修改agent工具描述，分批获取相似文档，返回文档数量进行修改

src/agent/rag_agent.py

@@ -22,7 +22,7 @@ class IssuanceArgs(BaseModel):
 
 class RAGQuery(BaseTool):
     name = "rag_query"
-    description = """这是一个区（县）级的水文气象地质知识库，当咨询区（县）的水文气象地质等相关信息的时候，你可以提供数据和资料。注意，这个查询只能获取单个区（县）的水文气象地质等相关信息，当需要查询省市的详细信息时，需要获取改省市下的具体行政规划，并一一获取具体的区（县）的水文气象地质等相关信息。这个知识库中信息并不全面，有可能缺失。"""
+    description = """这是一个区（县）级的水文气象地质知识库，当咨询区（县）的水文气象地质等相关信息的时候，你可以提供数据和资料。注意，这个查询只能获取一个区（县）的水文气象地质等相关信息。如果问题中有多个区县，请拆解出来，并一个区县一个区县的查询。当需要查询省市的详细信息时，需要获取改省市下的具体行政规划，并一一获取具体的区（县）的水文气象地质等相关信息。这个知识库中信息并不全面，有可能缺失。"""
     args_schema: Type[BaseModel] = IssuanceArgs
     rerank: Any  # 替换 Any 为适当的类型
     rerank_model: Any  # 替换 Any 为适当的类型
@@ -53,9 +53,10 @@ class RAGQuery(BaseTool):
         # split_list = []
         # for l in split_str:
         #     split_list.append(l)
-
+        split_docs_list = []
+        for l in location:
             start = time.time()
-        answer = self.db.find_like_doc(location)
+            answer = self.db.find_like_doc(l)
             end = time.time()
             print('find_like_doc time: %s Seconds' % (end - start))
             print(len(answer) if answer else 0)
@@ -64,6 +65,7 @@ class RAGQuery(BaseTool):
                 d = Document(page_content=a[0], metadata=json.loads(a[1]))
                 split_docs.append(d)
             print(len(split_docs))
+            split_docs_list.append(split_docs)
         # if len(split_docs)>10:
         #     split_docs= split_docs[:10]
 
@@ -77,7 +79,7 @@ class RAGQuery(BaseTool):
         print(matches)
         similarity = self.get_similarity_with_ext_origin(matches,_location=location)
         # cur_similarity = similarity.get_rerank(self.rerank_model)
-        cur_similarity = similarity.get_rerank_with_doc(self.rerank_model,split_docs)
+        cur_similarity = similarity.get_rerank_with_doc(self.rerank_model,split_docs_list)
         # docs = similarity.get_rerank_docs()
         # print(cur_similarity)
         # # geo_result = "以下是详细的水文气象地质资料："+cur_similarity+"\n 以下是原问题"+question

src/agent/tool_divisions.py

@@ -132,7 +132,7 @@ class AdministrativeDivisionArgs(BaseModel):
 
 class AdministrativeDivision(BaseTool):
     name = "administrative_division"
-    description = "根据用户提问中涉及到的地区信息补全其行政区划信息，明确具体的省、市、县信息。比如输入县，补全所属省市，输入市则补全省级以及下辖所有县区"
+    description = "根据用户提问中涉及到的地区信息补全其行政区划信息，明确具体的省、市、县信息。比如输入县，补全所属省市，输入市则补全省级以及下辖所有县区,当问题中涉及区县的时候，一定要优先调用此工具"
     args_schema: Type[BaseModel] = AdministrativeDivisionArgs
 
     def _run(self, input_text: str) -> str:

src/pgdb/knowledge/txt_doc_table.py

@@ -66,16 +66,9 @@ class TxtDoc:
             print("drop table txt_doc ok")
 
 
-    def find_like_doc(self,item:list):
-        print(item)
-        i0 = item[0]
-        if len(item)>=1:
-            item = item[1:]
-            print(item)
-        query = "select text,matadate FROM txt_doc WHERE matadate like '%"+i0+"%' or text like '%"+i0+"%' "
-        for i in item:
-            query+= "or matadate like '%"+i+"%' or text like '%"+i0+"%' "
-        print(query)
+    def find_like_doc(self,item:str):
+
+        query = "select text,matadate FROM txt_doc WHERE matadate like '%"+item+"%' or text like '%"+item+"%' "
         self.db.execute(query)
         answer = self.db.fetchall()
 

src/server/get_similarity.py

@@ -80,33 +80,31 @@ class GetSimilarityWithExt:
         result += "]"
         return result
 
-    def get_rerank_with_doc(self, reranker: BgeRerank,split_doc:list, top_k=5):
+    def get_rerank_with_doc(self, reranker: BgeRerank,split_docs_list:list):
+        top_k = self.get_doc_nums(len(split_docs_list))
         question = '\n'.join(self.question)
         print(question)
+        rerank_docs1_hash = []
+        rerank_docs2_hash = []
+        m = {}
+        result = []
+        for split_doc in split_docs_list:
             start = time.time()
             rerank_docs1 = reranker.compress_documents(split_doc, question)
             end = time.time()
             print('重排1 time: %s Seconds' % (end - start))
+            for doc in rerank_docs1:
+                m[hash(doc.page_content)] = doc
+                rerank_docs1_hash.append(hash(doc.page_content))
+            result.append((60, rerank_docs1_hash))
         start = time.time()
         rerank_docs2 = reranker.compress_documents(self.similarity_docs, question)
         end = time.time()
         print('重排2 time: %s Seconds' % (end - start))
-        rerank_docs1_hash = []
-        rerank_docs2_hash = []
-        m = {}
-        start = time.time()
-        for doc in rerank_docs1:
-            m[hash(doc.page_content)] = doc
-            rerank_docs1_hash.append(hash(doc.page_content))
-
         for doc in rerank_docs2:
             m[hash(doc.page_content)] = doc
             rerank_docs2_hash.append(hash(doc.page_content))
 
-        end = time.time()
-        step_time1 = end - start
-        result = []
-        result.append((60,rerank_docs1_hash))
         result.append((55,rerank_docs2_hash))
         print(len(rerank_docs1_hash))
         print(len(rerank_docs2_hash))
@@ -114,14 +112,11 @@ class GetSimilarityWithExt:
         rrf_doc = reciprocal_rank_fusion(result)
         end = time.time()
         print('混排 time: %s Seconds' % (end - start))
-        print(rrf_doc)
+        print("混排文档数量：", len(rrf_doc))
         d_list = []
-        start = time.time()
         for key in rrf_doc:
             d_list.append(m[key])
-        end = time.time()
-        step_time2 = end - start
-        print('文档去重 time: %s Seconds' % (step_time1 + step_time2))
+        print("返回文档数量：",top_k)
         self.rerank_docs = d_list[:top_k]
         return self.join_document(d_list[:top_k])
 
@@ -151,6 +146,15 @@ class GetSimilarityWithExt:
         print(len(unique_documents))
         return unique_documents
 
+    def get_doc_nums(self,num :int)->int:
+        num = num*3
+        if num<5:
+            return 5
+        elif num>30:
+            return 30
+        else:
+            return num
+
 class QAExt:
     llm = None
 

test/rag_agent_test.py

@@ -10,7 +10,8 @@ from langchain_core.prompts.chat import ChatPromptTemplate,HumanMessagePromptTem
 from langchain_core.prompts import PromptTemplate
 from langchain.chains import LLMChain
 import langchain_core
-
+from src.llm.ernie_with_sdk import ChatERNIESerLLM
+from qianfan import ChatCompletion
 from src.pgdb.knowledge.similarity import VectorStore_FAISS
 from src.server.get_similarity import QAExt
 from src.server.agent import create_chart_agent
@@ -41,7 +42,8 @@ base_llm = ChatOpenAI(
     verbose=True,
     temperature=0
 )
-
+# base_llm = ChatERNIESerLLM(
+#         chat_completion=ChatCompletion(ak="pT7sV1smp4AeDl0LjyZuHBV9", sk="b3N0ibo1IKTLZlSs7weZc8jdR0oHjyMu"))
 
 vecstore_faiss = VectorStore_FAISS(
         embedding_model_name=EMBEEDING_MODEL_PATH,
@@ -98,7 +100,7 @@ for h in history:
     prompt += "问：{}\n答：{}\n".format(h[0], h[1])
 print(prompt)
 # res = agent_executor.invoke({"input":"以下历史对话记录： "+prompt+"以下是问题："+"攸县、长沙县、化隆县和大通县谁的年平均降雨量大"})
-res = agent_executor.invoke({"input":"攸县、长沙县、化隆县和大通县谁的年平均降雨量大","histories":history})
+res = agent_executor.invoke({"input":"西宁市各区县年平均降雨量","histories":history})
 
 print("====== result: ======")
 print(res)