增加向量化的库

pom.xml

@@ -58,6 +58,7 @@
         <okhttp.version>4.9.0</okhttp.version>
         <springboot.ai>1.0.3</springboot.ai>
         <httpclient5.version>5.3.1</httpclient5.version>
+        <jedis.version>5.1.0</jedis.version>
     </properties>
 
     <dependencyManagement>
@@ -92,6 +93,7 @@
                 <scope>import</scope>
             </dependency>
 
+
             <!--spring-cloud-->
             <dependency>
                 <groupId>org.springframework.cloud</groupId>
@@ -181,6 +183,12 @@
                 <version>${okhttp.version}</version>
             </dependency>
 
+            <dependency>
+                <groupId>redis.clients</groupId>
+                <artifactId>jedis</artifactId>
+                <version>${jedis.version}</version>
+            </dependency>
+
 
         </dependencies>
     </dependencyManagement>

virtual-patient-common/pom.xml

@@ -48,20 +48,26 @@
             <artifactId>httpclient5</artifactId>
         </dependency>
 
+        <dependency>
+            <groupId>io.springboot.ai</groupId>
+            <artifactId>spring-ai-redis</artifactId>
+        </dependency>
+
         <dependency>
             <groupId>org.springframework.cloud</groupId>
             <artifactId>spring-cloud-starter-bootstrap</artifactId>
         </dependency>
 
-
         <dependency>
             <groupId>org.springframework.boot</groupId>
             <artifactId>spring-boot-starter-data-redis</artifactId>
         </dependency>
+
         <dependency>
             <groupId>redis.clients</groupId>
             <artifactId>jedis</artifactId>
         </dependency>
+
         <!--redis分布式锁 https://gitee.com/baomidou/lock4j -->
         <dependency>
             <groupId>com.baomidou</groupId>

virtual-patient-common/src/main/java/com/supervision/config/VectorEmbeddingClient.java

@@ -0,0 +1,53 @@
+package com.supervision.config;
+
+import cn.hutool.http.HttpUtil;
+import cn.hutool.json.JSONUtil;
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+import org.springframework.ai.document.Document;
+import org.springframework.ai.embedding.*;
+import org.springframework.util.Assert;
+
+import java.util.ArrayList;
+import java.util.List;
+import java.util.Map;
+import java.util.concurrent.atomic.AtomicInteger;
+
+public class VectorEmbeddingClient extends AbstractEmbeddingClient {
+
+    private final Logger logger = LoggerFactory.getLogger(getClass());
+
+
+    @Override
+    public List<Double> embed(Document document) {
+        List<List<Double>> list = this.call(new EmbeddingRequest(List.of(document.getContent()), EmbeddingOptions.EMPTY))
+                .getResults()
+                .stream()
+                .map(Embedding::getOutput)
+                .toList();
+        return list.iterator().next();
+    }
+
+    @Override
+    public EmbeddingResponse call(EmbeddingRequest request) {
+        Assert.notEmpty(request.getInstructions(), "At least one text is required!");
+        if (request.getInstructions().size() != 1) {
+            logger.warn(
+                    "Ollama Embedding does not support batch embedding. Will make multiple API calls to embed(Document)");
+        }
+        List<List<Double>> embeddingList = new ArrayList<>();
+        for (String inputContent : request.getInstructions()) {
+            // TODO 这里需要吧inputContent转化为向量数据
+            String post = HttpUtil.post("http://192.168.10.42:8000/embeddings/", JSONUtil.toJsonStr(Map.of("text", inputContent)));
+            String o = JSONUtil.parseObj(post).getStr("embeddings");
+            List<Double> embedding = JSONUtil.toList(o, Double.class);
+            //List<Double> embedding = List.of(1.0, 2.0, 3.0);
+            embeddingList.add(embedding);
+        }
+        var indexCounter = new AtomicInteger(0);
+        List<Embedding> embeddings = embeddingList.stream()
+                .map(e -> new Embedding(e, indexCounter.getAndIncrement()))
+                .toList();
+        return new EmbeddingResponse(embeddings);
+    }
+}

virtual-patient-common/src/main/java/com/supervision/config/VectorSimilarityConfiguration.java

@@ -0,0 +1,26 @@
+package com.supervision.config;
+
+import org.springframework.ai.vectorstore.RedisVectorStore;
+import org.springframework.context.annotation.Bean;
+import org.springframework.context.annotation.Configuration;
+
+@Configuration
+public class VectorSimilarityConfiguration {
+
+    @Bean
+    public VectorEmbeddingClient vectorEmbeddingClient() {
+        return new VectorEmbeddingClient();
+    }
+
+    @Bean
+    public RedisVectorStore redisVectorStore(VectorEmbeddingClient vectorEmbeddingClient) {
+        RedisVectorStore.RedisVectorStoreConfig config = RedisVectorStore.RedisVectorStoreConfig.builder()
+                .withURI("redis://:123456@192.168.10.137:6380")
+                // 定义搜索过滤器使用的元数据字段
+                .withMetadataFields(
+                        RedisVectorStore.MetadataField.tag("medicalId"),
+                        RedisVectorStore.MetadataField.tag("type"))
+                .build();
+        return new RedisVectorStore(config, vectorEmbeddingClient);
+    }
+}

virtual-patient-common/src/main/java/com/supervision/util/VectorSimilarityUtil.java

@@ -0,0 +1,123 @@
+package com.supervision.util;
+
+import cn.hutool.core.map.MapUtil;
+import lombok.Data;
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.ai.document.Document;
+
+import java.util.*;
+import java.util.concurrent.ConcurrentHashMap;
+
+@Slf4j
+public class VectorSimilarityUtil {
+
+
+    private static final Map<String, Map<String, Document>> storeMap = new ConcurrentHashMap<>();
+
+    private static final Double similarityThreshold = 0.5;
+
+    private static final Integer topK = 5;
+
+
+    public static void add(String storeId, Document document) {
+        // TODO 需要序列化成为索引
+        List<Double> embedding = null;
+        document.setEmbedding(embedding);
+        Map<String, Document> store = storeMap.computeIfAbsent(storeId, k -> new ConcurrentHashMap<>());
+        store.put(document.getId(), document);
+    }
+
+    public static void add(String storeId, List<Document> documents) {
+        for (Document document : documents) {
+            // TODO 需要序列化成为索引
+            List<Double> embedding = null;
+            document.setEmbedding(embedding);
+            Map<String, Document> store = storeMap.computeIfAbsent(storeId, k -> new ConcurrentHashMap<>());
+            store.put(document.getId(), document);
+        }
+    }
+
+    public static Optional<Boolean> delete(String storeId, List<String> idList) {
+        if (!storeMap.containsKey(storeId)) {
+            for (String id : idList) {
+                storeMap.get(storeId).remove(id);
+            }
+        }
+        return Optional.of(true);
+    }
+
+    public static List<Document> similaritySearch(String storeId, List<Double> userQueryEmbedding) {
+        Map<String, Document> store = storeMap.get(storeId);
+        if (MapUtil.isNotEmpty(store)) {
+            return store.values()
+                    .stream()
+                    .map(entry -> new Similarity(entry.getId(),
+                            EmbeddingMath.cosineSimilarity(userQueryEmbedding, entry.getEmbedding())))
+                    .filter(s -> s.score >= similarityThreshold)
+                    .sorted(Comparator.<Similarity>comparingDouble(s -> s.score).reversed())
+                    .limit(topK)
+                    .map(s -> store.get(s.key))
+                    .toList();
+        }
+        return new ArrayList<>();
+    }
+
+
+    @Data
+    public static class Similarity {
+
+        private String key;
+
+        private double score;
+
+        public Similarity(String key, double score) {
+            this.key = key;
+            this.score = score;
+        }
+
+    }
+
+    public static class EmbeddingMath {
+
+        private EmbeddingMath() {
+            throw new UnsupportedOperationException("This is a utility class and cannot be instantiated");
+        }
+
+        public static double cosineSimilarity(List<Double> vectorX, List<Double> vectorY) {
+            if (vectorX == null || vectorY == null) {
+                throw new RuntimeException("Vectors must not be null");
+            }
+            if (vectorX.size() != vectorY.size()) {
+                throw new IllegalArgumentException("Vectors lengths must be equal");
+            }
+
+            double dotProduct = dotProduct(vectorX, vectorY);
+            double normX = norm(vectorX);
+            double normY = norm(vectorY);
+
+            if (normX == 0 || normY == 0) {
+                throw new IllegalArgumentException("Vectors cannot have zero norm");
+            }
+
+            return dotProduct / (Math.sqrt(normX) * Math.sqrt(normY));
+        }
+
+        public static double dotProduct(List<Double> vectorX, List<Double> vectorY) {
+            if (vectorX.size() != vectorY.size()) {
+                throw new IllegalArgumentException("Vectors lengths must be equal");
+            }
+
+            double result = 0;
+            for (int i = 0; i < vectorX.size(); ++i) {
+                result += vectorX.get(i) * vectorY.get(i);
+            }
+
+            return result;
+        }
+
+        public static double norm(List<Double> vector) {
+            return dotProduct(vector, vector);
+        }
+
+    }
+}

virtual-patient-web/pom.xml

@@ -110,7 +110,6 @@
             <scope>test</scope>
         </dependency>
 
-
     </dependencies>
 
     <build>

virtual-patient-web/src/main/java/com/supervision/controller/TestController.java

@@ -5,18 +5,27 @@ import cn.hutool.http.HttpUtil;
 import cn.hutool.json.JSONObject;
 import cn.hutool.json.JSONUtil;
 import com.supervision.exception.BusinessException;
+import com.supervision.model.AskTemplateQuestionLibrary;
 import com.supervision.model.ConfigPhysicalTool;
+import com.supervision.service.AskTemplateQuestionLibraryService;
 import com.supervision.service.ConfigPhysicalToolService;
 import com.supervision.util.MinioUtil;
 import lombok.RequiredArgsConstructor;
-import org.springframework.web.bind.annotation.*;
+import lombok.extern.slf4j.Slf4j;
+import org.springframework.ai.document.Document;
+import org.springframework.ai.vectorstore.RedisVectorStore;
+import org.springframework.ai.vectorstore.SearchRequest;
+import org.springframework.web.bind.annotation.GetMapping;
+import org.springframework.web.bind.annotation.PostMapping;
+import org.springframework.web.bind.annotation.RequestMapping;
+import org.springframework.web.bind.annotation.RestController;
 import org.springframework.web.multipart.MultipartFile;
 
-import java.io.FileInputStream;
-import java.io.IOException;
 import java.io.InputStream;
 import java.util.*;
+import java.util.stream.Collectors;
 
+@Slf4j
 @RestController
 @RequestMapping("test")
 @RequiredArgsConstructor
@@ -24,6 +33,32 @@ public class TestController {
 
     private final ConfigPhysicalToolService configPhysicalToolService;
 
+    private final RedisVectorStore redisVectorStore;
+
+    private final AskTemplateQuestionLibraryService askTemplateQuestionLibraryService;
+
+    @GetMapping("testRedisVectorStore")
+    public void testRedisVectorStore() {
+        List<AskTemplateQuestionLibrary> list = askTemplateQuestionLibraryService.list();
+        for (AskTemplateQuestionLibrary askTemplateQuestionLibrary : list) {
+            String description = askTemplateQuestionLibrary.getDescription();
+            redisVectorStore.add(List.of(new Document(description, Map.of("type", "1", "medicalId", "222"))));
+            List<String> question = askTemplateQuestionLibrary.getQuestion();
+            for (String s : question) {
+                redisVectorStore.add(List.of(new Document(s, Map.of("type", "1", "medicalId", "222"))));
+            }
+//            log.info("处理完成:{}", description);
+        }
+    }
+
+    @GetMapping("testQuestion")
+    public List<Map<String, String>> testQuestion(String question) {
+        List<Document> documents = redisVectorStore.similaritySearch(SearchRequest
+                .query(question)
+                .withTopK(5));
+        return documents.stream().map(document -> Map.of("content", document.getContent(), "id", document.getId())).collect(Collectors.toList());
+    }
+
     @GetMapping("testExpireTime")
     public String testExpireTime() {
         return "OK";
@@ -36,18 +71,19 @@ public class TestController {
 
     /**
      * 数字人获取房间号
-     * @param key 数字人ID,张总那边提供
+     *
+     * @param key   数字人ID,张总那边提供
      * @param token 前端每打开一个页面,就给一个新的UUID
      * @return 房间号
      */
     @GetMapping("queryRoomId")
-    public String queryRoomId(String key,String token){
+    public String queryRoomId(String key, String token) {
         Map<String, Object> param = new HashMap<>();
-        param.put("token",token);
+        param.put("token", token);
-        param.put("key",key);
+        param.put("key", key);
         String s = HttpUtil.get("https://digital-human.jd.com/getRoomId", param);
         JSONObject entries = JSONUtil.parseObj(s);
-        if (200 != entries.getInt("code")){
+        if (200 != entries.getInt("code")) {
             throw new BusinessException(entries.getStr("data"));
         }
         return entries.getStr("data");
@@ -56,7 +92,8 @@ public class TestController {
 
     /**
      * 调用数字人接口进行语音的播放
-     * @param text 需要播放的文本
+     *
+     * @param text   需要播放的文本
      * @param roomId 房间ID
      */
     @GetMapping("shuZiRenSend")
@@ -102,5 +139,4 @@ public class TestController {
     }
 
 
-
 }

virtual-patient-web/src/main/resources/application.yml

@@ -12,6 +12,8 @@ server:
     # 是否分配的直接内存
     direct-buffers: true
 spring:
+  main:
+    allow-bean-definition-overriding: true
   servlet:
     multipart:
       max-file-size: 100MB
@@ -39,6 +41,12 @@ spring:
           log-slow-sql: true                                # 是否开启 慢SQL 记录，默认false
           slow-sql-millis: 5000                             # 慢 SQL 的标准，默认 3000，单位：毫秒
           merge-sql: false                                  # 合并多个连接池的监控数据，默认false
+  ai:
+    vectorstore:
+      redis:
+        uri: redis://:123456@192.168.10.137:6380
+        index: 1
+        prefix: 'vp:vector:'
 
 mybatis-plus:
   mapper-locations: classpath*:mapper/**/*.xml