import os
import asyncio
from pydantic import SecretStr
#ollamba_basurl = "http://localhost:11434"
#model_name = "deepseekr1:7b"
# llm = ChatOllama(
# temperature=0.0,
# model= model_name,
# base_url = ollamba_basurl
# )
llm = ChatOpenAI(
base_url=os.getenv("OPENAI_URL"),
model=os.getenv("QIANWEN_MODEL_NAME"),
api_key=os.getenv("OPENAI_API_KEY")
)
# embeddings = OllamaEmbeddings(model = "quentinz/bge-large-zh-v1.5:lates",
# base_url= ollamba_basurl)
embeddings = OpenAIEmbeddings(
base_url = "https://ark.cn-beijing.volces.com/api/v3",
api_key = os.getenv("OPENAI_EMBEDDING_API_KEY"),
model = "doubao-embedding-large-text-240915"
)
collection = 'pdf_document'
try:
db = Milvus(
embedding_function = embeddings,
collection_name = collection,
text_field="content",
vector_field="embedding",
connection_args={
"host":"192.168.64.145",
"port":"19530",
"user":"root",
"password":"Milvus",
"db_name":"default"
}
)
except Exception as e:
print(f"连接milvus失败 {e}")
raise e
# 创建相似度检索器
retriever = db.as_retriever(
search_type="similarity",
search_kwargs={
"k": 3
}
)
# 构建提示词
prompt = ChatPromptTemplate.from_template(
"""
请根据上下文之间输出最终答案，不要包含任何推理过程，如果不知道答案，请输出不知道
上下文: {content}
问题: {question}
答案:
"""
)
# 构建请求链
setup_retriever = RunnableParallel({"content": retriever, "question" : RunnablePassthrough()})
chain = setup_retriever|prompt | llm | StrOutputParser()
async def get_answer(question):
async for chunk in chain.astream(question):
print(chunk,end = " ",flush= True)
if __name__ == '__main__':
response = chain.invoke("bright grace 下面是什么")
print(response)
# loop = asyncio.get_event_loop()
# asyncio.set_event_loop(loop)
# loop.run_until_complete(get_answer("回答一个你会的问题"))
D:\code\python-start-demo\ai\ollama_embedding.py:43: LangChainDeprecationWarning: The class `Milvus` was deprecated in LangChain 0.2.0 and will be removed in 1.0. An updated version of the class exists in the :class:`~langchain-milvus package and should be used instead. To use it run `pip install -U :class:`~langchain-milvus` and import as `from :class:`~langchain_milvus import MilvusVectorStore``.
db = Milvus(
Traceback (most recent call last):
File "D:\code\python-start-demo\ai\ollama_embedding.py", line 86, in <module>
response = chain.invoke("bright grace 下面是什么")
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\langchain_core\runnables\base.py", line 3045, in invoke
input_ = context.run(step.invoke, input_, config, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\langchain_core\runnables\base.py", line 3774, in invoke
output = {key: future.result() for key, future in zip(steps, futures)}
^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\concurrent\futures\_base.py", line 456, in result
return self.__get_result()
^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\concurrent\futures\_base.py", line 401, in __get_result
raise self._exception
File "D:\ProgramData\conda_envs\myenv\Lib\concurrent\futures\thread.py", line 59, in run
result = self.fn(*self.args, **self.kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\langchain_core\runnables\base.py", line 3758, in _invoke_step
return context.run(
^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\langchain_core\retrievers.py", line 259, in invoke
result = self._get_relevant_documents(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\langchain_core\vectorstores\base.py", line 1079, in _get_relevant_documents
docs = self.vectorstore.similarity_search(query, **_kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\xieme\AppData\Roaming\Python\Python312\site-packages\langchain_community\vectorstores\milvus.py", line 663, in similarity_search
res = self.similarity_search_with_score(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\xieme\AppData\Roaming\Python\Python312\site-packages\langchain_community\vectorstores\milvus.py", line 734, in similarity_search_with_score
embedding = self.embedding_func.embed_query(query)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\langchain_openai\embeddings\base.py", line 638, in embed_query
return self.embed_documents([text], **kwargs)[0]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\langchain_openai\embeddings\base.py", line 590, in embed_documents
return self._get_len_safe_embeddings(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\langchain_openai\embeddings\base.py", line 478, in _get_len_safe_embeddings
response = self.client.create(
^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\openai\resources\embeddings.py", line 129, in create
return self._post(
^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\openai\_base_client.py", line 1242, in post
return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "D:\ProgramData\conda_envs\myenv\Lib\site-packages\openai\_base_client.py", line 1037, in request
raise self._make_status_error_from_response(err.response) from None
openai.BadRequestError: Error code: 400 - {'error': {'code': 'InvalidParameter', 'message': 'The parameter `input[0]` specified in the request are not valid: expected a string, but got `[73216 21507 40195 28190 21043 6271 222 82696]` instead. Request id: 02174999054139521b512236e6af9dc2992721609ebcc354653b4', 'param': 'input[0]', 'type': 'BadRequest'}}
