训练 Edge 可导出模型

创建自定义模型的方法,是使用准备好的数据集对其进行训练。AutoML API 使用数据集中的条目来训练、测试模型并评估其性能。您可以查看结果、根据需要调整训练数据集,并使用改进后的数据集训练新的模型。

训练模型可能需要几个小时才能完成。借助 AutoML API,您可以检查训练的状态

每次开始训练时,AutoML Vision 都会创建新模型,因此您的项目可能包含大量模型。您可以获取项目中模型的列表删除不再需要的模型。或者,您也可以使用 Cloud AutoML Vision 界面来列出和删除通过 AutoML API 创建且不再需要的模型。

模型基于 Google 最先进的研究。您的模型将采用 TF Lite 软件包的形式。如需详细了解如何使用 TensorFlow Lite SDK 集成 TensorFlow Lite 模型,请点击此处针对 iOSAndroid 的相应链接。

训练 Edge 模型

如果您有一个包含一组带标签的固定训练项的数据集,就可以创建和训练自定义 Edge 模型了。

其他语言

C#: 请按照客户端库页面上的 C# 设置说明操作,然后访问 .NET 版 AutoML Vision 参考文档。

PHP: 请按照客户端库页面上的 PHP 设置说明操作,然后访问 PHP 版 AutoML Vision 参考文档。

Ruby 版: 请按照客户端库页面上的 Ruby 设置说明操作,然后访问 Ruby 版 AutoML Vision 参考文档。

TensorFlow Serving 和 TF Lite 模型

训练时,您可以根据具体使用场景选择所需的 Edge 模型类型:

  • 低延时 (mobile-low-latency-1)
  • 通用目的 (mobile-versatile-1)
  • 更高预测质量 (mobile-high-accuracy-1)

网页界面

  1. 打开 Vision Dashboard

    数据集页面显示当前项目的可用数据集。

    列出数据集页面
  2. 选择要用于训练自定义模型的数据集。
  3. 准备好数据集后,选择训练标签页和训练新模型按钮。

    此操作会打开包含训练选项的训练新模型侧边窗口。

  4. 定义模型训练部分中,更改模型名称(或使用默认值)并选择 Edge 作为模型类型。选择训练 Edge 模型后,选择继续训练 Edge 模型单选按钮图片
  5. 在接下来的模型优化选项部分中,选择所需的优化条件:较高准确率 (Higher accuracy)、最佳权衡 (Best tradeoff) 或较快预测速度 (Faster prediction)。选择优化规范后,再选择继续

    最佳权衡单选按钮图片
  6. 在接下来的设置节点时预算部分中,使用建议的节点时预算或指定其他值。

    默认情况下,对于大多数数据集来说,24 节点时足以训练模型。此建议值是使模型完全收敛的估算值。但是,您可以选择其他数值。图像分类的最小节点时为 8。对于 Object Detection,此最小值为 20。

    设置节点预算部分
  7. 选择开始训练以开始训练模型。

训练模型可能需要几个小时才能完成。模型训练成功后,您用于 Google Cloud Platform 项目的电子邮件地址会收到一封邮件。

REST

在训练 Edge 模型时,您可以在 modelType 字段中指定三个不同的值,具体取决于您的模型需求:

  • mobile-low-latency-1 表示低延时
  • mobile-versatile-1 表示通用目的
  • mobile-high-accuracy-1 表示较高预测质量

在使用任何请求数据之前,请先进行以下替换:

  • project-id:您的 GCP 项目 ID。
  • display-name:您选择的字符串显示名。
  • dataset-id:您的数据集的 ID。此 ID 是数据集名称的最后一个元素。例如:
    • 数据集名称:projects/project-id/locations/location-id/datasets/3104518874390609379
    • 数据集 ID:3104518874390609379

HTTP 方法和网址:

POST https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/models

请求 JSON 正文:

{
  "displayName": "DISPLAY_NAME",
  "datasetId": "DATASET_ID",
  "imageClassificationModelMetadata": {
    "trainBudget": "1",
    "modelType": "mobile-low-latency-1"
  }
}

如需发送请求,请选择以下方式之一:

curl

将请求正文保存在名为 request.json 的文件中,然后执行以下命令:

curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "x-goog-user-project: project-id" \
-H "Content-Type: application/json; charset=utf-8" \
-d @request.json \
"https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/models"

PowerShell

将请求正文保存在名为 request.json 的文件中,然后执行以下命令:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred"; "x-goog-user-project" = "project-id" }

Invoke-WebRequest `
-Method POST `
-Headers $headers `
-ContentType: "application/json; charset=utf-8" `
-InFile request.json `
-Uri "https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/models" | Select-Object -Expand Content

您应该会看到类似如下所示的输出。可以使用操作 ID(本例中为 ICN2106290444865378475)来获取任务的状态。如需查看示例,请参阅处理长时间运行的操作

{
  "name": "projects/PROJECT_ID/locations/us-central1/operations/ICN2106290444865378475",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.automl.v1.OperationMetadata",
    "createTime": "2019-10-30T20:06:08.253243Z",
    "updateTime": "2019-10-30T20:06:08.253243Z",
    "createModelDetails": {}
  }
}

Go

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

import (
	"context"
	"fmt"
	"io"

	automl "cloud.google.com/go/automl/apiv1"
	"cloud.google.com/go/automl/apiv1/automlpb"
)

// visionClassificationCreateModel creates a model for image classification.
func visionClassificationCreateModel(w io.Writer, projectID string, location string, datasetID string, modelName string) error {
	// projectID := "my-project-id"
	// location := "us-central1"
	// datasetID := "ICN123456789..."
	// modelName := "model_display_name"

	ctx := context.Background()
	client, err := automl.NewClient(ctx)
	if err != nil {
		return fmt.Errorf("NewClient: %w", err)
	}
	defer client.Close()

	req := &automlpb.CreateModelRequest{
		Parent: fmt.Sprintf("projects/%s/locations/%s", projectID, location),
		Model: &automlpb.Model{
			DisplayName: modelName,
			DatasetId:   datasetID,
			ModelMetadata: &automlpb.Model_ImageClassificationModelMetadata{
				ImageClassificationModelMetadata: &automlpb.ImageClassificationModelMetadata{
					TrainBudgetMilliNodeHours: 1000, // 1000 milli-node hours are 1 hour
				},
			},
		},
	}

	op, err := client.CreateModel(ctx, req)
	if err != nil {
		return fmt.Errorf("CreateModel: %w", err)
	}
	fmt.Fprintf(w, "Processing operation name: %q\n", op.Name())
	fmt.Fprintf(w, "Training started...\n")

	return nil
}

Java

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

import com.google.api.gax.longrunning.OperationFuture;
import com.google.cloud.automl.v1.AutoMlClient;
import com.google.cloud.automl.v1.ImageClassificationModelMetadata;
import com.google.cloud.automl.v1.LocationName;
import com.google.cloud.automl.v1.Model;
import com.google.cloud.automl.v1.OperationMetadata;
import java.io.IOException;
import java.util.concurrent.ExecutionException;

class VisionClassificationCreateModel {

  public static void main(String[] args)
      throws IOException, ExecutionException, InterruptedException {
    // TODO(developer): Replace these variables before running the sample.
    String projectId = "YOUR_PROJECT_ID";
    String datasetId = "YOUR_DATASET_ID";
    String displayName = "YOUR_DATASET_NAME";
    createModel(projectId, datasetId, displayName);
  }

  // Create a model
  static void createModel(String projectId, String datasetId, String displayName)
      throws IOException, ExecutionException, InterruptedException {
    // 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 (AutoMlClient client = AutoMlClient.create()) {
      // A resource that represents Google Cloud Platform location.
      LocationName projectLocation = LocationName.of(projectId, "us-central1");
      // Set model metadata.
      ImageClassificationModelMetadata metadata =
          ImageClassificationModelMetadata.newBuilder().setTrainBudgetMilliNodeHours(24000).build();
      Model model =
          Model.newBuilder()
              .setDisplayName(displayName)
              .setDatasetId(datasetId)
              .setImageClassificationModelMetadata(metadata)
              .build();

      // Create a model with the model metadata in the region.
      OperationFuture<Model, OperationMetadata> future =
          client.createModelAsync(projectLocation, model);
      // OperationFuture.get() will block until the model is created, which may take several hours.
      // You can use OperationFuture.getInitialFuture to get a future representing the initial
      // response to the request, which contains information while the operation is in progress.
      System.out.format("Training operation name: %s\n", future.getInitialFuture().get().getName());
      System.out.println("Training started...");
    }
  }
}

Node.js

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

/**
 * TODO(developer): Uncomment these variables before running the sample.
 */
// const projectId = 'YOUR_PROJECT_ID';
// const location = 'us-central1';
// const dataset_id = 'YOUR_DATASET_ID';
// const displayName = 'YOUR_DISPLAY_NAME';

// Imports the Google Cloud AutoML library
const {AutoMlClient} = require('@google-cloud/automl').v1;

// Instantiates a client
const client = new AutoMlClient();

async function createModel() {
  // Construct request
  const request = {
    parent: client.locationPath(projectId, location),
    model: {
      displayName: displayName,
      datasetId: datasetId,
      imageClassificationModelMetadata: {
        trainBudgetMilliNodeHours: 24000,
      },
    },
  };

  // Don't wait for the LRO
  const [operation] = await client.createModel(request);
  console.log(`Training started... ${operation}`);
  console.log(`Training operation name: ${operation.name}`);
}

createModel();

Python

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

from google.cloud import automl

# TODO(developer): Uncomment and set the following variables
# project_id = "YOUR_PROJECT_ID"
# dataset_id = "YOUR_DATASET_ID"
# display_name = "your_models_display_name"

client = automl.AutoMlClient()

# A resource that represents Google Cloud Platform location.
project_location = f"projects/{project_id}/locations/us-central1"
# Leave model unset to use the default base model provided by Google
# train_budget_milli_node_hours: The actual train_cost will be equal or
# less than this value.
# https://cloud.google.com/automl/docs/reference/rpc/google.cloud.automl.v1#imageclassificationmodelmetadata
metadata = automl.ImageClassificationModelMetadata(
    train_budget_milli_node_hours=24000
)
model = automl.Model(
    display_name=display_name,
    dataset_id=dataset_id,
    image_classification_model_metadata=metadata,
)

# Create a model with the model metadata in the region.
response = client.create_model(parent=project_location, model=model)

print(f"Training operation name: {response.operation.name}")
print("Training started...")

列出操作状态

使用以下代码示例列出项目的操作并过滤结果。

REST

在使用任何请求数据之前,请先进行以下替换:

  • project-id:您的 GCP 项目 ID。

HTTP 方法和网址:

GET https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations

如需发送请求,请选择以下方式之一:

curl

执行以下命令:

curl -X GET \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "x-goog-user-project: project-id" \
"https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations"

PowerShell

执行以下命令:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred"; "x-goog-user-project" = "project-id" }

Invoke-WebRequest `
-Method GET `
-Headers $headers `
-Uri "https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations" | Select-Object -Expand Content

您看到的输出会因您请求的操作而异。

您还可以使用选择查询参数(operationIddoneworksOn)过滤返回的操作。例如,如需返回已完成运行的操作列表,请修改网址:

GET https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations?filter="done=true"

Go

在试用此示例之前,请按照 API 与参考文档 > 客户端库页面上与此编程语言对应的设置说明进行操作。

import (
	"context"
	"fmt"
	"io"

	automl "cloud.google.com/go/automl/apiv1"
	longrunning "cloud.google.com/go/longrunning/autogen/longrunningpb"
	"google.golang.org/api/iterator"
)

// listOperationStatus lists existing operations' status.
func listOperationStatus(w io.Writer, projectID string, location string) error {
	// projectID := "my-project-id"
	// location := "us-central1"

	ctx := context.Background()
	client, err := automl.NewClient(ctx)
	if err != nil {
		return fmt.Errorf("NewClient: %w", err)
	}
	defer client.Close()

	req := &longrunning.ListOperationsRequest{
		Name: fmt.Sprintf("projects/%s/locations/%s", projectID, location),
	}

	it := client.LROClient.ListOperations(ctx, req)

	// Iterate over all results
	for {
		op, err := it.Next()
		if err == iterator.Done {
			break
		}
		if err != nil {
			return fmt.Errorf("ListOperations.Next: %w", err)
		}

		fmt.Fprintf(w, "Name: %v\n", op.GetName())
		fmt.Fprintf(w, "Operation details:\n")
		fmt.Fprintf(w, "%v", op)
	}

	return nil
}

Java

在试用此示例之前,请按照 API 与参考文档 > 客户端库页面上与此编程语言对应的设置说明进行操作。

import com.google.cloud.automl.v1.AutoMlClient;
import com.google.cloud.automl.v1.LocationName;
import com.google.longrunning.ListOperationsRequest;
import com.google.longrunning.Operation;
import java.io.IOException;

class ListOperationStatus {

  static void listOperationStatus() throws IOException {
    // TODO(developer): Replace these variables before running the sample.
    String projectId = "YOUR_PROJECT_ID";
    listOperationStatus(projectId);
  }

  // Get the status of an operation
  static void listOperationStatus(String projectId) throws IOException {
    // 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 (AutoMlClient client = AutoMlClient.create()) {
      // A resource that represents Google Cloud Platform location.
      LocationName projectLocation = LocationName.of(projectId, "us-central1");

      // Create list operations request.
      ListOperationsRequest listrequest =
          ListOperationsRequest.newBuilder().setName(projectLocation.toString()).build();

      // List all the operations names available in the region by applying filter.
      for (Operation operation :
          client.getOperationsClient().listOperations(listrequest).iterateAll()) {
        System.out.println("Operation details:");
        System.out.format("\tName: %s\n", operation.getName());
        System.out.format("\tMetadata Type Url: %s\n", operation.getMetadata().getTypeUrl());
        System.out.format("\tDone: %s\n", operation.getDone());
        if (operation.hasResponse()) {
          System.out.format("\tResponse Type Url: %s\n", operation.getResponse().getTypeUrl());
        }
        if (operation.hasError()) {
          System.out.println("\tResponse:");
          System.out.format("\t\tError code: %s\n", operation.getError().getCode());
          System.out.format("\t\tError message: %s\n\n", operation.getError().getMessage());
        }
      }
    }
  }
}

Node.js

在试用此示例之前,请按照 API 与参考文档 > 客户端库页面上与此编程语言对应的设置说明进行操作。

/**
 * TODO(developer): Uncomment these variables before running the sample.
 */
// const projectId = 'YOUR_PROJECT_ID';
// const location = 'us-central1';

// Imports the Google Cloud AutoML library
const {AutoMlClient} = require('@google-cloud/automl').v1;

// Instantiates a client
const client = new AutoMlClient();

async function listOperationStatus() {
  // Construct request
  const request = {
    name: client.locationPath(projectId, location),
    filter: `worksOn=projects/${projectId}/locations/${location}/models/*`,
  };

  const [response] = await client.operationsClient.listOperations(request);

  console.log('List of operation status:');
  for (const operation of response) {
    console.log(`Name: ${operation.name}`);
    console.log('Operation details:');
    console.log(`${operation}`);
  }
}

listOperationStatus();

Python

在试用此示例之前,请按照 API 与参考文档 > 客户端库页面上与此编程语言对应的设置说明进行操作。

from google.cloud import automl

# TODO(developer): Uncomment and set the following variables
# project_id = "YOUR_PROJECT_ID"

client = automl.AutoMlClient()
# A resource that represents Google Cloud Platform location.
project_location = f"projects/{project_id}/locations/us-central1"
# List all the operations names available in the region.
response = client._transport.operations_client.list_operations(
    name=project_location, filter_="", timeout=5
)

print("List of operations:")
for operation in response:
    print(f"Name: {operation.name}")
    print("Operation details:")
    print(operation)

其他语言

C#: 请按照客户端库页面上的 C# 设置说明操作,然后访问 .NET 版 AutoML Vision Object Detection 参考文档。

PHP: 请按照客户端库页面上的 PHP 设置说明操作,然后访问 PHP 版 AutoML Vision Object Detection 参考文档。

Ruby 版: 请按照客户端库页面上的 Ruby 设置说明操作,然后访问 Ruby 版 AutoML Vision Object Detection 参考文档。

处理长时间运行的操作

REST

在使用任何请求数据之前,请先进行以下替换:

  • project-id:您的 GCP 项目 ID。
  • operation-id:您的操作的 ID。此 ID 是操作名称的最后一个元素。例如:
    • 操作名称:projects/project-id/locations/location-id/operations/IOD5281059901324392598
    • 操作 ID:IOD5281059901324392598

HTTP 方法和网址:

GET https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations/OPERATION_ID

如需发送请求,请选择以下方式之一:

curl

执行以下命令:

curl -X GET \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "x-goog-user-project: project-id" \
"https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations/OPERATION_ID"

PowerShell

执行以下命令:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred"; "x-goog-user-project" = "project-id" }

Invoke-WebRequest `
-Method GET `
-Headers $headers `
-Uri "https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations/OPERATION_ID" | Select-Object -Expand Content
完成导入操作后,您应该会看到类似如下所示的输出:
{
  "name": "projects/PROJECT_ID/locations/us-central1/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.automl.v1.OperationMetadata",
    "createTime": "2018-10-29T15:56:29.176485Z",
    "updateTime": "2018-10-29T16:10:41.326614Z",
    "importDataDetails": {}
  },
  "done": true,
  "response": {
    "@type": "type.googleapis.com/google.protobuf.Empty"
  }
}

完成创建模型操作后,您应会看到如下输出:

{
  "name": "projects/PROJECT_ID/locations/us-central1/operations/OPERATION_ID",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.automl.v1.OperationMetadata",
    "createTime": "2019-07-22T18:35:06.881193Z",
    "updateTime": "2019-07-22T19:58:44.972235Z",
    "createModelDetails": {}
  },
  "done": true,
  "response": {
    "@type": "type.googleapis.com/google.cloud.automl.v1.Model",
    "name": "projects/PROJECT_ID/locations/us-central1/models/MODEL_ID"
  }
}

Go

在试用此示例之前,请按照 API 与参考文档 > 客户端库页面上与此编程语言对应的设置说明进行操作。

import (
	"context"
	"fmt"
	"io"

	automl "cloud.google.com/go/automl/apiv1"
	"cloud.google.com/go/automl/apiv1/automlpb"
)

// getOperationStatus gets an operation's status.
func getOperationStatus(w io.Writer, projectID string, location string, datasetID string, modelName string) error {
	// projectID := "my-project-id"
	// location := "us-central1"
	// datasetID := "ICN123456789..."
	// modelName := "model_display_name"

	ctx := context.Background()
	client, err := automl.NewClient(ctx)
	if err != nil {
		return fmt.Errorf("NewClient: %w", err)
	}
	defer client.Close()

	req := &automlpb.CreateModelRequest{
		Parent: fmt.Sprintf("projects/%s/locations/%s", projectID, location),
		Model: &automlpb.Model{
			DisplayName: modelName,
			DatasetId:   datasetID,
			ModelMetadata: &automlpb.Model_ImageClassificationModelMetadata{
				ImageClassificationModelMetadata: &automlpb.ImageClassificationModelMetadata{
					TrainBudgetMilliNodeHours: 1000, // 1000 milli-node hours are 1 hour
				},
			},
		},
	}

	op, err := client.CreateModel(ctx, req)
	if err != nil {
		return err
	}
	fmt.Fprintf(w, "Name: %v\n", op.Name())

	// Wait for the longrunning operation complete.
	resp, err := op.Wait(ctx)
	if err != nil && !op.Done() {
		fmt.Println("failed to fetch operation status", err)
		return err
	}
	if err != nil && op.Done() {
		fmt.Println("operation completed with error", err)
		return err
	}
	fmt.Fprintf(w, "Response: %v\n", resp)

	return nil
}

Java

在试用此示例之前,请按照 API 与参考文档 > 客户端库页面上与此编程语言对应的设置说明进行操作。

import com.google.cloud.automl.v1.AutoMlClient;
import com.google.longrunning.Operation;
import java.io.IOException;

class GetOperationStatus {

  static void getOperationStatus() throws IOException {
    // TODO(developer): Replace these variables before running the sample.
    String operationFullId = "projects/[projectId]/locations/us-central1/operations/[operationId]";
    getOperationStatus(operationFullId);
  }

  // Get the status of an operation
  static void getOperationStatus(String operationFullId) throws IOException {
    // 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 (AutoMlClient client = AutoMlClient.create()) {
      // Get the latest state of a long-running operation.
      Operation operation = client.getOperationsClient().getOperation(operationFullId);

      // Display operation details.
      System.out.println("Operation details:");
      System.out.format("\tName: %s\n", operation.getName());
      System.out.format("\tMetadata Type Url: %s\n", operation.getMetadata().getTypeUrl());
      System.out.format("\tDone: %s\n", operation.getDone());
      if (operation.hasResponse()) {
        System.out.format("\tResponse Type Url: %s\n", operation.getResponse().getTypeUrl());
      }
      if (operation.hasError()) {
        System.out.println("\tResponse:");
        System.out.format("\t\tError code: %s\n", operation.getError().getCode());
        System.out.format("\t\tError message: %s\n", operation.getError().getMessage());
      }
    }
  }
}

Node.js

在试用此示例之前,请按照 API 与参考文档 > 客户端库页面上与此编程语言对应的设置说明进行操作。

/**
 * TODO(developer): Uncomment these variables before running the sample.
 */
// const projectId = 'YOUR_PROJECT_ID';
// const location = 'us-central1';
// const operationId = 'YOUR_OPERATION_ID';

// Imports the Google Cloud AutoML library
const {AutoMlClient} = require('@google-cloud/automl').v1;

// Instantiates a client
const client = new AutoMlClient();

async function getOperationStatus() {
  // Construct request
  const request = {
    name: `projects/${projectId}/locations/${location}/operations/${operationId}`,
  };

  const [response] = await client.operationsClient.getOperation(request);

  console.log(`Name: ${response.name}`);
  console.log('Operation details:');
  console.log(`${response}`);
}

getOperationStatus();

Python

在试用此示例之前,请按照 API 与参考文档 > 客户端库页面上与此编程语言对应的设置说明进行操作。

from google.cloud import automl

# TODO(developer): Uncomment and set the following variables
# operation_full_id = \
#     "projects/[projectId]/locations/us-central1/operations/[operationId]"

client = automl.AutoMlClient()
# Get the latest state of a long-running operation.
response = client._transport.operations_client.get_operation(operation_full_id)

print(f"Name: {response.name}")
print("Operation details:")
print(response)

其他语言

C#: 请按照客户端库页面上的 C# 设置说明操作,然后访问 .NET 版 AutoML Vision 参考文档。

PHP: 请按照客户端库页面上的 PHP 设置说明操作,然后访问 PHP 版 AutoML Vision 参考文档。

Ruby 版: 请按照客户端库页面上的 Ruby 设置说明操作,然后访问 Ruby 版 AutoML Vision 参考文档。

取消操作

您可以使用操作 ID 取消导入任务或训练任务。

REST

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

在使用任何请求数据之前,请先进行以下替换:

  • project-id:您的 GCP 项目 ID。
  • operation-id:您的操作的 ID。此 ID 是操作名称的最后一个元素。例如:
    • 操作名称:projects/project-id/locations/location-id/operations/IOD5281059901324392598
    • 操作 ID:IOD5281059901324392598

HTTP 方法和网址:

POST https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations/OPERATION_ID:cancel

如需发送请求,请选择以下方式之一:

curl

执行以下命令:

curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "x-goog-user-project: project-id" \
-H "Content-Type: application/json; charset=utf-8" \
-d "" \
"https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations/OPERATION_ID:cancel"

PowerShell

执行以下命令:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred"; "x-goog-user-project" = "project-id" }

Invoke-WebRequest `
-Method POST `
-Headers $headers `
-Uri "https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/operations/OPERATION_ID:cancel" | Select-Object -Expand Content
如果请求成功,则将返回空的 JSON 对象:
{}

获取模型的相关信息

使用以下代码示例获取有关特定的训练后模型的信息。您可以使用此请求返回的信息来修改模式或发送预测请求。

REST

在使用任何请求数据之前,请先进行以下替换:

  • project-id:您的 GCP 项目 ID。
  • model-id:您的模型的 ID(从创建模型时返回的响应中获取)。此 ID 是模型名称的最后一个元素。 例如:
    • 模型名称:projects/project-id/locations/location-id/models/IOD4412217016962778756
    • 模型 ID:IOD4412217016962778756

HTTP 方法和网址:

GET https://automl.googleapis.com/v1/projects/project-id/locations/us-central1/models/model-id

如需发送请求,请选择以下方式之一:

curl

执行以下命令:

curl -X GET \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "x-goog-user-project: project-id" \
"https://automl.googleapis.com/v1/projects/project-id/locations/us-central1/models/model-id"

PowerShell

执行以下命令:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred"; "x-goog-user-project" = "project-id" }

Invoke-WebRequest `
-Method GET `
-Headers $headers `
-Uri "https://automl.googleapis.com/v1/projects/project-id/locations/us-central1/models/model-id" | Select-Object -Expand Content

您应该收到类似以下内容的 JSON 响应:



    {
  "name": "projects/project-id/locations/us-central1/models/model-id",
  "displayName": "display-name",
  "datasetId": "dataset-id",
  "createTime": "2019-10-30T20:06:08.253243Z",
  "deploymentState": "UNDEPLOYED",
  "updateTime": "2019-10-30T20:54:50.472328Z",
  "imageClassificationModelMetadata": {
    "trainBudget": "1",
    "modelType": "mobile-low-latency-1",
    "nodeQps": 3.2
  }
}

Go

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

import (
	"context"
	"fmt"
	"io"

	automl "cloud.google.com/go/automl/apiv1"
	"cloud.google.com/go/automl/apiv1/automlpb"
)

// getModel gets a model.
func getModel(w io.Writer, projectID string, location string, modelID string) error {
	// projectID := "my-project-id"
	// location := "us-central1"
	// modelID := "TRL123456789..."

	ctx := context.Background()
	client, err := automl.NewClient(ctx)
	if err != nil {
		return fmt.Errorf("NewClient: %w", err)
	}
	defer client.Close()

	req := &automlpb.GetModelRequest{
		Name: fmt.Sprintf("projects/%s/locations/%s/models/%s", projectID, location, modelID),
	}

	model, err := client.GetModel(ctx, req)
	if err != nil {
		return fmt.Errorf("GetModel: %w", err)
	}

	// Retrieve deployment state.
	deploymentState := "undeployed"
	if model.GetDeploymentState() == automlpb.Model_DEPLOYED {
		deploymentState = "deployed"
	}

	// Display the model information.
	fmt.Fprintf(w, "Model name: %v\n", model.GetName())
	fmt.Fprintf(w, "Model display name: %v\n", model.GetDisplayName())
	fmt.Fprintf(w, "Model create time:\n")
	fmt.Fprintf(w, "\tseconds: %v\n", model.GetCreateTime().GetSeconds())
	fmt.Fprintf(w, "\tnanos: %v\n", model.GetCreateTime().GetNanos())
	fmt.Fprintf(w, "Model deployment state: %v\n", deploymentState)

	return nil
}

Java

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

import com.google.cloud.automl.v1.AutoMlClient;
import com.google.cloud.automl.v1.Model;
import com.google.cloud.automl.v1.ModelName;
import java.io.IOException;

class GetModel {

  static void getModel() throws IOException {
    // TODO(developer): Replace these variables before running the sample.
    String projectId = "YOUR_PROJECT_ID";
    String modelId = "YOUR_MODEL_ID";
    getModel(projectId, modelId);
  }

  // Get a model
  static void getModel(String projectId, String modelId) throws IOException {
    // 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 (AutoMlClient client = AutoMlClient.create()) {
      // Get the full path of the model.
      ModelName modelFullId = ModelName.of(projectId, "us-central1", modelId);
      Model model = client.getModel(modelFullId);

      // Display the model information.
      System.out.format("Model name: %s\n", model.getName());
      // To get the model id, you have to parse it out of the `name` field. As models Ids are
      // required for other methods.
      // Name Format: `projects/{project_id}/locations/{location_id}/models/{model_id}`
      String[] names = model.getName().split("/");
      String retrievedModelId = names[names.length - 1];
      System.out.format("Model id: %s\n", retrievedModelId);
      System.out.format("Model display name: %s\n", model.getDisplayName());
      System.out.println("Model create time:");
      System.out.format("\tseconds: %s\n", model.getCreateTime().getSeconds());
      System.out.format("\tnanos: %s\n", model.getCreateTime().getNanos());
      System.out.format("Model deployment state: %s\n", model.getDeploymentState());
    }
  }
}

Node.js

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

/**
 * TODO(developer): Uncomment these variables before running the sample.
 */
// const projectId = 'YOUR_PROJECT_ID';
// const location = 'us-central1';
// const modelId = 'YOUR_MODEL_ID';

// Imports the Google Cloud AutoML library
const {AutoMlClient} = require('@google-cloud/automl').v1;

// Instantiates a client
const client = new AutoMlClient();

async function getModel() {
  // Construct request
  const request = {
    name: client.modelPath(projectId, location, modelId),
  };

  const [response] = await client.getModel(request);

  console.log(`Model name: ${response.name}`);
  console.log(
    `Model id: ${
      response.name.split('/')[response.name.split('/').length - 1]
    }`
  );
  console.log(`Model display name: ${response.displayName}`);
  console.log('Model create time');
  console.log(`\tseconds ${response.createTime.seconds}`);
  console.log(`\tnanos ${response.createTime.nanos / 1e9}`);
  console.log(`Model deployment state: ${response.deploymentState}`);
}

getModel();

Python

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

from google.cloud import automl

# TODO(developer): Uncomment and set the following variables
# project_id = "YOUR_PROJECT_ID"
# model_id = "YOUR_MODEL_ID"

client = automl.AutoMlClient()
# Get the full path of the model.
model_full_id = client.model_path(project_id, "us-central1", model_id)
model = client.get_model(name=model_full_id)

# Retrieve deployment state.
if model.deployment_state == automl.Model.DeploymentState.DEPLOYED:
    deployment_state = "deployed"
else:
    deployment_state = "undeployed"

# Display the model information.
print(f"Model name: {model.name}")
print("Model id: {}".format(model.name.split("/")[-1]))
print(f"Model display name: {model.display_name}")
print(f"Model create time: {model.create_time}")
print(f"Model deployment state: {deployment_state}")

其他语言

C#: 请按照客户端库页面上的 C# 设置说明操作,然后访问 .NET 版 AutoML Vision Object Detection 参考文档。

PHP: 请按照客户端库页面上的 PHP 设置说明操作,然后访问 PHP 版 AutoML Vision Object Detection 参考文档。

Ruby 版: 请按照客户端库页面上的 Ruby 设置说明操作,然后访问 Ruby 版 AutoML Vision Object Detection 参考文档。

可恢复训练

您现在可以暂停并在后续恢复大型数据集(包含一千张以上的图片)的自定义模型训练。暂停模型训练不是涉及暂停按钮的人工处置措施,而是用户将节点时预算设置为指定时间,例如 2 小时。达到该时间后,无论模型状态如何,训练都会自动停止。

后续训练具有以下限制:

  • 基本模型时间限制 - 您只能对最近 14 天内训练过的模型恢复训练;在请求发出 14 天之前创建的基本模型不符合后续训练的条件。
  • 不可修改标签 - 如果您更改基本模型数据集中的标签,则后续训练将失败。
  • 无法保证更好的性能 - 对模型进行后续训练并不能保证获得更好的模型性能。

网页界面

在试用此示例之前,请按照客户端库页面中与此编程语言对应的设置说明执行操作。

  1. 打开 AutoML Vision UI

    数据集页面显示当前项目的可用数据集。

    数据集列表页面
  2. 选择要用于训练自定义模型的数据集。

    所选数据集的显示名会显示在标题栏中,该页面还会列出数据集中的各个训练项及其标签。

    图片项页面
  3. 查看完数据集后,选择标题栏正下方的训练标签页。

    如果您以前训练过模型,则它们将显示在此标签页的列表中。

    已训练的模型列表
  4. 如果可以对某个模型进行可恢复的训练,则该训练将在查看完整评估选项旁显示为一个选项。

    选择恢复训练,对适用模型恢复训练。

    “继续训练”选项
  5. 与训练原始模型类似,选择继续训练将打开“训练新模型”侧边窗口。在定义模型部分中,您可以更改新模型的名称。您还将看到用于训练此模型的基本模型的名称。

    “定义模型”侧边窗口

    定义好模型后,选择继续以转到下一部分。

  6. 在下面的设置节点时预算部分中,指定节点时预算。与初始模型训练类似,此预算受价格结构和配额限制的约束。

    指定训练预算后,选择开始训练。完成可恢复的训练后,您会收到一封电子邮件。

REST

在使用任何请求数据之前,请先进行以下替换:

  • base-model-id:基本(现有)模型的 ID。新模型基于这个基本模型创建。以下新模型规范必须与基本模型匹配:

    * 即使新模型请求中省略了这个字段,系统也会检查基本和新的 modelType

  • project-id:您的 GCP 项目 ID。
  • display-name:您选择的字符串显示名。
  • dataset-id:您的数据集的 ID。此 ID 是数据集名称的最后一个元素。例如:
    • 数据集名称:projects/project-id/locations/location-id/datasets/3104518874390609379
    • 数据集 ID:3104518874390609379

HTTP 方法和网址:

POST https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/models

请求 JSON 正文:

{
  "baseModelId": BASE_MODEL_ID,
  "displayName": "DISPLAY_NAME",
  "datasetId": "DATASET_ID",
  "imageClassificationModelMetadata": {
    "trainBudget": "1",
    "modelType": "mobile-low-latency-1"
  }
}

如需发送请求,请选择以下方式之一:

curl

将请求正文保存在名为 request.json 的文件中,然后执行以下命令:

curl -X POST \
-H "Authorization: Bearer $(gcloud auth print-access-token)" \
-H "x-goog-user-project: project-id" \
-H "Content-Type: application/json; charset=utf-8" \
-d @request.json \
"https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/models"

PowerShell

将请求正文保存在名为 request.json 的文件中,然后执行以下命令:

$cred = gcloud auth print-access-token
$headers = @{ "Authorization" = "Bearer $cred"; "x-goog-user-project" = "project-id" }

Invoke-WebRequest `
-Method POST `
-Headers $headers `
-ContentType: "application/json; charset=utf-8" `
-InFile request.json `
-Uri "https://automl.googleapis.com/v1/projects/PROJECT_ID/locations/us-central1/models" | Select-Object -Expand Content

您应该会看到类似如下所示的输出。可以使用操作 ID(本例中为 ICN2106290444865378475)来获取任务的状态。如需查看示例,请参阅处理长时间运行的操作

{
  "name": "projects/PROJECT_ID/locations/us-central1/operations/ICN2106290444865378475",
  "metadata": {
    "@type": "type.googleapis.com/google.cloud.automl.v1.OperationMetadata",
    "createTime": "2019-10-30T20:06:08.253243Z",
    "updateTime": "2019-10-30T20:06:08.253243Z",
    "createModelDetails": {}
  }
}