代码提交

.idea/knowledge.iml

@@ -4,7 +4,7 @@
     <content url="file://$MODULE_DIR$">
       <excludeFolder url="file://$MODULE_DIR$/.venv" />
     </content>
-    <orderEntry type="jdk" jdkName="C:\ProgramData\anaconda3" jdkType="Python SDK" />
+    <orderEntry type="jdk" jdkName="openKnowledge" jdkType="Python SDK" />
     <orderEntry type="sourceFolder" forTests="false" />
   </component>
   <component name="PyDocumentationSettings">

.idea/misc.xml

@@ -3,5 +3,5 @@
   <component name="Black">
     <option name="sdkName" value="Python 3.9" />
   </component>
-  <component name="ProjectRootManager" version="2" project-jdk-name="C:\ProgramData\anaconda3" project-jdk-type="Python SDK" />
+  <component name="ProjectRootManager" version="2" project-jdk-name="openKnowledge" project-jdk-type="Python SDK" />
 </project>

requirements.txt

@@ -1,6 +1,6 @@
 fastapi==0.115.12
-#networkx==3.4.2
-#numpy==2.2.5
+sentence-transformers==4.0.1
+numpy==1.26.4
 pgvector==0.1.8
 pydantic==2.11.1
 Requests==2.31.0

src/knowledge/.env

@@ -0,0 +1,8 @@
+DB_HOST = 173.18.12.203
+DB_NAME = medkg
+DB_PORT = 5432
+DB_USER = knowledge
+DB_PASSWORD = qwer1234.
+
+license=E:\project\knowledge\license_issued
+EMBEDDING_MODEL=E:\project\knowledge2\bge-m3

src/knowledge/service/kg_node_service.py

@@ -23,7 +23,7 @@ class KGNodeService:
         if cache_key in self._cache:
             return self._cache[cache_key]
 
-        query_embedding = Vectorizer.get_embedding(title)
+        query_embedding = Vectorizer.get_instance().get_embedding(title)
         db = next(get_db())
         # 执行向量搜索
         results = (
@@ -67,7 +67,7 @@ class KGNodeService:
         if limit < 1:
             limit = 10
 
-        embedding = Vectorizer.get_embedding(keyword)
+        embedding = Vectorizer.get_instance().get_embedding(keyword)
         offset = (page_no - 1) * limit
 
         try:
@@ -207,7 +207,7 @@ class KGNodeService:
                 updated_nodes = []
                 for node in nodes:
                     if not node.embedding:
-                        embedding = Vectorizer.get_embedding(node.name)
+                        embedding = Vectorizer.get_instance().get_embedding(node.name)
                         node.embedding = embedding
                         updated_nodes.append(node)
                 if updated_nodes: