲线下的面积。优化不常见类别的预测结果。
特定召回率下的精确率 maximize-precision-at-recall 优化特定召回值下的精确率。
特定精确率下的召回率 maximize-recall-at-precision 优化特定精确率下的召回率。

回归模型最适合的情况:

优化目标 API 值 在什么情况下使用该目标
均方根误差 minimize-rmse 最大限度降低均方根误差 (RMSE)。准确捕捉更多极值。默认值。
MAE minimize-mae 最大限度降低平均绝对误差 (MAE)。将极值视为对模型影响较小的离群值。
RMSLE minimize-rmsle 最大限度降低均方根对数误差 (RMSLE)。根据相对误差而不是绝对误差来判错。适用于预测值和实际值都非常大的情况。

准备工作

在训练模型之前,您必须先完成以下操作:

训练 AutoML 模型

Cloud Console

  1. 在 Google Cloud 控制台的 Vertex AI 部分中,转到数据集页面。

    转到“数据集”页面

  2. 点击要用于训练模型的数据集的名称,以打开其详情页面。

  3. 如果您的数据类型使用注释集,请选择要用于此模型的注释集。

  4. 点击训练新模型

  5. 训练新模型窗口中,完成以下步骤

    1. 选择 AutoML 作为训练方法,然后点击继续

    2. 输入新模型的显示名。

    3. 选择目标列。

      目标列是模型将预测的值。

      详细了解目标列要求

    4. 如果您要将测试数据集导出到 BigQuery,请勾选将测试数据集导出到 BigQuery (Export test dataset to BigQuery) 并提供表的名称。

    5. 如果您想手动控制数据拆分,请打开高级选项

      默认数据拆分是随机的。根据您的数据,您可以选择手动来使用数据拆分列、控制数据拆分的百分比或提供时间列。详细了解数据拆分

    6. 点击继续

    7. 点击生成统计信息(如果您尚未生成)。

      生成统计信息会填充转换下拉菜单。

    8. 在“训练选项”页面上,查看列列表,并从训练中排除任何不应用于训练模型的列。

      如果您要使用数据拆分列,则应包含该列。

    9. 查看为包含的特征选择的转换,以及是否允许无效数据,并进行任何所需更新。

      详细了解转换无效数据

    10. 如果要指定权重列,或更改默认的优化目标,请打开高级选项并进行选择。

      详细了解权重列优化目标

    11. 点击继续

    12. 计算和价格窗口中,输入模型训练小时数的上限。

      此设置有助于限制训练费用。实际所用的时间可能超过此值,因为创建新模型涉及其他操作。

      建议的训练时间与训练数据的大小有关。 下表按行数显示了建议的训练时间范围;列数较多则所需训练时间更长。

      建议的训练时间
      少于 10 万 1 - 3 小时
      10 万 - 100 万 1 - 6 小时
      100 万 - 1000 万 1 - 12 小时
      超过 1000 万 3 - 24 小时

      如需了解训练价格,请参阅价格页面

    13. 点击开始训练

      模型训练可能需要几个小时,具体取决于数据的大小和复杂性,以及训练预算(如果指定)。您可以关闭此标签页,稍后再返回。模型完成训练后,您会收到电子邮件。

API

选择表格数据类型目标。

分类

选择语言或环境标签页:

REST 和命令行

您可以使用 trainingPipelines.create 命令训练模型。

训练模型。

在使用任何请求数据之前,请先进行以下替换:

  • LOCATION:您的区域。
  • PROJECT:您的项目 ID 或项目编号。
  • TRAININGPIPELINE_DISPLAY_NAME:为此操作创建的训练流水线的显示名称。
  • TARGET_COLUMN:您希望此模型预测的列(值)。
  • WEIGHT_COLUMN:(可选)权重列。了解详情
  • TRAINING_BUDGET:您希望模型训练的最长时间,以毫节点时为单位(1000 毫节点时等于一节点时)。
  • OPTIMIZATION_OBJECTIVE:仅当您不希望预测类型的默认优化目标时,才需要。了解详情
  • TRANSFORMATION_TYPE:将为用于训练模型的每一列提供转换类型。了解详情
  • COLUMN_NAME:具有指定转换类型的列的名称。必须指定用于训练模型的每一列。
  • MODEL_DISPLAY_NAME:新训练模型的名称。
  • DATASET_ID:训练数据集的 ID。
  • 您可以提供 Split 对象来控制数据拆分。如需了解如何控制数据拆分,请参阅使用 REST 控制数据拆分
  • PROJECT_NUMBER:您的项目的项目编号(显示在响应中)

HTTP 方法和网址:

POST https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT/locations/LOCATION/trainingPipelines

请求 JSON 正文:

{
    "displayName": "TRAININGPIPELINE_DISPLAY_NAME",
    "trainingTaskDefinition": "gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_tabular_1.0.0.yaml",
    "trainingTaskInputs": {
        "targetColumn": "TARGET_COLUMN",
        "weightColumn": "WEIGHT_COLUMN",
        "predictionType": "classification",
        "trainBudgetMilliNodeHours": TRAINING_BUDGET,
        "optimizationObjective": "OPTIMIZATION_OBJECTIVE",
        "transformations": [
            {"TRANSFORMATION_TYPE_1":  {"column_name" : "COLUMN_NAME_1"} },
            {"TRANSFORMATION_TYPE_2":  {"column_name" : "COLUMN_NAME_2"} },
            ...
    },
    "modelToUpload": {"displayName": "MODEL_DISPLAY_NAME"},
    "inputDataConfig": {
      "datasetId": "DATASET_ID",
    }
}

如需发送您的请求,请展开以下选项之一:

您应会收到如下所示的 JSON 响应:

{
  "name": "projects/PROJECT_NUMBER/locations/us-central1/trainingPipelines/4567",
  "displayName": "myModelName",
  "trainingTaskDefinition": "gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_tabular_1.0.0.yaml",
  "modelToUpload": {
    "displayName": "myModelName"
  },
  "state": "PIPELINE_STATE_PENDING",
  "createTime": "2020-08-18T01:22:57.479336Z",
  "updateTime": "2020-08-18T01:22:57.479336Z"
}

Java

如需了解如何安装和使用 Vertex AI 客户端库,请参阅 Vertex AI 客户端库。如需了解详情,请参阅 Vertex AI Java API 参考文档


import com.google.cloud.aiplatform.util.ValueConverter;
import com.google.cloud.aiplatform.v1.DeployedModelRef;
import com.google.cloud.aiplatform.v1.EnvVar;
import com.google.cloud.aiplatform.v1.FilterSplit;
import com.google.cloud.aiplatform.v1.FractionSplit;
import com.google.cloud.aiplatform.v1.InputDataConfig;
import com.google.cloud.aiplatform.v1.LocationName;
import com.google.cloud.aiplatform.v1.Model;
import com.google.cloud.aiplatform.v1.ModelContainerSpec;
import com.google.cloud.aiplatform.v1.PipelineServiceClient;
import com.google.cloud.aiplatform.v1.PipelineServiceSettings;
import com.google.cloud.aiplatform.v1.Port;
import com.google.cloud.aiplatform.v1.PredefinedSplit;
import com.google.cloud.aiplatform.v1.PredictSchemata;
import com.google.cloud.aiplatform.v1.TimestampSplit;
import com.google.cloud.aiplatform.v1.TrainingPipeline;
import com.google.cloud.aiplatform.v1.schema.trainingjob.definition.AutoMlTablesInputs;
import com.google.cloud.aiplatform.v1.schema.trainingjob.definition.AutoMlTablesInputs.Transformation;
import com.google.cloud.aiplatform.v1.schema.trainingjob.definition.AutoMlTablesInputs.Transformation.AutoTransformation;
import com.google.rpc.Status;
import java.io.IOException;
import java.util.ArrayList;

public class CreateTrainingPipelineTabularClassificationSample {

  public static void main(String[] args) throws IOException {
    // TODO(developer): Replace these variables before running the sample.
    String project = "YOUR_PROJECT_ID";
    String modelDisplayName = "YOUR_DATASET_DISPLAY_NAME";
    String datasetId = "YOUR_DATASET_ID";
    String targetColumn = "TARGET_COLUMN";
    createTrainingPipelineTableClassification(project, modelDisplayName, datasetId, targetColumn);
  }

  static void createTrainingPipelineTableClassification(
      String project, String modelDisplayName, String datasetId, String targetColumn)
      throws IOException {
    PipelineServiceSettings pipelineServiceSettings =
        PipelineServiceSettings.newBuilder()
            .setEndpoint("us-central1-aiplatform.googleapis.com:443")
            .build();

    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (PipelineServiceClient pipelineServiceClient =
        PipelineServiceClient.create(pipelineServiceSettings)) {
      String location = "us-central1";
      LocationName locationName = LocationName.of(project, location);
      String trainingTaskDefinition =
          "gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_tables_1.0.0.yaml";

      // Set the columns used for training and their data types
      Transformation transformation1 =
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("sepal_width").build())
              .build();
      Transformation transformation2 =
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("sepal_length").build())
              .build();
      Transformation transformation3 =
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("petal_length").build())
              .build();
      Transformation transformation4 =
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("petal_width").build())
              .build();

      ArrayList<Transformation> transformationArrayList = new ArrayList<>();
      transformationArrayList.add(transformation1);
      transformationArrayList.add(transformation2);
      transformationArrayList.add(transformation3);
      transformationArrayList.add(transformation4);

      AutoMlTablesInputs autoMlTablesInputs =
          AutoMlTablesInputs.newBuilder()
              .setTargetColumn(targetColumn)
              .setPredictionType("classification")
              .addAllTransformations(transformationArrayList)
              .setTrainBudgetMilliNodeHours(8000)
              .build();

      FractionSplit fractionSplit =
          FractionSplit.newBuilder()
              .setTrainingFraction(0.8)
              .setValidationFraction(0.1)
              .setTestFraction(0.1)
              .build();

      InputDataConfig inputDataConfig =
          InputDataConfig.newBuilder()
              .setDatasetId(datasetId)
              .setFractionSplit(fractionSplit)
              .build();
      Model modelToUpload = Model.newBuilder().setDisplayName(modelDisplayName).build();

      TrainingPipeline trainingPipeline =
          TrainingPipeline.newBuilder()
              .setDisplayName(modelDisplayName)
              .setTrainingTaskDefinition(trainingTaskDefinition)
              .setTrainingTaskInputs(ValueConverter.toValue(autoMlTablesInputs))
              .setInputDataConfig(inputDataConfig)
              .setModelToUpload(modelToUpload)
              .build();

      TrainingPipeline trainingPipelineResponse =
          pipelineServiceClient.createTrainingPipeline(locationName, trainingPipeline);

      System.out.println("Create Training Pipeline Tabular Classification Response");
      System.out.format("\tName: %s\n", trainingPipelineResponse.getName());
      System.out.format("\tDisplay Name: %s\n", trainingPipelineResponse.getDisplayName());
      System.out.format(
          "\tTraining Task Definition: %s\n", trainingPipelineResponse.getTrainingTaskDefinition());
      System.out.format(
          "\tTraining Task Inputs: %s\n", trainingPipelineResponse.getTrainingTaskInputs());
      System.out.format(
          "\tTraining Task Metadata: %s\n", trainingPipelineResponse.getTrainingTaskMetadata());

      System.out.format("\tState: %s\n", trainingPipelineResponse.getState());
      System.out.format("\tCreate Time: %s\n", trainingPipelineResponse.getCreateTime());
      System.out.format("\tStart Time: %s\n", trainingPipelineResponse.getStartTime());
      System.out.format("\tEnd Time: %s\n", trainingPipelineResponse.getEndTime());
      System.out.format("\tUpdate Time: %s\n", trainingPipelineResponse.getUpdateTime());
      System.out.format("\tLabels: %s\n", trainingPipelineResponse.getLabelsMap());

      InputDataConfig inputDataConfigResponse = trainingPipelineResponse.getInputDataConfig();
      System.out.println("\tInput Data Config");
      System.out.format("\t\tDataset Id: %s\n", inputDataConfigResponse.getDatasetId());
      System.out.format(
          "\t\tAnnotations Filter: %s\n", inputDataConfigResponse.getAnnotationsFilter());

      FractionSplit fractionSplitResponse = inputDataConfigResponse.getFractionSplit();
      System.out.println("\t\tFraction Split");
      System.out.format(
          "\t\t\tTraining Fraction: %s\n", fractionSplitResponse.getTrainingFraction());
      System.out.format(
          "\t\t\tValidation Fraction: %s\n", fractionSplitResponse.getValidationFraction());
      System.out.format("\t\t\tTest Fraction: %s\n", fractionSplitResponse.getTestFraction());

      FilterSplit filterSplit = inputDataConfigResponse.getFilterSplit();
      System.out.println("\t\tFilter Split");
      System.out.format("\t\t\tTraining Fraction: %s\n", filterSplit.getTrainingFilter());
      System.out.format("\t\t\tValidation Fraction: %s\n", filterSplit.getValidationFilter());
      System.out.format("\t\t\tTest Fraction: %s\n", filterSplit.getTestFilter());

      PredefinedSplit predefinedSplit = inputDataConfigResponse.getPredefinedSplit();
      System.out.println("\t\tPredefined Split");
      System.out.format("\t\t\tKey: %s\n", predefinedSplit.getKey());

      TimestampSplit timestampSplit = inputDataConfigResponse.getTimestampSplit();
      System.out.println("\t\tTimestamp Split");
      System.out.format("\t\t\tTraining Fraction: %s\n", timestampSplit.getTrainingFraction());
      System.out.format("\t\t\tValidation Fraction: %s\n", timestampSplit.getValidationFraction());
      System.out.format("\t\t\tTest Fraction: %s\n", timestampSplit.getTestFraction());
      System.out.format("\t\t\tKey: %s\n", timestampSplit.getKey());

      Model modelResponse = trainingPipelineResponse.getModelToUpload();
      System.out.println("\tModel To Upload");
      System.out.format("\t\tName: %s\n", modelResponse.getName());
      System.out.format("\t\tDisplay Name: %s\n", modelResponse.getDisplayName());
      System.out.format("\t\tDescription: %s\n", modelResponse.getDescription());
      System.out.format("\t\tMetadata Schema Uri: %s\n", modelResponse.getMetadataSchemaUri());
      System.out.format("\t\tMeta Data: %s\n", modelResponse.getMetadata());
      System.out.format("\t\tTraining Pipeline: %s\n", modelResponse.getTrainingPipeline());
      System.out.format("\t\tArtifact Uri: %s\n", modelResponse.getArtifactUri());

      System.out.format(
          "\t\tSupported Deployment Resources Types: %s\n",
          modelResponse.getSupportedDeploymentResourcesTypesList().toString());
      System.out.format(
          "\t\tSupported Input Storage Formats: %s\n",
          modelResponse.getSupportedInputStorageFormatsList().toString());
      System.out.format(
          "\t\tSupported Output Storage Formats: %s\n",
          modelResponse.getSupportedOutputStorageFormatsList().toString());

      System.out.format("\t\tCreate Time: %s\n", modelResponse.getCreateTime());
      System.out.format("\t\tUpdate Time: %s\n", modelResponse.getUpdateTime());
      System.out.format("\t\tLables: %s\n", modelResponse.getLabelsMap());
      PredictSchemata predictSchemata = modelResponse.getPredictSchemata();

      System.out.println("\tPredict Schemata");
      System.out.format("\t\tInstance Schema Uri: %s\n", predictSchemata.getInstanceSchemaUri());
      System.out.format(
          "\t\tParameters Schema Uri: %s\n", predictSchemata.getParametersSchemaUri());
      System.out.format(
          "\t\tPrediction Schema Uri: %s\n", predictSchemata.getPredictionSchemaUri());

      for (Model.ExportFormat supportedExportFormat :
          modelResponse.getSupportedExportFormatsList()) {
        System.out.println("\tSupported Export Format");
        System.out.format("\t\tId: %s\n", supportedExportFormat.getId());
      }
      ModelContainerSpec containerSpec = modelResponse.getContainerSpec();

      System.out.println("\tContainer Spec");
      System.out.format("\t\tImage Uri: %s\n", containerSpec.getImageUri());
      System.out.format("\t\tCommand: %s\n", containerSpec.getCommandList());
      System.out.format("\t\tArgs: %s\n", containerSpec.getArgsList());
      System.out.format("\t\tPredict Route: %s\n", containerSpec.getPredictRoute());
      System.out.format("\t\tHealth Route: %s\n", containerSpec.getHealthRoute());

      for (EnvVar envVar : containerSpec.getEnvList()) {
        System.out.println("\t\tEnv");
        System.out.format("\t\t\tName: %s\n", envVar.getName());
        System.out.format("\t\t\tValue: %s\n", envVar.getValue());
      }

      for (Port port : containerSpec.getPortsList()) {
        System.out.println("\t\tPort");
        System.out.format("\t\t\tContainer Port: %s\n", port.getContainerPort());
      }

      for (DeployedModelRef deployedModelRef : modelResponse.getDeployedModelsList()) {
        System.out.println("\tDeployed Model");
        System.out.format("\t\tEndpoint: %s\n", deployedModelRef.getEndpoint());
        System.out.format("\t\tDeployed Model Id: %s\n", deployedModelRef.getDeployedModelId());
      }

      Status status = trainingPipelineResponse.getError();
      System.out.println("\tError");
      System.out.format("\t\tCode: %s\n", status.getCode());
      System.out.format("\t\tMessage: %s\n", status.getMessage());
    }
  }
}

Node.js

如需了解如何安装和使用 Vertex AI 客户端库,请参阅 Vertex AI 客户端库。如需了解详情,请参阅 Vertex AI Node.js API 参考文档

/**
 * TODO(developer): Uncomment these variables before running the sample.\
 * (Not necessary if passing values as arguments)
 */

// const datasetId = 'YOUR_DATASET_ID';
// const modelDisplayName = 'YOUR_MODEL_DISPLAY_NAME';
// const trainingPipelineDisplayName = 'YOUR_TRAINING_PIPELINE_DISPLAY_NAME';
// const targetColumn = 'YOUR_TARGET_COLUMN';
// const project = 'YOUR_PROJECT_ID';
// const location = 'YOUR_PROJECT_LOCATION';
const aiplatform = require('@google-cloud/aiplatform');
const {definition} =
  aiplatform.protos.google.cloud.aiplatform.v1.schema.trainingjob;

// Imports the Google Cloud Pipeline Service Client library
const {PipelineServiceClient} = aiplatform.v1;
// Specifies the location of the api endpoint
const clientOptions = {
  apiEndpoint: 'us-central1-aiplatform.googleapis.com',
};

// Instantiates a client
const pipelineServiceClient = new PipelineServiceClient(clientOptions);

async function createTrainingPipelineTablesClassification() {
  // Configure the parent resource
  const parent = `projects/${project}/locations/${location}`;

  const transformations = [
    {auto: {column_name: 'sepal_width'}},
    {auto: {column_name: 'sepal_length'}},
    {auto: {column_name: 'petal_length'}},
    {auto: {column_name: 'petal_width'}},
  ];
  const trainingTaskInputsObj = new definition.AutoMlTablesInputs({
    targetColumn: targetColumn,
    predictionType: 'classification',
    transformations: transformations,
    trainBudgetMilliNodeHours: 8000,
    disableEarlyStopping: false,
    optimizationObjective: 'minimize-log-loss',
  });
  const trainingTaskInputs = trainingTaskInputsObj.toValue();

  const modelToUpload = {displayName: modelDisplayName};
  const inputDataConfig = {
    datasetId: datasetId,
    fractionSplit: {
      trainingFraction: 0.8,
      validationFraction: 0.1,
      testFraction: 0.1,
    },
  };
  const trainingPipeline = {
    displayName: trainingPipelineDisplayName,
    trainingTaskDefinition:
      'gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_tables_1.0.0.yaml',
    trainingTaskInputs,
    inputDataConfig,
    modelToUpload,
  };
  const request = {
    parent,
    trainingPipeline,
  };

  // Create training pipeline request
  const [response] = await pipelineServiceClient.createTrainingPipeline(
    request
  );

  console.log('Create training pipeline tabular classification response');
  console.log(`Name : ${response.name}`);
  console.log('Raw response:');
  console.log(JSON.stringify(response, null, 2));
}
createTrainingPipelineTablesClassification();

Python

如需了解如何安装和使用 Vertex AI 客户端库,请参阅 Vertex AI 客户端库。如需了解详情,请参阅 Vertex AI Python API 参考文档

def create_training_pipeline_tabular_classification_sample(
    project: str,
    display_name: str,
    dataset_id: int,
    location: str = "us-central1",
    model_display_name: str = None,
    training_fraction_split: float = 0.8,
    validation_fraction_split: float = 0.1,
    test_fraction_split: float = 0.1,
    budget_milli_node_hours: int = 8000,
    disable_early_stopping: bool = False,
    sync: bool = True,
):
    aiplatform.init(project=project, location=location)

    tabular_classification_job = aiplatform.AutoMLTabularTrainingJob(
        display_name=display_name,
        optimization_prediction_type="classification"
    )

    my_tabular_dataset = aiplatform.TabularDataset(dataset_id)

    model = tabular_classification_job.run(
        dataset=my_tabular_dataset,
        training_fraction_split=training_fraction_split,
        validation_fraction_split=validation_fraction_split,
        test_fraction_split=test_fraction_split,
        budget_milli_node_hours=budget_milli_node_hours,
        model_display_name=model_display_name,
        disable_early_stopping=disable_early_stopping,
        sync=sync,
    )

    model.wait()

    print(model.display_name)
    print(model.resource_name)
    print(model.uri)
    return model

回归

选择语言或环境标签页:

REST 和命令行

您可以使用 trainingPipelines.create 命令训练模型。

训练模型。

在使用任何请求数据之前,请先进行以下替换:

  • LOCATION:您的区域。
  • PROJECT:您的项目 ID 或项目编号。
  • TRAININGPIPELINE_DISPLAY_NAME:为此操作创建的训练流水线的显示名称。
  • TARGET_COLUMN:您希望此模型预测的列(值)。
  • WEIGHT_COLUMN:(可选)权重列。了解详情
  • TRAINING_BUDGET:您希望模型训练的最长时间,以毫节点时为单位(1000 毫节点时等于一节点时)。
  • OPTIMIZATION_OBJECTIVE:仅当您不希望预测类型的默认优化目标时,才需要。了解详情
  • TRANSFORMATION_TYPE:将为用于训练模型的每一列提供转换类型。了解详情
  • COLUMN_NAME:具有指定转换类型的列的名称。必须指定用于训练模型的每一列。
  • MODEL_DISPLAY_NAME:新训练模型的名称。
  • DATASET_ID:训练数据集的 ID。
  • 您可以提供 Split 对象来控制数据拆分。如需了解如何控制数据拆分,请参阅使用 REST 控制数据拆分
  • PROJECT_NUMBER:您的项目的项目编号(显示在响应中)

HTTP 方法和网址:

POST https://LOCATION-aiplatform.googleapis.com/v1/projects/PROJECT/locations/LOCATION/trainingPipelines

请求 JSON 正文:

{
    "displayName": "TRAININGPIPELINE_DISPLAY_NAME",
    "trainingTaskDefinition": "gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_tabular_1.0.0.yaml",
    "trainingTaskInputs": {
        "targetColumn": "TARGET_COLUMN",
        "weightColumn": "WEIGHT_COLUMN",
        "predictionType": "regression",
        "trainBudgetMilliNodeHours": TRAINING_BUDGET,
        "optimizationObjective": "OPTIMIZATION_OBJECTIVE",
        "transformations": [
            {"TRANSFORMATION_TYPE_1":  {"column_name" : "COLUMN_NAME_1"} },
            {"TRANSFORMATION_TYPE_2":  {"column_name" : "COLUMN_NAME_2"} },
            ...
    },
    "modelToUpload": {"displayName": "MODEL_DISPLAY_NAME"},
    "inputDataConfig": {
      "datasetId": "DATASET_ID",
    }
}

如需发送您的请求,请展开以下选项之一:

您应会收到如下所示的 JSON 响应:

{
  "name": "projects/PROJECT_NUMBER/locations/us-central1/trainingPipelines/4567",
  "displayName": "myModelName",
  "trainingTaskDefinition": "gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_tabular_1.0.0.yaml",
  "modelToUpload": {
    "displayName": "myModelName"
  },
  "state": "PIPELINE_STATE_PENDING",
  "createTime": "2020-08-18T01:22:57.479336Z",
  "updateTime": "2020-08-18T01:22:57.479336Z"
}

Java

如需了解如何安装和使用 Vertex AI 客户端库,请参阅 Vertex AI 客户端库。如需了解详情,请参阅 Vertex AI Java API 参考文档


import com.google.cloud.aiplatform.util.ValueConverter;
import com.google.cloud.aiplatform.v1.DeployedModelRef;
import com.google.cloud.aiplatform.v1.EnvVar;
import com.google.cloud.aiplatform.v1.FilterSplit;
import com.google.cloud.aiplatform.v1.FractionSplit;
import com.google.cloud.aiplatform.v1.InputDataConfig;
import com.google.cloud.aiplatform.v1.LocationName;
import com.google.cloud.aiplatform.v1.Model;
import com.google.cloud.aiplatform.v1.ModelContainerSpec;
import com.google.cloud.aiplatform.v1.PipelineServiceClient;
import com.google.cloud.aiplatform.v1.PipelineServiceSettings;
import com.google.cloud.aiplatform.v1.Port;
import com.google.cloud.aiplatform.v1.PredefinedSplit;
import com.google.cloud.aiplatform.v1.PredictSchemata;
import com.google.cloud.aiplatform.v1.TimestampSplit;
import com.google.cloud.aiplatform.v1.TrainingPipeline;
import com.google.cloud.aiplatform.v1beta1.schema.trainingjob.definition.AutoMlTablesInputs;
import com.google.cloud.aiplatform.v1beta1.schema.trainingjob.definition.AutoMlTablesInputs.Transformation;
import com.google.cloud.aiplatform.v1beta1.schema.trainingjob.definition.AutoMlTablesInputs.Transformation.AutoTransformation;
import com.google.cloud.aiplatform.v1beta1.schema.trainingjob.definition.AutoMlTablesInputs.Transformation.TimestampTransformation;
import com.google.rpc.Status;
import java.io.IOException;
import java.util.ArrayList;

public class CreateTrainingPipelineTabularRegressionSample {

  public static void main(String[] args) throws IOException {
    // TODO(developer): Replace these variables before running the sample.
    String project = "YOUR_PROJECT_ID";
    String modelDisplayName = "YOUR_DATASET_DISPLAY_NAME";
    String datasetId = "YOUR_DATASET_ID";
    String targetColumn = "TARGET_COLUMN";
    createTrainingPipelineTableRegression(project, modelDisplayName, datasetId, targetColumn);
  }

  static void createTrainingPipelineTableRegression(
      String project, String modelDisplayName, String datasetId, String targetColumn)
      throws IOException {
    PipelineServiceSettings pipelineServiceSettings =
        PipelineServiceSettings.newBuilder()
            .setEndpoint("us-central1-aiplatform.googleapis.com:443")
            .build();

    // Initialize client that will be used to send requests. This client only needs to be created
    // once, and can be reused for multiple requests. After completing all of your requests, call
    // the "close" method on the client to safely clean up any remaining background resources.
    try (PipelineServiceClient pipelineServiceClient =
        PipelineServiceClient.create(pipelineServiceSettings)) {
      String location = "us-central1";
      LocationName locationName = LocationName.of(project, location);
      String trainingTaskDefinition =
          "gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_tables_1.0.0.yaml";

      // Set the columns used for training and their data types
      ArrayList<Transformation> tranformations = new ArrayList<>();
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("STRING_5000unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("INTEGER_5000unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("FLOAT_5000unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("FLOAT_5000unique_REPEATED"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("NUMERIC_5000unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("BOOLEAN_2unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setTimestamp(
                  TimestampTransformation.newBuilder()
                      .setColumnName("TIMESTAMP_1unique_NULLABLE")
                      .setInvalidValuesAllowed(true))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("DATE_1unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(AutoTransformation.newBuilder().setColumnName("TIME_1unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setTimestamp(
                  TimestampTransformation.newBuilder()
                      .setColumnName("DATETIME_1unique_NULLABLE")
                      .setInvalidValuesAllowed(true))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(
                  AutoTransformation.newBuilder()
                      .setColumnName("STRUCT_NULLABLE.STRING_5000unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(
                  AutoTransformation.newBuilder()
                      .setColumnName("STRUCT_NULLABLE.INTEGER_5000unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(
                  AutoTransformation.newBuilder()
                      .setColumnName("STRUCT_NULLABLE.FLOAT_5000unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(
                  AutoTransformation.newBuilder()
                      .setColumnName("STRUCT_NULLABLE.FLOAT_5000unique_REQUIRED"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(
                  AutoTransformation.newBuilder()
                      .setColumnName("STRUCT_NULLABLE.FLOAT_5000unique_REPEATED"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(
                  AutoTransformation.newBuilder()
                      .setColumnName("STRUCT_NULLABLE.NUMERIC_5000unique_NULLABLE"))
              .build());
      tranformations.add(
          Transformation.newBuilder()
              .setAuto(
                  AutoTransformation.newBuilder()
                      .setColumnName("STRUCT_NULLABLE.TIMESTAMP_1unique_NULLABLE"))
              .build());

      AutoMlTablesInputs trainingTaskInputs =
          AutoMlTablesInputs.newBuilder()
              .addAllTransformations(tranformations)
              .setTargetColumn(targetColumn)
              .setPredictionType("regression")
              .setTrainBudgetMilliNodeHours(8000)
              .setDisableEarlyStopping(false)
              // supported regression optimisation objectives: minimize-rmse,
              // minimize-mae, minimize-rmsle
              .setOptimizationObjective("minimize-rmse")
              .build();

      FractionSplit fractionSplit =
          FractionSplit.newBuilder()
              .setTrainingFraction(0.8)
              .setValidationFraction(0.1)
              .setTestFraction(0.1)
              .build();

      InputDataConfig inputDataConfig =
          InputDataConfig.newBuilder()
              .setDatasetId(datasetId)
              .setFractionSplit(fractionSplit)
              .build();
      Model modelToUpload = Model.newBuilder().setDisplayName(modelDisplayName).build();

      TrainingPipeline trainingPipeline =
          TrainingPipeline.newBuilder()
              .setDisplayName(modelDisplayName)
              .setTrainingTaskDefinition(trainingTaskDefinition)
              .setTrainingTaskInputs(ValueConverter.toValue(trainingTaskInputs))
              .setInputDataConfig(inputDataConfig)
              .setModelToUpload(modelToUpload)
              .build();

      TrainingPipeline trainingPipelineResponse =
          pipelineServiceClient.createTrainingPipeline(locationName, trainingPipeline);

      System.out.println("Create Training Pipeline Tabular Regression Response");
      System.out.format("\tName: %s\n", trainingPipelineResponse.getName());
      System.out.format("\tDisplay Name: %s\n", trainingPipelineResponse.getDisplayName());
      System.out.format(
          "\tTraining Task Definition: %s\n", trainingPipelineResponse.getTrainingTaskDefinition());
      System.out.format(
          "\tTraining Task Inputs: %s\n", trainingPipelineResponse.getTrainingTaskInputs());
      System.out.format(
          "\tTraining Task Metadata: %s\n", trainingPipelineResponse.getTrainingTaskMetadata());

      System.out.format("\tState: %s\n", trainingPipelineResponse.getState());
      System.out.format("\tCreate Time: %s\n", trainingPipelineResponse.getCreateTime());
      System.out.format("\tStart Time: %s\n", trainingPipelineResponse.getStartTime());
      System.out.format("\tEnd Time: %s\n", trainingPipelineResponse.getEndTime());
      System.out.format("\tUpdate Time: %s\n", trainingPipelineResponse.getUpdateTime());
      System.out.format("\tLabels: %s\n", trainingPipelineResponse.getLabelsMap());

      InputDataConfig inputDataConfigResponse = trainingPipelineResponse.getInputDataConfig();
      System.out.println("\tInput Data Config");
      System.out.format("\t\tDataset Id: %s\n", inputDataConfigResponse.getDatasetId());
      System.out.format(
          "\t\tAnnotations Filter: %s\n", inputDataConfigResponse.getAnnotationsFilter());

      FractionSplit fractionSplitResponse = inputDataConfigResponse.getFractionSplit();
      System.out.println("\t\tFraction Split");
      System.out.format(
          "\t\t\tTraining Fraction: %s\n", fractionSplitResponse.getTrainingFraction());
      System.out.format(
          "\t\t\tValidation Fraction: %s\n", fractionSplitResponse.getValidationFraction());
      System.out.format("\t\t\tTest Fraction: %s\n", fractionSplitResponse.getTestFraction());

      FilterSplit filterSplit = inputDataConfigResponse.getFilterSplit();
      System.out.println("\t\tFilter Split");
      System.out.format("\t\t\tTraining Fraction: %s\n", filterSplit.getTrainingFilter());
      System.out.format("\t\t\tValidation Fraction: %s\n", filterSplit.getValidationFilter());
      System.out.format("\t\t\tTest Fraction: %s\n", filterSplit.getTestFilter());

      PredefinedSplit predefinedSplit = inputDataConfigResponse.getPredefinedSplit();
      System.out.println("\t\tPredefined Split");
      System.out.format("\t\t\tKey: %s\n", predefinedSplit.getKey());

      TimestampSplit timestampSplit = inputDataConfigResponse.getTimestampSplit();
      System.out.println("\t\tTimestamp Split");
      System.out.format("\t\t\tTraining Fraction: %s\n", timestampSplit.getTrainingFraction());
      System.out.format("\t\t\tValidation Fraction: %s\n", timestampSplit.getValidationFraction());
      System.out.format("\t\t\tTest Fraction: %s\n", timestampSplit.getTestFraction());
      System.out.format("\t\t\tKey: %s\n", timestampSplit.getKey());

      Model modelResponse = trainingPipelineResponse.getModelToUpload();
      System.out.println("\tModel To Upload");
      System.out.format("\t\tName: %s\n", modelResponse.getName());
      System.out.format("\t\tDisplay Name: %s\n", modelResponse.getDisplayName());
      System.out.format("\t\tDescription: %s\n", modelResponse.getDescription());
      System.out.format("\t\tMetadata Schema Uri: %s\n", modelResponse.getMetadataSchemaUri());
      System.out.format("\t\tMeta Data: %s\n", modelResponse.getMetadata());
      System.out.format("\t\tTraining Pipeline: %s\n", modelResponse.getTrainingPipeline());
      System.out.format("\t\tArtifact Uri: %s\n", modelResponse.getArtifactUri());

      System.out.format(
          "\t\tSupported Deployment Resources Types: %s\n",
          modelResponse.getSupportedDeploymentResourcesTypesList().toString());
      System.out.format(
          "\t\tSupported Input Storage Formats: %s\n",
          modelResponse.getSupportedInputStorageFormatsList().toString());
      System.out.format(
          "\t\tSupported Output Storage Formats: %s\n",
          modelResponse.getSupportedOutputStorageFormatsList().toString());

      System.out.format("\t\tCreate Time: %s\n", modelResponse.getCreateTime());
      System.out.format("\t\tUpdate Time: %s\n", modelResponse.getUpdateTime());
      System.out.format("\t\tLables: %s\n", modelResponse.getLabelsMap());
      PredictSchemata predictSchemata = modelResponse.getPredictSchemata();

      System.out.println("\tPredict Schemata");
      System.out.format("\t\tInstance Schema Uri: %s\n", predictSchemata.getInstanceSchemaUri());
      System.out.format(
          "\t\tParameters Schema Uri: %s\n", predictSchemata.getParametersSchemaUri());
      System.out.format(
          "\t\tPrediction Schema Uri: %s\n", predictSchemata.getPredictionSchemaUri());

      for (Model.ExportFormat supportedExportFormat :
          modelResponse.getSupportedExportFormatsList()) {
        System.out.println("\tSupported Export Format");
        System.out.format("\t\tId: %s\n", supportedExportFormat.getId());
      }
      ModelContainerSpec containerSpec = modelResponse.getContainerSpec();

      System.out.println("\tContainer Spec");
      System.out.format("\t\tImage Uri: %s\n", containerSpec.getImageUri());
      System.out.format("\t\tCommand: %s\n", containerSpec.getCommandList());
      System.out.format("\t\tArgs: %s\n", containerSpec.getArgsList());
      System.out.format("\t\tPredict Route: %s\n", containerSpec.getPredictRoute());
      System.out.format("\t\tHealth Route: %s\n", containerSpec.getHealthRoute());

      for (EnvVar envVar : containerSpec.getEnvList()) {
        System.out.println("\t\tEnv");
        System.out.format("\t\t\tName: %s\n", envVar.getName());
        System.out.format("\t\t\tValue: %s\n", envVar.getValue());
      }

      for (Port port : containerSpec.getPortsList()) {
        System.out.println("\t\tPort");
        System.out.format("\t\t\tContainer Port: %s\n", port.getContainerPort());
      }

      for (DeployedModelRef deployedModelRef : modelResponse.getDeployedModelsList()) {
        System.out.println("\tDeployed Model");
        System.out.format("\t\tEndpoint: %s\n", deployedModelRef.getEndpoint());
        System.out.format("\t\tDeployed Model Id: %s\n", deployedModelRef.getDeployedModelId());
      }

      Status status = trainingPipelineResponse.getError();
      System.out.println("\tError");
      System.out.format("\t\tCode: %s\n", status.getCode());
      System.out.format("\t\tMessage: %s\n", status.getMessage());
    }
  }
}

Node.js

如需了解如何安装和使用 Vertex AI 客户端库,请参阅 Vertex AI 客户端库。如需了解详情,请参阅 Vertex AI Node.js API 参考文档

/**
 * TODO(developer): Uncomment these variables before running the sample.\
 * (Not necessary if passing values as arguments)
 */

// const datasetId = 'YOUR_DATASET_ID';
// const modelDisplayName = 'YOUR_MODEL_DISPLAY_NAME';
// const trainingPipelineDisplayName = 'YOUR_TRAINING_PIPELINE_DISPLAY_NAME';
// const targetColumn = 'YOUR_TARGET_COLUMN';
// const project = 'YOUR_PROJECT_ID';
// const location = 'YOUR_PROJECT_LOCATION';
const aiplatform = require('@google-cloud/aiplatform');
const {definition} =
  aiplatform.protos.google.cloud.aiplatform.v1.schema.trainingjob;

// Imports the Google Cloud Pipeline Service Client library
const {PipelineServiceClient} = aiplatform.v1;
// Specifies the location of the api endpoint
const clientOptions = {
  apiEndpoint: 'us-central1-aiplatform.googleapis.com',
};

// Instantiates a client
const pipelineServiceClient = new PipelineServiceClient(clientOptions);

async function createTrainingPipelineTablesRegression() {
  // Configure the parent resource
  const parent = `projects/${project}/locations/${location}`;

  const transformations = [
    {auto: {column_name: 'STRING_5000unique_NULLABLE'}},
    {auto: {column_name: 'INTEGER_5000unique_NULLABLE'}},
    {auto: {column_name: 'FLOAT_5000unique_NULLABLE'}},
    {auto: {column_name: 'FLOAT_5000unique_REPEATED'}},
    {auto: {column_name: 'NUMERIC_5000unique_NULLABLE'}},
    {auto: {column_name: 'BOOLEAN_2unique_NULLABLE'}},
    {
      timestamp: {
        column_name: 'TIMESTAMP_1unique_NULLABLE',
        invalid_values_allowed: true,
      },
    },
    {auto: {column_name: 'DATE_1unique_NULLABLE'}},
    {auto: {column_name: 'TIME_1unique_NULLABLE'}},
    {
      timestamp: {
        column_name: 'DATETIME_1unique_NULLABLE',
        invalid_values_allowed: true,
      },
    },
    {auto: {column_name: 'STRUCT_NULLABLE.STRING_5000unique_NULLABLE'}},
    {auto: {column_name: 'STRUCT_NULLABLE.INTEGER_5000unique_NULLABLE'}},
    {auto: {column_name: 'STRUCT_NULLABLE.FLOAT_5000unique_NULLABLE'}},
    {auto: {column_name: 'STRUCT_NULLABLE.FLOAT_5000unique_REQUIRED'}},
    {auto: {column_name: 'STRUCT_NULLABLE.FLOAT_5000unique_REPEATED'}},
    {auto: {column_name: 'STRUCT_NULLABLE.NUMERIC_5000unique_NULLABLE'}},
    {auto: {column_name: 'STRUCT_NULLABLE.BOOLEAN_2unique_NULLABLE'}},
    {auto: {column_name: 'STRUCT_NULLABLE.TIMESTAMP_1unique_NULLABLE'}},
  ];

  const trainingTaskInputsObj = new definition.AutoMlTablesInputs({
    transformations,
    targetColumn,
    predictionType: 'regression',
    trainBudgetMilliNodeHours: 8000,
    disableEarlyStopping: false,
    optimizationObjective: 'minimize-rmse',
  });
  const trainingTaskInputs = trainingTaskInputsObj.toValue();

  const modelToUpload = {displayName: modelDisplayName};
  const inputDataConfig = {
    datasetId: datasetId,
    fractionSplit: {
      trainingFraction: 0.8,
      validationFraction: 0.1,
      testFraction: 0.1,
    },
  };
  const trainingPipeline = {
    displayName: trainingPipelineDisplayName,
    trainingTaskDefinition:
      'gs://google-cloud-aiplatform/schema/trainingjob/definition/automl_tables_1.0.0.yaml',
    trainingTaskInputs,
    inputDataConfig,
    modelToUpload,
  };
  const request = {
    parent,
    trainingPipeline,
  };

  // Create training pipeline request
  const [response] = await pipelineServiceClient.createTrainingPipeline(
    request
  );

  console.log('Create training pipeline tabular regression response');
  console.log(`Name : ${response.name}`);
  console.log('Raw response:');
  console.log(JSON.stringify(response, null, 2));
}
createTrainingPipelineTablesRegression();

Python

如需了解如何安装和使用 Vertex AI 客户端库,请参阅 Vertex AI 客户端库。如需了解详情,请参阅 Vertex AI Python API 参考文档

def create_training_pipeline_tabular_regression_sample(
    project: str,
    display_name: str,
    dataset_id: int,
    location: str = "us-central1",
    model_display_name: str = None,
    training_fraction_split: float = 0.8,
    validation_fraction_split: float = 0.1,
    test_fraction_split: float = 0.1,
    budget_milli_node_hours: int = 8000,
    disable_early_stopping: bool = False,
    sync: bool = True,
):
    aiplatform.init(project=project, location=location)

    tabular_regression_job = aiplatform.AutoMLTabularTrainingJob(
        display_name=display_name,
        optimization_prediction_type="regression"
    )

    my_tabular_dataset = aiplatform.TabularDataset(dataset_id)

    model = tabular_regression_job.run(
        dataset=my_tabular_dataset,
        training_fraction_split=training_fraction_split,
        validation_fraction_split=validation_fraction_split,
        test_fraction_split=test_fraction_split,
        budget_milli_node_hours=budget_milli_node_hours,
        model_display_name=model_display_name,
        disable_early_stopping=disable_early_stopping,
        sync=sync,
    )

    model.wait()

    print(model.display_name)
    print(model.resource_name)
    print(model.uri)
    return model

使用 REST 控制数据拆分

您可以控制在训练集、验证集和测试集之间拆分训练数据的方式。使用 Vertex AI API 时,请使用 Split 对象来确定数据拆分。Split 对象可以包含在 InputConfig 对象中作为多种对象类型中的一种,其中每种类型都提供一种不同的训练数据拆分方式。

可用于拆分数据的方法取决于数据类型:

  • FractionSplit

    • TRAINING_FRACTION:要用于训练集的训练数据的比例。
    • VALIDATION_FRACTION:要用于验证集的训练数据的比例。不用于视频数据。
    • TEST_FRACTION:要用于测试集的训练数据的比例。

    如果指定了任一比例,则必须指定所有比例。这些比例之和必须等于 1.0。比例的默认值会因数据类型而异。了解详情

    "fractionSplit": {
      "trainingFraction": TRAINING_FRACTION,
      "validationFraction": VALIDATION_FRACTION,
      "testFraction": TEST_FRACTION
    },
    

  • PredefinedSplit

    • DATA_SPLIT_COLUMN:包含数据拆分值(TRAINVALIDATIONTEST)的列。

    使用拆分列为每行手动指定数据拆分。 了解详情

    "predefinedSplit": {
      "key": DATA_SPLIT_COLUMN
    },
    
  • TimestampSplit

    • TRAINING_FRACTION:要用于训练集的训练数据的百分比。默认值为 0.80。
    • VALIDATION_FRACTION:要用于验证集的训练数据的百分比。默认值为 0.10。
    • TEST_FRACTION:要用于测试集的训练数据的百分比。默认值为 0.10。
    • TIME_COLUMN:包含时间戳的列。

    如果指定了任一比例,则必须指定所有比例。这些比例之和必须等于 1.0。了解详情

    "timestampSplit": {
      "trainingFraction": TRAINING_FRACTION,
      "validationFraction": VALIDATION_FRACTION,
      "testFraction": TEST_FRACTION,
      "key": TIME_COLUMN
    }
    

后续步骤