rag工具优化

src/pgdb/knowledge/txt_doc_table.py

@@ -69,12 +69,12 @@ class TxtDoc:
     def find_like_doc(self,item:list):
         print(item)
         i0 = item[0]
-        if len(item)>1:
+        if len(item)>=1:
             item = item[1:]
             print(item)
-        query = "select text,matadate FROM txt_doc WHERE text like '%"+i0+"%' "
+        query = "select text,matadate FROM txt_doc WHERE matadate like '%"+i0+"%' or text like '%"+i0+"%' "
         for i in item:
-            query+= "or text like '%"+i+"%' "
+            query+= "or matadate like '%"+i+"%' or text like '%"+i0+"%' "
         print(query)
         self.db.execute(query)
         answer = self.db.fetchall()

test/rag_agent_test.py

+from typing import List,Union
+from langchain_openai import ChatOpenAI
+from langchain.agents import AgentExecutor,create_structured_chat_agent
+from langchain_core.prompts.chat import ChatPromptTemplate,HumanMessagePromptTemplate,SystemMessagePromptTemplate,MessagesPlaceholder
+from langchain_core.prompts import PromptTemplate
+import langchain_core
+from src.pgdb.knowledge.similarity import VectorStore_FAISS
+from src.server.get_similarity import QAExt
+from src.pgdb.knowledge.k_db import PostgresDB
+from src.pgdb.knowledge.txt_doc_table import TxtDoc
+from langchain.chains import LLMChain
+from src.agent.tool_divisions import AdministrativeDivision
+from src.agent.rag_agent import RAGQuery
+from src.config.consts import (
+    EMBEEDING_MODEL_PATH,
+    FAISS_STORE_PATH,
+    INDEX_NAME,
+    VEC_DB_HOST,
+    VEC_DB_PASSWORD,
+    VEC_DB_PORT,
+    VEC_DB_USER,
+    VEC_DB_DBNAME,
+    SIMILARITY_SHOW_NUMBER,
+    prompt_enhancement_history_template,
+    prompt1
+)
+base_llm = ChatOpenAI(
+    openai_api_key='xxxxxxxxxxxxx',
+    openai_api_base='http://192.168.10.14:8000/v1',
+    model_name='Qwen2-7B',
+    verbose=True,
+    temperature=0
+)
+
+
+vecstore_faiss = VectorStore_FAISS(
+        embedding_model_name=EMBEEDING_MODEL_PATH,
+        store_path=FAISS_STORE_PATH,
+        index_name=INDEX_NAME,
+        info={"port": VEC_DB_PORT, "host": VEC_DB_HOST, "dbname": VEC_DB_DBNAME, "username": VEC_DB_USER,
+              "password": VEC_DB_PASSWORD},
+        show_number=SIMILARITY_SHOW_NUMBER,
+        reset=False)
+ext = QAExt(base_llm)
+
+k_db = PostgresDB(host=VEC_DB_HOST, database=VEC_DB_DBNAME, user=VEC_DB_USER, password=VEC_DB_PASSWORD, port=VEC_DB_PORT)
+k_db.connect()
+
+llm_chain = LLMChain(llm=base_llm, prompt=PromptTemplate(input_variables=["history","context", "question"], template=prompt1), llm_kwargs= {"temperature": 0})
+
+tools = [AdministrativeDivision(),RAGQuery(vecstore_faiss,ext,PromptTemplate(input_variables=["history","context", "question"], template=prompt_enhancement_history_template),_db=TxtDoc(k_db),_llm_chain=llm_chain)]
+input_variables=['agent_scratchpad', 'input', 'tool_names', 'tools']
+input_types={'chat_history': List[Union[langchain_core.messages.ai.AIMessage, langchain_core.messages.human.HumanMessage, langchain_core.messages.chat.ChatMessage, langchain_core.messages.system.SystemMessage, langchain_core.messages.function.FunctionMessage, langchain_core.messages.tool.ToolMessage]]}
+metadata={'lc_hub_owner': 'hwchase17', 'lc_hub_repo': 'structured-chat-agent', 'lc_hub_commit_hash': 'ea510f70a5872eb0f41a4e3b7bb004d5711dc127adee08329c664c6c8be5f13c'}
+messages=[
+  SystemMessagePromptTemplate(prompt=PromptTemplate(input_variables=['tool_names', 'tools'], template='Respond to the human as helpfully and accurately as possible. You have access to the following tools:\n\n{tools}\n\nUse a json blob to specify a tool by providing an action key (tool name) and an action_input key (tool input).\n\nValid "action" values: "Final Answer" or {tool_names}\n\nProvide only ONE action per $JSON_BLOB, as shown:\n\n```\n{{\n  "action": $TOOL_NAME,\n  "action_input": $INPUT\n}}\n```\n\nFollow this format:\n\nQuestion: input question to answer\nThought: consider previous and subsequent steps\nAction:\n```\n$JSON_BLOB\n```\nObservation: action result\n... (repeat Thought/Action/Observation N times)\nThought: I know what to respond\nAction:\n```\n{{\n  "action": "Final Answer",\n  "action_input": "Final response to human"\n}}\n\nBegin! Reminder to ALWAYS respond with a valid json blob of a single action. Use tools if necessary. Respond directly if appropriate. Format is Action:```$JSON_BLOB```then Observation')),
+  MessagesPlaceholder(variable_name='chat_history', optional=True),
+  HumanMessagePromptTemplate(prompt=PromptTemplate(input_variables=['agent_scratchpad', 'input'], template='{input}\n\n{agent_scratchpad}\n (reminder to respond in a JSON blob no matter what)'))
+]
+
+prompt = ChatPromptTemplate(
+    input_variables=input_variables,
+    input_types=input_types,
+    metadata=metadata,
+    messages=messages
+)
+
+
+agent = create_structured_chat_agent(llm=base_llm, tools=tools, prompt=prompt)
+agent_executor = AgentExecutor(agent=agent, tools=tools,verbose=True,handle_parsing_errors=True)
+history = []
+h1 = []
+h1.append("攸县年降雨量")
+h1.append("攸县年雨量平均为30ml")
+history.append(h1)
+h1 = []
+h1.append("长沙县年降雨量")
+h1.append("长沙县年雨量平均为50ml")
+history.append(h1)
+prompt = ""
+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"})
+
+print("====== result: ======")
+print(res)