I was using djl version 0.25, if I upgrade to version 0.32, do I need to change the method?

[phyjar]

this is my POM

<!-- DJL API -->
        <dependency>
            <groupId>ai.djl</groupId>
            <artifactId>api</artifactId>
            <version>${djl.version}</version>
        </dependency>

        <!-- DJL PyTorch Engine -->
        <dependency>
            <groupId>ai.djl.pytorch</groupId>
            <artifactId>pytorch-engine</artifactId>
            <version>${djl.version}</version>
        </dependency>

        <!-- PyTorch Native Library -->
        <dependency>
            <groupId>ai.djl.pytorch</groupId>
            <artifactId>pytorch-native-cpu</artifactId>
            <classifier>win-x86_64</classifier>
            <version>2.0.1</version>
            <scope>runtime</scope>
        </dependency>

        <!-- DJL PyTorch Model Zoo -->
        <dependency>
            <groupId>ai.djl.pytorch</groupId>
            <artifactId>pytorch-model-zoo</artifactId>
            <version>${djl.version}</version>
        </dependency>

This is my code

public void init() {
        try {            
            String modelPath = Paths.get(modelBasePath, "fire.torchscript.pt").toString();
            String synsetPath = Paths.get(modelBasePath, "fire.txt").toString();
            
            logger.info("正在加载模型，路径: {}", modelPath);
            File modelFile = new File(modelPath);
            if (!modelFile.exists()) {
                throw new RuntimeException("模型文件不存在: " + modelPath);
            }

            Translator<Image, DetectedObjects> translator = YoloV5Translator.builder()
                    .addTransform(new Resize(640, 480))
                    .addTransform(new ToTensor())
                    .optSynsetArtifactName(synsetPath)
                    .build();

            Criteria<Image, DetectedObjects> criteria = Criteria.builder()
                    .setTypes(Image.class, DetectedObjects.class)
                    .optModelPath(Paths.get(modelPath))
                    .optTranslator(translator)
                    .build();

            model = ModelZoo.loadModel(criteria);
            logger.info("模型加载成功");
            
        } catch (Exception e) {
            logger.error("无法初始化对象检测服务: {}", e.getMessage(), e);
            throw new RuntimeException("Failed to initialize object detection service", e);
        }
    }
    
    public DetectedObjects detectObjects(Image image) throws Exception {
        try (Predictor<Image, DetectedObjects> predictor = model.newPredictor()) {
            return predictor.predict(image);
        }
    }

I'm using windows, my cuda version is 12.4, and I see that the documentation supports gpu acceleration, what should I do? It's not clear enough for me to view the document

<dependency>
    <groupId>ai.djl.pytorch</groupId>
    <artifactId>pytorch-native-cu124</artifactId>
    <classifier>win-x86_64</classifier>
    <version>2.4.0</version>
    <scope>runtime</scope>
</dependency>
<dependency>
<groupId>ai.djl.pytorch</groupId>
<artifactId>pytorch-jni</artifactId>
<version>2.4.0-0.30.0</version>
<scope>runtime</scope>
</dependency>