在 Android 上使用機器學習套件擷取實體

如要分析一段文字並擷取其中的實體，請叫用 annotate() 方法，並傳遞文字字串或 EntityExtractionParams 例項，藉此指定參照時間、時區或篩選條件等其他選項，以限制搜尋一部分的實體類型。API 會傳回 EntityAnnotation 物件清單，其中包含每個實體的相關資訊。

立即體驗

• 請試用範例應用程式，查看這個 API 的使用範例。

事前準備

1. 在專案層級的 build.gradle 檔案中，請確認 Google 的 Maven 存放區包含在 buildscript 和 allprojects 區段中。

2. 將 ML Kit 實體擷取程式庫的依附元件新增至模組的應用程式層級 Gradle 檔案 (通常名為 app/build.gradle)：

   dependencies {
   // …
   
   implementation 'com.google.mlkit:entity-extraction:16.0.0-beta4'
   }
   

擷取實體

建立 EntityExtractor 物件，並使用 EntityExtractorOptions 進行設定

val entityExtractor =
   EntityExtraction.getClient(
       EntityExtractorOptions.Builder(EntityExtractorOptions.ENGLISH)
           .build())

EntityExtractor entityExtractor =
        EntityExtraction.getClient(
            new EntityExtractorOptions.Builder(EntityExtractorOptions.ENGLISH)
                .build());

如果需要，EntityExtractorOptions 也會接受使用者定義的 Executor，否則機器學習套件會使用預設的 Executor

確認必要的模型已下載到裝置上。

entityExtractor
  .downloadModelIfNeeded()
  .addOnSuccessListener { _ ->
    /* Model downloading succeeded, you can call extraction API here. */
  }
  .addOnFailureListener { _ -> /* Model downloading failed. */ }

entityExtractor
    .downloadModelIfNeeded()
    .addOnSuccessListener(
        aVoid -> {
          // Model downloading succeeded, you can call the extraction API here. 
        })
    .addOnFailureListener(
        exception -> {
          // Model downloading failed.
        });

確認下載模型後，將字串或 EntityExtractionParams 傳遞至 annotate()。除非知道模型可用，否則請勿呼叫 annotate()。

val params =
      EntityExtractionParams.Builder("My flight is LX373, please pick me up at 8am tomorrow.")
        .setEntityTypesFilter((/* optional entity type filter */)
        .setPreferredLocale(/* optional preferred locale */)
        .setReferenceTime(/* optional reference date-time */)
        .setReferenceTimeZone(/* optional reference timezone */)
        .build()
entityExtractor
      .annotate(params)
      .addOnSuccessListener {
        // Annotation process was successful, you can parse the EntityAnnotations list here.
      }
      .addOnFailureListener {
        // Check failure message here.
      }

EntityExtractionParams params = new EntityExtractionParams
        .Builder("My flight is LX373, please pick me up at 8am tomorrow.")
        .setEntityTypesFilter(/* optional entity type filter */)
        .setPreferredLocale(/* optional preferred locale */)
        .setReferenceTime(/* optional reference date-time */)
        .setReferenceTimeZone(/* optional reference timezone */)
        .build();
entityExtractor
        .annotate(params)
        .addOnSuccessListener(new OnSuccessListener<List<EntityAnnotation>>() {
          @Override
          public void onSuccess(List<EntityAnnotation> entityAnnotations) {
            // Annotation process was successful, you can parse the EntityAnnotations list here.
          }
        })
        .addOnFailureListener(new OnFailureListener() {
          @Override
          public void onFailure(@NonNull Exception e) {
            // Check failure message here.
          }
        });

PreferredLocale、ReferenceTime 和 ReferenceTimeZone 只會影響 DateTime 實體。如果未明確設定，則預設值為使用者裝置中的值。

循環註解註解以擷取可辨識實體的相關資訊。

for (entityAnnotation in entityAnnotations) {
  val entities: List<Entity> = entityAnnotation.entities

  Log.d(TAG, "Range: ${entityAnnotation.start} - ${entityAnnotation.end}")
  for (entity in entities) {
    when (entity) {
      is DateTimeEntity -> {
        Log.d(TAG, "Granularity: ${entity.dateTimeGranularity}")
        Log.d(TAG, "TimeStamp: ${entity.timestampMillis}")
      }
      is FlightNumberEntity -> {
        Log.d(TAG, "Airline Code: ${entity.airlineCode}")
        Log.d(TAG, "Flight number: ${entity.flightNumber}")
      }
      is MoneyEntity -> {
        Log.d(TAG, "Currency: ${entity.unnormalizedCurrency}")
        Log.d(TAG, "Integer part: ${entity.integerPart}")
        Log.d(TAG, "Fractional Part: ${entity.fractionalPart}")
      }
      else -> {
        Log.d(TAG, "  $entity")
      }
    }
  }
}

List<EntityAnnotation> entityAnnotations = /* Get from EntityExtractor */;
for (EntityAnnotation entityAnnotation : entityAnnotations) {
  List<Entity> entities = entityAnnotation.getEntities();

  Log.d(TAG, String.format("Range: [%d, %d)", entityAnnotation.getStart(), entityAnnotation.getEnd()));
  for (Entity entity : entities) {
    switch (entity.getType()) {
      case Entity.TYPE_DATE_TIME:
        DateTimeEntity dateTimeEntity = entity.asDateTimeEntity();
        Log.d(TAG, "Granularity: " + dateTimeEntity.getDateTimeGranularity());
        Log.d(TAG, "Timestamp: " + dateTimeEntity.getTimestampMillis());
      case Entity.TYPE_FLIGHT_NUMBER:
        FlightNumberEntity flightNumberEntity = entity.asFlightNumberEntity();
        Log.d(TAG, "Airline Code: " + flightNumberEntity.getAirlineCode());
        Log.d(TAG, "Flight number: " + flightNumberEntity.getFlightNumber());
      case Entity.TYPE_MONEY:
        MoneyEntity moneyEntity = entity.asMoneyEntity();
        Log.d(TAG, "Currency: " + moneyEntity.getUnnormalizedCurrency());
        Log.d(TAG, "Integer Part: " + moneyEntity.getIntegerPart());
        Log.d(TAG, "Fractional Part: " + moneyEntity.getFractionalPart());
      case Entity.TYPE_UNKNOWN:
      default:
        Log.d(TAG, "Entity: " + entity);
    }
  }
}

如果不再需要 EntityExtractor 物件，請呼叫 close() 方法。如果您在 Fragment 或 AppCompatActivity 中使用 EntityExtractor，您可以在 Fragment 或 AppCompatActivity 中呼叫 LifecycleOwner.getLifecycle()，然後呼叫 Lifecycle.addObserver。例如：

val options = …
val extractor = EntityExtraction.getClient(options);
getLifecycle().addObserver(extractor);

EntityExtractorOptions options = …
EntityExtractor extractor = EntityExtraction.getClient(options);
getLifecycle().addObserver(extractor);

明確管理實體擷取模型

如上所述，當您使用實體擷取 API 時，機器學習套件會視需要自動將語言專用的模型下載到裝置上 (當您呼叫 downloadModelIfNeeded() 時)。您也可以使用 ML Kit 的模型管理 API，明確管理要在裝置中使用的模型。如果您想事先下載模型，這個方法就很實用。這個 API 也可讓您刪除不再需要的模型。

val modelManager = RemoteModelManager.getInstance()

// Get entity extraction models stored on the device.
modelManager.getDownloadedModels(EntityExtractionRemoteModel::class.java)
  .addOnSuccessListener {
    // ...
  }
  .addOnFailureListener({
    // Error.
  })
    
// Delete the German model if it's on the device.
val germanModel =
  EntityExtractionRemoteModel.Builder(EntityExtractorOptions.GERMAN).build()
modelManager.deleteDownloadedModel(germanModel)
  .addOnSuccessListener({
    // Model deleted.
  })
  .addOnFailureListener({
    // Error.
  })
    
// Download the French model.
val frenchModel =
  EntityExtractionRemoteModel.Builder(EntityExtractorOptions.FRENCH).build()
val conditions =
  DownloadConditions.Builder()
    .requireWifi()
    .build()
modelManager.download(frenchModel, conditions)
  .addOnSuccessListener({
    // Model downloaded.
  })
  .addOnFailureListener({
    // Error.
  })

// Get entity extraction models stored on the device.
modelManager.getDownloadedModels(EntityExtractionRemoteModel.class)
    .addOnSuccessListener(new OnSuccessListener<Set<EntityExtractionRemoteModel>>() {
      @Override
      public void onSuccess(Set<EntityExtractionRemoteModel> models) {
        // ...
      }
    })
    .addOnFailureListener(new OnFailureListener() {
      @Override
      public void onFailure(@NonNull Exception e) {
        // Error.
      }
    });

// Delete the German model if it's on the device.
EntityExtractionRemoteModel germanModel = new EntityExtractionRemoteModel.Builder(EntityExtractorOptions.GERMAN).build();
    modelManager.deleteDownloadedModel(germanModel)
        .addOnSuccessListener(new OnSuccessListener<Void>() {
          @Override
          public void onSuccess(Void v) {
            // Model deleted.
          }
        })
        .addOnFailureListener(new OnFailureListener() {
          @Override
          public void onFailure(@NonNull Exception e) {
            // Error.
          }
        });

// Download the French model.
EntityExtractionRemoteModel frenchModel = new EntityExtractionRemoteModel.Builder(EntityExtractorOptions.FRENCH).build();
    DownloadConditions conditions = new DownloadConditions.Builder()
        .requireWifi()
        .build();
    modelManager.download(frenchModel, conditions)
        .addOnSuccessListener(new OnSuccessListener<Void>() {
          @Override
          public void onSuccess(Void v) {
            // Model downloaded.
          }
        })
        .addOnFailureListener(new OnFailureListener() {
          @Override
          public void onFailure(@NonNull Exception e) {
            // Error.
          }
        });