更新上下文参数

app.py

@@ -69,6 +69,7 @@ def route_embed():
 
 def route_query(msg):
     response = query(msg)
+    print(response)
     if response:
         resObj = {}
         resObj["data"] = response

get_vector_db.py

@@ -7,7 +7,7 @@ COLLECTION_NAME = os.getenv('COLLECTION_NAME', 'siwei_ai')
 TEXT_EMBEDDING_MODEL = os.getenv('TEXT_EMBEDDING_MODEL', 'nomic-embed-text')
 
 def get_vector_db():
-    embedding = OllamaEmbeddings(model=TEXT_EMBEDDING_MODEL,show_progress=True,num_gpu=0)
+    embedding = OllamaEmbeddings(model=TEXT_EMBEDDING_MODEL,show_progress=True,num_ctx=32000,num_gpu=0)
 
     db = Chroma(
         collection_name=COLLECTION_NAME,

query.py

@@ -17,23 +17,24 @@ def get_prompt():
         矢量数据库。通过对用户问题生成多个视角
         目标是帮助用户克服基于距离的一些局限性
         相似性搜索。请提供这些用换行符分隔的备选问题。
-        原始问题：{问题}""",
+        Original question: {question}""",
     )
 
     template = """仅根据以下上下文用中文回答问题：
     {context},请严格以markdown格式输出并保障寄送格式正确无误，
     Question: {question}
     """
+    # Question: {question}
 
-    prompt = ChatPromptTemplate.from_template(template)
 
+    prompt = ChatPromptTemplate.from_template(template)
     return QUERY_PROMPT, prompt
 
 # Main function to handle the query process
 def query(input):
     if input:
         # Initialize the language model with the specified model name
-        llm = ChatOllama(model=LLM_MODEL)
+        llm = ChatOllama(model=LLM_MODEL,keep_alive=-1,num_ctx=32000,num_gpu=0)
         # Get the vector database instance
         db = get_vector_db()
         # Get the prompt templates