Java 函数如何简化人工智能开发，降低工程成本？

import com.google.api.gax.batching.BatchingSettings;
import com.google.cloud.functions.BackgroundFunction;
import com.google.cloud.functions.Context;
import functions.eventpojos.PubsubMessage;
import java.nio.charset.StandardCharsets;
import java.util.Base64;
import java.util.Collections;
import java.util.logging.Logger;
import org.tensorflow.TensorFlow;

public class ImageClassifier implements BackgroundFunction<PubsubMessage> {

  // 预加载 TensorFlow 模型
  private static final Logger logger = Logger.getLogger(ImageClassifier.class.getName());
  public static final TensorFlow MODEL = TensorFlow.model();
  public static final BatchingSettings BATCHING_SETTINGS =
      BatchingSettings.newBuilder().setElementCountThreshold(100).build();

  @Override
  public void accept(PubsubMessage message, Context context) {
    try {
      final String body = message.getData().toString(StandardCharsets.UTF_8);
      final String[] parts = body.split(",");
      final String image = new String(Base64.getDecoder().decode(parts[0]));
      final long prediction = MODEL.execute(image, "serving_default");
      logger.info("Prediction: " + prediction);
    } catch (Exception e) {
      logger.severe(e.getMessage());
      throw new FunctionsException("Failed to classify image", e);
    }
  }
}

runtime: java11
env_variables:
  TF_CPP_MIN_LOG_LEVEL: 3 # 抑制 TensorFlow 日志