【Elasticsearch】-图片向量化存储

需要结合深度学习模型

1、pom依赖

注意结尾的webp-imageio 包，用于解决ImageIO.read读取部分图片返回为null的问题


        <dependency>
            <groupId>org.openpnp</groupId>
            <artifactId>opencv</artifactId>
            <version>4.7.0-0</version>
        </dependency>

        <dependency>
            <groupId>com.microsoft.onnxruntime</groupId>
            <artifactId>onnxruntime</artifactId>
            <version>1.17.1</version>
        </dependency>

        <!-- 服务器端推理引擎 -->
        <dependency>
            <groupId>ai.djl</groupId>
            <artifactId>api</artifactId>
            <version>${djl.version}</version>
        </dependency>
        <dependency>
            <groupId>ai.djl</groupId>
            <artifactId>basicdataset</artifactId>
            <version>${djl.version}</version>
        </dependency>
        <dependency>
            <groupId>ai.djl</groupId>
            <artifactId>model-zoo</artifactId>
            <version>${djl.version}</version>
        </dependency>
        <!-- Pytorch -->
        <dependency>
            <groupId>ai.djl.pytorch</groupId>
            <artifactId>pytorch-engine</artifactId>
            <version>${djl.version}</version>
        </dependency>
        <dependency>
            <groupId>ai.djl.pytorch</groupId>
            <artifactId>pytorch-model-zoo</artifactId>
            <version>${djl.version}</version>
        </dependency>
        <!-- ONNX -->
        <dependency>
            <groupId>ai.djl.onnxruntime</groupId>
            <artifactId>onnxruntime-engine</artifactId>
            <version>${djl.version}</version>
        </dependency>


        <!-- 解决ImageIO.read 读取为null -->
        <dependency>
            <groupId>org.sejda.imageio</groupId>
            <artifactId>webp-imageio</artifactId>
            <version>0.1.6</version>
        </dependency>

2、加载模型

注意提前设置环境变量，pytorch依赖环境dll文件，如果不存在，则默认下载

System.setProperty("ENGINE_CACHE_DIR", modelPath);

import ai.djl.Device;
import ai.djl.modality.cv.Image;
import ai.djl.repository.zoo.Criteria;
import ai.djl.training.util.ProgressBar;
import ai.djl.translate.Translator;

 public Criteria<Image, T> criteria() {
        Translator<Image, T> translator = getTranslator(arguments);
        try {
            JarFileUtils.copyFileFromJar("/onnx/models/" + modelName, PathConstants.ONNX, null, false, true);
        } catch (IOException e) {
            throw new RuntimeException(e);
        }
//        String model_path = PathConstants.TEMP_DIR + PathConstants.ONNX + "/" + modelName;
        String modelPath = PathConstants.TEMP_DIR + File.separator+PathConstants.ONNX_NAME+ File.separator + modelName;
        log.info("路径修改前:{}",modelPath);
        modelPath= DjlHandlerUtil.getFixedModelPath(modelPath);
        log.info("路径修改后:{}",modelPath);
        Criteria<Image, T> criteria =
                Criteria.builder()
                        .setTypes(Image.class, getClassOfT())
                        .optModelUrls(modelPath)
                        .optTranslator(translator)
                        .optDevice(Device.cpu())
                        .optEngine(getEngine()) // Use PyTorch engine
                        .optProgress(new ProgressBar())
                        .build();
        return criteria;
    }


protected Translator<Image, float[]> getTranslator(Map<String, Object> arguments) {
        BaseImageTranslator.BaseBuilder<?> builder=new BaseImageTranslator.BaseBuilder<BaseImageTranslator.BaseBuilder>() {
            @Override
            protected BaseImageTranslator.BaseBuilder self() {
                return this;
            }
        };
        return new BaseImageTranslator<float[]>(builder) {
            @Override
            public float[] processOutput(TranslatorContext translatorContext, NDList ndList) throws Exception {
                return ndList.get(0).toFloatArray();

            }
        };
    }

3、FaceFeatureTranslator



import ai.djl.modality.cv.Image;
import ai.djl.modality.cv.transform.Normalize;
import ai.djl.modality.cv.transform.ToTensor;
import ai.djl.ndarray.NDArray;
import ai.djl.ndarray.NDList;
import ai.djl.translate.Batchifier;
import ai.djl.translate.Pipeline;
import ai.djl.translate.Translator;
import ai.djl.translate.TranslatorContext;

/**
 * @author gc.x
 * @date 2022-04
 */
public final class FaceFeatureTranslator implements Translator<Image, float[]> {

    public FaceFeatureTranslator() {
    }

    @Override
    public NDList processInput(TranslatorContext ctx, Image input) {
        NDArray array = input.toNDArray(ctx.getNDManager(), Image.Flag.COLOR);
        Pipeline pipeline = new Pipeline();
        pipeline
                // .add(new Resize(160))
                .add(new ToTensor())
                .add(
                        new Normalize(
                                new float[]{127.5f / 255.0f, 127.5f / 255.0f, 127.5f / 255.0f},
                                new float[]{128.0f / 255.0f, 128.0f / 255.0f, 128.0f / 255.0f}));

        return pipeline.transform(new NDList(array));
    }

    @Override
    public float[] processOutput(TranslatorContext ctx, NDList list) {
        NDList result = new NDList();
        long numOutputs = list.singletonOrThrow().getShape().get(0);
        for (int i = 0; i < numOutputs; i++) {
            result.add(list.singletonOrThrow().get(i));
        }
        float[][] embeddings = result.stream().map(NDArray::toFloatArray).toArray(float[][]::new);
        float[] feature = new float[embeddings.length];
        for (int i = 0; i < embeddings.length; i++) {
            feature[i] = embeddings[i][0];
        }
        return feature;
    }

    @Override
    public Batchifier getBatchifier() {
        return Batchifier.STACK;
    }
}

4、BaseImageTranslator


import ai.djl.Model;
import ai.djl.modality.cv.Image;
import ai.djl.modality.cv.transform.CenterCrop;
import ai.djl.modality.cv.transform.Normalize;
import ai.djl.modality.cv.transform.Resize;
import ai.djl.modality.cv.transform.ToTensor;
import ai.djl.modality.cv.util.NDImageUtils;
import ai.djl.ndarray.NDArray;
import ai.djl.ndarray.NDList;
import ai.djl.translate.*;
import ai.djl.util.Utils;

import java.io.IOException;
import java.io.InputStream;
import java.net.MalformedURLException;
import java.net.URL;
import java.util.Arrays;
import java.util.List;
import java.util.Map;


public abstract class BaseImageTranslator<T> implements Translator<Image, T> {

    private static final float[] MEAN = {0.485f, 0.456f, 0.406f};
    private static final float[] STD = {0.229f, 0.224f, 0.225f};

    private Image.Flag flag;
    private Pipeline pipeline;
    private Batchifier batchifier;

    /**
     * Constructs an ImageTranslator with the provided builder.
     *
     * @param builder the data to build with
     */
    public BaseImageTranslator(BaseBuilder<?> builder) {
        flag = builder.flag;
        pipeline = builder.pipeline;
        batchifier = builder.batchifier;
    }

    /** {@inheritDoc} */
    @Override
    public Batchifier getBatchifier() {
        return batchifier;
    }

    /**
     * Processes the {@link Image} input and converts it to NDList.
     *
     * @param ctx the toolkit that helps create the input NDArray
     * @param input the {@link Image} input
     * @return a {@link NDList}
     */
    @Override
    public NDList processInput(TranslatorContext ctx, Image input) {
        NDArray array = input.toNDArray(ctx.getNDManager(), flag);
        array = NDImageUtils.resize(array, 640, 640);
        array = array.transpose(2, 0, 1); // HWC -> CHW RGB -> BGR
//        array = array.expandDims(0);
        array = array.div(255f);
        return new NDList(array);
//        return pipeline.transform(new NDList(array));
    }

    protected static String getStringValue(Map<String, ?> arguments, String key, String def) {
        Object value = arguments.get(key);
        if (value == null) {
            return def;
        }
        return value.toString();
    }

    protected static int getIntValue(Map<String, ?> arguments, String key, int def) {
        Object value = arguments.get(key);
        if (value == null) {
            return def;
        }
        return (int) Double.parseDouble(value.toString());
    }

    protected static float getFloatValue(Map<String, ?> arguments, String key, float def) {
        Object value = arguments.get(key);
        if (value == null) {
            return def;
        }
        return (float) Double.parseDouble(value.toString());
    }

    protected static boolean getBooleanValue(Map<String, ?> arguments, String key, boolean def) {
        Object value = arguments.get(key);
        if (value == null) {
            return def;
        }
        return Boolean.parseBoolean(value.toString());
    }

    /**
     * A builder to extend for all classes extending the {@link BaseImageTranslator}.
     *
     * @param <T> the concrete builder type
     */
    @SuppressWarnings("rawtypes")
    public abstract static class BaseBuilder<T extends BaseBuilder> {

        protected int width = 224;
        protected int height = 224;
        protected Image.Flag flag = Image.Flag.COLOR;
        protected Pipeline pipeline;
        protected Batchifier batchifier = Batchifier.STACK;

        /**
         * Sets the optional {@link Image.Flag} (default is {@link
         * Image.Flag#COLOR}).
         *
         * @param flag the color mode for the images
         * @return this builder
         */
        public T optFlag(Image.Flag flag) {
            this.flag = flag;
            return self();
        }

        /**
         * Sets the {@link Pipeline} to use for pre-processing the image.
         *
         * @param pipeline the pre-processing pipeline
         * @return this builder
         */
        public T setPipeline(Pipeline pipeline) {
            this.pipeline = pipeline;
            return self();
        }

        /**
         * Adds the {@link Transform} to the {@link Pipeline} use for pre-processing the image.
         *
         * @param transform the {@link Transform} to be added
         * @return this builder
         */
        public T addTransform(Transform transform) {
            if (pipeline == null) {
                pipeline = new Pipeline();
            }
            pipeline.add(transform);
            return self();
        }

        /**
         * Sets the {@link Batchifier} for the {@link Translator}.
         *
         * @param batchifier the {@link Batchifier} to be set
         * @return this builder
         */
        public T optBatchifier(Batchifier batchifier) {
            this.batchifier = batchifier;
            return self();
        }

        protected abstract T self();

        protected void validate() {
            if (pipeline == null) {
                throw new IllegalArgumentException("pipeline is required.");
            }
        }

        protected void configPreProcess(Map<String, ?> arguments) {
            if (pipeline == null) {
                pipeline = new Pipeline();
            }
            width = getIntValue(arguments, "width", 224);
            height = getIntValue(arguments, "height", 224);
            if (arguments.containsKey("flag")) {
                flag = Image.Flag.valueOf(arguments.get("flag").toString());
            }
            if (getBooleanValue(arguments, "centerCrop", false)) {
                addTransform(new CenterCrop());
            }
            if (getBooleanValue(arguments, "resize", false)) {
                addTransform(new Resize(width, height));
            }
            if (getBooleanValue(arguments, "toTensor", true)) {
                addTransform(new ToTensor());
            }
            String normalize = getStringValue(arguments, "normalize", "false");
            if ("true".equals(normalize)) {
                addTransform(new Normalize(MEAN, STD));
            } else if (!"false".equals(normalize)) {
                String[] tokens = normalize.split("\\s*,\\s*");
                if (tokens.length != 6) {
                    throw new IllegalArgumentException("Invalid normalize value: " + normalize);
                }
                float[] mean = {
                    Float.parseFloat(tokens[0]),
                    Float.parseFloat(tokens[1]),
                    Float.parseFloat(tokens[2])
                };
                float[] std = {
                    Float.parseFloat(tokens[3]),
                    Float.parseFloat(tokens[4]),
                    Float.parseFloat(tokens[5])
                };
                addTransform(new Normalize(mean, std));
            }
            String range = (String) arguments.get("range");
            if ("0,1".equals(range)) {
                addTransform(a -> a.div(255f));
            } else if ("-1,1".equals(range)) {
                addTransform(a -> a.div(128f).sub(1));
            }
            if (arguments.containsKey("batchifier")) {
                batchifier = Batchifier.fromString((String) arguments.get("batchifier"));
            }
        }

        protected void configPostProcess(Map<String, ?> arguments) {}
    }

    /** A Builder to construct a {@code ImageClassificationTranslator}. */
    @SuppressWarnings("rawtypes")
    public abstract static class ClassificationBuilder<T extends BaseBuilder>
            extends BaseBuilder<T> {

        protected SynsetLoader synsetLoader;

        /**
         * Sets the name of the synset file listing the potential classes for an image.
         *
         * @param synsetArtifactName a file listing the potential classes for an image
         * @return the builder
         */
        public T optSynsetArtifactName(String synsetArtifactName) {
            synsetLoader = new SynsetLoader(synsetArtifactName);
            return self();
        }

        /**
         * Sets the URL of the synset file.
         *
         * @param synsetUrl the URL of the synset file
         * @return the builder
         */
        public T optSynsetUrl(String synsetUrl) {
            try {
                this.synsetLoader = new SynsetLoader(new URL(synsetUrl));
            } catch (MalformedURLException e) {
                throw new IllegalArgumentException("Invalid synsetUrl: " + synsetUrl, e);
            }
            return self();
        }

        /**
         * Sets the potential classes for an image.
         *
         * @param synset the potential classes for an image
         * @return the builder
         */
        public T optSynset(List<String> synset) {
            synsetLoader = new SynsetLoader(synset);
            return self();
        }

        /** {@inheritDoc} */
        @Override
        protected void validate() {
            super.validate();
            if (synsetLoader == null) {
                synsetLoader = new SynsetLoader("synset.txt");
            }
        }

        /** {@inheritDoc} */
        @Override
        protected void configPostProcess(Map<String, ?> arguments) {
            String synset = (String) arguments.get("synset");
            if (synset != null) {
                optSynset(Arrays.asList(synset.split(",")));
            }
            String synsetUrl = (String) arguments.get("synsetUrl");
            if (synsetUrl != null) {
                optSynsetUrl(synsetUrl);
            }
            String synsetFileName = (String) arguments.get("synsetFileName");
            if (synsetFileName != null) {
                optSynsetArtifactName(synsetFileName);
            }
        }
    }

    protected static final class SynsetLoader {

        private String synsetFileName;
        private URL synsetUrl;
        private List<String> synset;

        public SynsetLoader(List<String> synset) {
            this.synset = synset;
        }

        public SynsetLoader(URL synsetUrl) {
            this.synsetUrl = synsetUrl;
        }

        public SynsetLoader(String synsetFileName) {
            this.synsetFileName = synsetFileName;
        }

        public List<String> load(Model model) throws IOException {
            if (synset != null) {
                return synset;
            } else if (synsetUrl != null) {
                try (InputStream is = synsetUrl.openStream()) {
                    return Utils.readLines(is);
                }
            }
            return model.getArtifact(synsetFileName, Utils::readLines);
        }
    }
}

5、创建向量索引字段

需要在索引库创建的时候，一并创建对应字段。

import co.elastic.clients.elasticsearch.ElasticsearchClient;
import co.elastic.clients.elasticsearch._types.mapping.Property;
import co.elastic.clients.elasticsearch._types.mapping.TypeMapping;
import co.elastic.clients.elasticsearch.indices.Alias;
import co.elastic.clients.elasticsearch.indices.CreateIndexRequest;
import co.elastic.clients.elasticsearch.indices.CreateIndexResponse;
import co.elastic.clients.elasticsearch.indices.ExistsRequest;


CreateIndexResponse response = null;
        try {

            TypeMapping.Builder tmBuilder = new TypeMapping.Builder();
            // 图片相似检索，采用点积运算  文本相似采用余线相似
            tmBuilder.properties('_img_vector', new Property.Builder().denseVector(builder -> builder.index(true).dims(1024).similarity("dot_product")
                    .indexOptions(opBuilder -> opBuilder.type("hnsw").m(12).efConstruction(100))).build());

            TypeMapping typeMapping = tmBuilder.build();

            CreateIndexRequest request = CreateIndexRequest.of(builder -> builder.index(indexName)
                    .aliases(indexName + "_alias", new Alias.Builder().isWriteIndex(true).build())
                    .mappings(typeMapping));

            response = esClient.indices().create(request);

            log.info("acknowledged: {}", response.acknowledged());
            log.info("index: {}", response.index());
            log.info("shardsAcknowledged: {}", response.shardsAcknowledged());

            flag = response.acknowledged();
        } catch (IOException e) {
            e.printStackTrace();
        }

创建后生成的结构数据如下

6、添加到ES

float[] feature;
// 自定义属性字段数据，构建文档
            Map<String, Object> dataMap = req.getDataMap();
            // 自定义内置参数
            dataMap.put("_es_doc_type", "IMAGE");

            dataMap.put("_img_vector", feature);
            IndexRequest<Map> request = IndexRequest.of(i -> i
                    .index(req.getIndexLib())
                    .id(req.getDocId())
                    .document(dataMap)
            );
            IndexResponse response = esClient.index(request);
            boolean flag = response.result() == Result.Created;
            log.info("添加文档id={},结果={}", req.getDocId(), flag);

实际存储的数据结构如下图

7、pytorch环境依赖

cpu/linux-x86_64/native/lib/libc10.so.gz
cpu/linux-x86_64/native/lib/libtorch_cpu.so.gz
cpu/linux-x86_64/native/lib/libtorch.so.gz
cpu/linux-x86_64/native/lib/libgomp-52f2fd74.so.1.gz
cpu/osx-aarch64/native/lib/libtorch_cpu.dylib.gz
cpu/osx-aarch64/native/lib/libtorch.dylib.gz
cpu/osx-aarch64/native/lib/libc10.dylib.gz
cpu/osx-x86_64/native/lib/libtorch_cpu.dylib.gz
cpu/osx-x86_64/native/lib/libiomp5.dylib.gz
cpu/osx-x86_64/native/lib/libtorch.dylib.gz
cpu/osx-x86_64/native/lib/libc10.dylib.gz
cpu/win-x86_64/native/lib/torch.dll.gz
cpu/win-x86_64/native/lib/uv.dll.gz
cpu/win-x86_64/native/lib/torch_cpu.dll.gz
cpu/win-x86_64/native/lib/c10.dll.gz
cpu/win-x86_64/native/lib/fbgemm.dll.gz
cpu/win-x86_64/native/lib/libiomp5md.dll.gz
cpu/win-x86_64/native/lib/asmjit.dll.gz
cpu/win-x86_64/native/lib/libiompstubs5md.dll.gz
cpu-precxx11/linux-aarch64/native/lib/libc10.so.gz
cpu-precxx11/linux-aarch64/native/lib/libtorch_cpu.so.gz
cpu-precxx11/linux-aarch64/native/lib/libarm_compute-973e5a6b.so.gz
cpu-precxx11/linux-aarch64/native/lib/libopenblasp-r0-56e95da7.3.24.so.gz
cpu-precxx11/linux-aarch64/native/lib/libtorch.so.gz
cpu-precxx11/linux-aarch64/native/lib/libarm_compute_graph-6990f339.so.gz
cpu-precxx11/linux-aarch64/native/lib/libstdc%2B%2B.so.6.gz
cpu-precxx11/linux-aarch64/native/lib/libarm_compute_core-0793f69d.so.gz
cpu-precxx11/linux-aarch64/native/lib/libgfortran-b6d57c85.so.5.0.0.gz
cpu-precxx11/linux-aarch64/native/lib/libgomp-6e1a1d1b.so.1.0.0.gz
cpu-precxx11/linux-x86_64/native/lib/libgomp-a34b3233.so.1.gz
cpu-precxx11/linux-x86_64/native/lib/libc10.so.gz
cpu-precxx11/linux-x86_64/native/lib/libtorch_cpu.so.gz
cpu-precxx11/linux-x86_64/native/lib/libtorch.so.gz
cpu-precxx11/linux-x86_64/native/lib/libstdc%2B%2B.so.6.gz
cu121/linux-x86_64/native/lib/libc10_cuda.so.gz
cu121/linux-x86_64/native/lib/libcudnn.so.8.gz
cu121/linux-x86_64/native/lib/libnvfuser_codegen.so.gz
cu121/linux-x86_64/native/lib/libc10.so.gz
cu121/linux-x86_64/native/lib/libtorch_cpu.so.gz
cu121/linux-x86_64/native/lib/libcaffe2_nvrtc.so.gz
cu121/linux-x86_64/native/lib/libcudnn_adv_infer.so.8.gz
cu121/linux-x86_64/native/lib/libcudnn_cnn_train.so.8.gz
cu121/linux-x86_64/native/lib/libcudnn_ops_infer.so.8.gz
cu121/linux-x86_64/native/lib/libnvrtc-builtins-6c5639ce.so.12.1.gz
cu121/linux-x86_64/native/lib/libnvrtc-b51b459d.so.12.gz
cu121/linux-x86_64/native/lib/libtorch.so.gz
cu121/linux-x86_64/native/lib/libtorch_cuda_linalg.so.gz
cu121/linux-x86_64/native/lib/libcublas-37d11411.so.12.gz
cu121/linux-x86_64/native/lib/libtorch_cuda.so.gz
cu121/linux-x86_64/native/lib/libcudnn_adv_train.so.8.gz
cu121/linux-x86_64/native/lib/libcublasLt-f97bfc2c.so.12.gz
cu121/linux-x86_64/native/lib/libnvToolsExt-847d78f2.so.1.gz
cu121/linux-x86_64/native/lib/libcudnn_ops_train.so.8.gz
cu121/linux-x86_64/native/lib/libcudnn_cnn_infer.so.8.gz
cu121/linux-x86_64/native/lib/libgomp-52f2fd74.so.1.gz
cu121/linux-x86_64/native/lib/libcudart-9335f6a2.so.12.gz
cu121/win-x86_64/native/lib/zlibwapi.dll.gz
cu121/win-x86_64/native/lib/cudnn_ops_train64_8.dll.gz
cu121/win-x86_64/native/lib/torch.dll.gz
cu121/win-x86_64/native/lib/nvrtc-builtins64_121.dll.gz
cu121/win-x86_64/native/lib/cufftw64_11.dll.gz
cu121/win-x86_64/native/lib/cudnn_adv_infer64_8.dll.gz
cu121/win-x86_64/native/lib/nvrtc64_120_0.dll.gz
cu121/win-x86_64/native/lib/cusolverMg64_11.dll.gz
cu121/win-x86_64/native/lib/torch_cuda.dll.gz
cu121/win-x86_64/native/lib/cufft64_11.dll.gz
cu121/win-x86_64/native/lib/cublas64_12.dll.gz
cu121/win-x86_64/native/lib/cudnn64_8.dll.gz
cu121/win-x86_64/native/lib/uv.dll.gz
cu121/win-x86_64/native/lib/cudnn_cnn_train64_8.dll.gz
cu121/win-x86_64/native/lib/caffe2_nvrtc.dll.gz
cu121/win-x86_64/native/lib/torch_cpu.dll.gz
cu121/win-x86_64/native/lib/c10.dll.gz
cu121/win-x86_64/native/lib/cudnn_cnn_infer64_8.dll.gz
cu121/win-x86_64/native/lib/c10_cuda.dll.gz
cu121/win-x86_64/native/lib/cudart64_12.dll.gz
cu121/win-x86_64/native/lib/nvfuser_codegen.dll.gz
cu121/win-x86_64/native/lib/fbgemm.dll.gz
cu121/win-x86_64/native/lib/curand64_10.dll.gz
cu121/win-x86_64/native/lib/libiomp5md.dll.gz
cu121/win-x86_64/native/lib/cusolver64_11.dll.gz
cu121/win-x86_64/native/lib/cudnn_adv_train64_8.dll.gz
cu121/win-x86_64/native/lib/cublasLt64_12.dll.gz
cu121/win-x86_64/native/lib/nvToolsExt64_1.dll.gz
cu121/win-x86_64/native/lib/nvJitLink_120_0.dll.gz
cu121/win-x86_64/native/lib/cusparse64_12.dll.gz
cu121/win-x86_64/native/lib/asmjit.dll.gz
cu121/win-x86_64/native/lib/cudnn_ops_infer64_8.dll.gz
cu121/win-x86_64/native/lib/libiompstubs5md.dll.gz
cu121/win-x86_64/native/lib/cupti64_2023.1.1.dll.gz
cu121-precxx11/linux-x86_64/native/lib/libgomp-a34b3233.so.1.gz
cu121-precxx11/linux-x86_64/native/lib/libc10_cuda.so.gz
cu121-precxx11/linux-x86_64/native/lib/libcudnn.so.8.gz
cu121-precxx11/linux-x86_64/native/lib/libnvfuser_codegen.so.gz
cu121-precxx11/linux-x86_64/native/lib/libc10.so.gz
cu121-precxx11/linux-x86_64/native/lib/libtorch_cpu.so.gz
cu121-precxx11/linux-x86_64/native/lib/libcaffe2_nvrtc.so.gz
cu121-precxx11/linux-x86_64/native/lib/libcudnn_adv_infer.so.8.gz
cu121-precxx11/linux-x86_64/native/lib/libcudnn_cnn_train.so.8.gz
cu121-precxx11/linux-x86_64/native/lib/libcudnn_ops_infer.so.8.gz
cu121-precxx11/linux-x86_64/native/lib/libnvrtc-builtins-6c5639ce.so.12.1.gz
cu121-precxx11/linux-x86_64/native/lib/libnvrtc-b51b459d.so.12.gz
cu121-precxx11/linux-x86_64/native/lib/libtorch.so.gz
cu121-precxx11/linux-x86_64/native/lib/libtorch_cuda_linalg.so.gz
cu121-precxx11/linux-x86_64/native/lib/libcublas-37d11411.so.12.gz
cu121-precxx11/linux-x86_64/native/lib/libtorch_cuda.so.gz
cu121-precxx11/linux-x86_64/native/lib/libstdc%2B%2B.so.6.gz
cu121-precxx11/linux-x86_64/native/lib/libcudnn_adv_train.so.8.gz
cu121-precxx11/linux-x86_64/native/lib/libcublasLt-f97bfc2c.so.12.gz
cu121-precxx11/linux-x86_64/native/lib/libnvToolsExt-847d78f2.so.1.gz
cu121-precxx11/linux-x86_64/native/lib/libcudnn_ops_train.so.8.gz
cu121-precxx11/linux-x86_64/native/lib/libcudnn_cnn_infer.so.8.gz
cu121-precxx11/linux-x86_64/native/lib/libcudart-9335f6a2.so.12.gz

【Elasticsearch】-图片向量化存储

1、pom依赖

2、加载模型

3、FaceFeatureTranslator

4、BaseImageTranslator

5、创建向量索引字段

6、添加到ES

7、pytorch环境依赖

悦读