hhĂ ddlZddlZddlZddlmZddlmZmZmZm Z m Z m Z m Z m Z mZmZmZmZddlmZmZddlmZddlmZddlmZmZmZmZmZmZm Z m!Z!ddl"m#Z#dd l$m%Z%dd l&m'Z'dd l(m)Z)m*Z*dd l+m,Z,m-Z-m.Z.dd l/m0Z0m1Z1m2Z2ddl3m4Z4ddl5m6Z6m7Z7ddl8m9Z9ddl:m;Z;mZ>m?Z?m@Z@mAZAmBZBejCeDZEdedeFfdZGdeeHe eeHe eefffdeHfdZIde eHefdefdZJGddeZKdS)N) itemgetter) Any AsyncIteratorCallableDictIteratorListMappingOptionalSequenceTypeUnioncast)AsyncCallbackManagerForLLMRunCallbackManagerForLLMRun)LanguageModelInput) BaseChatModel) AIMessageAIMessageChunk BaseMessage ChatMessageFunctionMessage HumanMessage SystemMessage ToolMessage) UsageMetadata)tool_call_chunk)OutputParserLike)JsonOutputKeyToolsParserPydanticToolsParser)ChatGenerationChatGenerationChunk ChatResult)Runnable RunnableMapRunnablePassthrough)BaseTool)convert_to_secret_strget_from_dict_or_envconvert_to_openai_tool) get_fieldsis_basemodel_subclass) BaseModel ConfigDictField SecretStrmodel_validatormessagereturnc@t|tr|j|jd}nt|tr d|jd}nt|t r]d|jd}t |jdkr:|jd}|dtj |ddd |d <d |d <nft|ttfr8d t|j|j p|jdd}nt!d||S)z@Convert a message to a dictionary that can be passed to the API.)rolecontentuser assistantrnameargsF) ensure_ascii)r: arguments function_callNr7function)r6r7r:zGot unknown type ) isinstancerr6r7rrlen tool_callsjsondumpsrr_create_tool_contentr:additional_kwargsget TypeError)r3 message_dict tool_calls s/var/www/FlaskApp/flask-venv/lib/python3.11/site-packages/langchain_community/chat_models/baidu_qianfan_endpoint.pyconvert_message_to_dictrL<s<';''7 ' II G\ * *7 &7?CC GY ' '7 +HH w! " "a ' '*1-I!&)!Z &(9NNN--L ) '+L # Go{; < <7+GO<<LIG$=$A$A&$I$I  5G55666 r7c.t|trk ttj|tr|Stjd|iS#tj$rtjd|icYSwxYwtjd|iS)z$Convert tool content to dict scheme. tool_result)r@strrCloadsdictrDJSONDecodeError)r7s rKrErEYs'3 4 8$*W--t44 <z='":;;;# 8 8 8:}g677 7 7 7 8z='2333s(AA%A>=A>_dictc |ddpd}i}|drthoughtsbody finish_reasonidobject search_info)rZ request_idr\r]r?)typer?r[rB)r7rFusage prompt_tokensrcompletion_tokens total_tokens) input_tokens output_tokensrc) rGrRpoprPuuiduuid4rrusage_metadata)rTr7rFmsg_additional_kwargsretr`s rK_convert_dict_to_messagerlgsii"%%+G+- yy!!?,d53I.J.JK *?; ; ; o . 2 2: > > >G599VR00F4EF '++OR@@$T* $$Xr22%))-<< _b11  1B1F1F R2 2 o. #-11/2FF$*,,''  / l+ /   C "%%gt444u *?A66))$7;;>155    JrMc eZdZUdZeeZeee fe d< eeZ eee fe d< dZ e e d<eddZ eee d < edd Zeee d < d Zeee d < eddZeee d< dZeee d< dZeee d< dZeee d< edZeee d< dZeee d< edZededed e fd!Ze d eee fffd" Z!e d efd#Z"e d eee ffd$Z#d%e$e%d&e d eee ffd'Z& d4d%e$e%d(ee$ed)ee'd&e d e(f d*Z) d4d%e$e%d(ee$ed)ee*d&e d e(f d+Z+ d4d%e$e%d(ee$ed)ee'd&e d e,e-f d,Z. d4d%e$e%d(ee$ed)ee*d&e d e/e-f d-Z0d.e1e2eee fe3e4e5e6fd&e d e7e8e%fffd/ Z9d d0d1e2ee3e4fd2ed&e d e7e8e2ee4fffd3Z:xZ;S)5QianfanChatEndpointuBaidu Qianfan chat model integration. Setup: Install ``qianfan`` and set environment variables ``QIANFAN_AK``, ``QIANFAN_SK``. .. code-block:: bash pip install qianfan export QIANFAN_AK="your-api-key" export QIANFAN_SK="your-secret_key" Key init args — completion params: model: str Name of Qianfan model to use. temperature: Optional[float] Sampling temperature. endpoint: Optional[str] Endpoint of the Qianfan LLM top_p: Optional[float] What probability mass to use. Key init args — client params: timeout: Optional[int] Timeout for requests. api_key: Optional[str] Qianfan API KEY. If not passed in will be read from env var QIANFAN_AK. secret_key: Optional[str] Qianfan SECRET KEY. If not passed in will be read from env var QIANFAN_SK. See full list of supported init args and their descriptions in the params section. Instantiate: .. code-block:: python from langchain_community.chat_models import QianfanChatEndpoint qianfan_chat = QianfanChatEndpoint( model="ERNIE-3.5-8K", temperature=0.2, timeout=30, # api_key="...", # secret_key="...", # top_p="...", # other params... ) Invoke: .. code-block:: python messages = [ ("system", "你是一名专业的翻译家，可以将用户的中文翻译为英文。"), ("human", "我喜欢编程。"), ] qianfan_chat.invoke(messages) .. code-block:: python AIMessage(content='I enjoy programming.', additional_kwargs={'finish_reason': 'normal', 'request_id': 'as-7848zeqn1c', 'object': 'chat.completion', 'search_info': []}, response_metadata={'token_usage': {'prompt_tokens': 16, 'completion_tokens': 4, 'total_tokens': 20}, 'model_name': 'ERNIE-3.5-8K', 'finish_reason': 'normal', 'id': 'as-7848zeqn1c', 'object': 'chat.completion', 'created': 1719153606, 'result': 'I enjoy programming.', 'is_truncated': False, 'need_clear_history': False, 'usage': {'prompt_tokens': 16, 'completion_tokens': 4, 'total_tokens': 20}}, id='run-4bca0c10-5043-456b-a5be-2f62a980f3f0-0') Stream: .. code-block:: python for chunk in qianfan_chat.stream(messages): print(chunk) .. code-block:: python content='I enjoy' response_metadata={'finish_reason': 'normal', 'request_id': 'as-yz0yz1w1rq', 'object': 'chat.completion', 'search_info': []} id='run-0fa9da50-003e-4a26-ba16-dbfe96249b8b' role='assistant' content=' programming.' response_metadata={'finish_reason': 'normal', 'request_id': 'as-yz0yz1w1rq', 'object': 'chat.completion', 'search_info': []} id='run-0fa9da50-003e-4a26-ba16-dbfe96249b8b' role='assistant' .. code-block:: python stream = chat.stream(messages) full = next(stream) for chunk in stream: full += chunk full .. code-block:: AIMessageChunk(content='I enjoy programming.', response_metadata={'finish_reason': 'normalnormal', 'request_id': 'as-p63cnn3ppnas-p63cnn3ppn', 'object': 'chat.completionchat.completion', 'search_info': []}, id='run-09a8cbbd-5ded-4529-981d-5bc9d1206404') Async: .. code-block:: python await qianfan_chat.ainvoke(messages) # stream: # async for chunk in qianfan_chat.astream(messages): # print(chunk) # batch: # await qianfan_chat.abatch([messages]) .. code-block:: python [AIMessage(content='I enjoy programming.', additional_kwargs={'finish_reason': 'normal', 'request_id': 'as-mpqa8qa1qb', 'object': 'chat.completion', 'search_info': []}, response_metadata={'token_usage': {'prompt_tokens': 16, 'completion_tokens': 4, 'total_tokens': 20}, 'model_name': 'ERNIE-3.5-8K', 'finish_reason': 'normal', 'id': 'as-mpqa8qa1qb', 'object': 'chat.completion', 'created': 1719155120, 'result': 'I enjoy programming.', 'is_truncated': False, 'need_clear_history': False, 'usage': {'prompt_tokens': 16, 'completion_tokens': 4, 'total_tokens': 20}}, id='run-443b2231-08f9-4725-b807-b77d0507ad44-0')] Tool calling: .. code-block:: python from pydantic import BaseModel, Field class GetWeather(BaseModel): '''Get the current weather in a given location''' location: str = Field( ..., description="The city and state, e.g. San Francisco, CA" ) class GetPopulation(BaseModel): '''Get the current population in a given location''' location: str = Field( ..., description="The city and state, e.g. San Francisco, CA" ) chat_with_tools = qianfan_chat.bind_tools([GetWeather, GetPopulation]) ai_msg = chat_with_tools.invoke( "Which city is hotter today and which is bigger: LA or NY?" ) ai_msg.tool_calls .. code-block:: python [ { 'name': 'GetWeather', 'args': {'location': 'Los Angeles, CA'}, 'id': '533e5f63-a3dc-40f2-9d9c-22b1feee62e0' } ] Structured output: .. code-block:: python from typing import Optional from pydantic import BaseModel, Field class Joke(BaseModel): '''Joke to tell user.''' setup: str = Field(description="The setup of the joke") punchline: str = Field(description="The punchline to the joke") rating: Optional[int] = Field(description="How funny the joke is, from 1 to 10") structured_chat = qianfan_chat.with_structured_output(Joke) structured_chat.invoke("Tell me a joke about cats") .. code-block:: python Joke( setup='A cat is sitting in front of a mirror and sees another cat. What does the cat think?', punchline="The cat doesn't think it's another cat, it thinks it's another mirror.", rating=None ) Response metadata .. code-block:: python ai_msg = qianfan_chat.invoke(messages) ai_msg.response_metadata .. code-block:: python { 'token_usage': { 'prompt_tokens': 16, 'completion_tokens': 4, 'total_tokens': 20}, 'model_name': 'ERNIE-3.5-8K', 'finish_reason': 'normal', 'id': 'as-qbzwtydqmi', 'object': 'chat.completion', 'created': 1719158153, 'result': 'I enjoy programming.', 'is_truncated': False, 'need_clear_history': False, 'usage': { 'prompt_tokens': 16, 'completion_tokens': 4, 'total_tokens': 20 } } )default_factory init_kwargs model_kwargsNclientapi_key)defaultalias qianfan_ak secret_key qianfan_skF streaming<timeout)rurequest_timeoutg?top_pgffffff? temperature penalty_scorertmodelendpointT)populate_by_namebefore)modevaluesr4cJtt|ddgdd|d<tt|ddgdd|d<d t|D}||i|d i|d |d d }|ddkr|d|d<|ddkr|d|d<|d|ddkr |d|d< ddl}|jdi||d<n#t$rtdwxYw|S)Nrvrs QIANFAN_AKrWrrxrw QIANFAN_SKc2i|]\}}|j ||jSNr).0r:fields rK z.s2   e}( %-(((rMrprry)rstreamakskrrrrzGqianfan package not found, please install it with `pip install qianfan`) r(r)r,itemsupdaterGget_secret_valueqianfanChatCompletion ImportError)clsrdefault_valuesparamsrs rKvalidate_environmentz(QianfanChatEndpoint.validate_environments  5 y1<     | 5 |4lB     |   )#4466    f%%% jj++ #''00$((55    ,  0 0 2 2b 8 8!,/@@BBF4L ,  0 0 2 2b 8 8!,/@@BBF4L   z * * 6z*b00!/ !;F:   NNN5w5????F8     (    s 1FF cLi|j|jdtjS)N)rr)rrsuper_identifying_params)self __class__s rKrz'QianfanChatEndpoint._identifying_paramss0 =4:>> gg)  rMcdS)zReturn type of chat_model.zbaidu-qianfan-chatr)rs rK _llm_typezQianfanChatEndpoint._llm_types $#rMcr|j|j|j|j|j|j|jd}i||jS)z3Get the default parameters for calling Qianfan API.)rrrr|r}r~r)rrryr|r}r~rrq)r normal_paramss rK_default_paramsz#QianfanChatEndpoint._default_paramssKZ n#3Z+!/  6-54#455rMmessageskwargsc dd|Di}dt|DD]<}d|vrd|d<|dxxtt||jdzz cc<=i||j|S)a Converts a list of messages into a dictionary containing the message content and default parameters. Args: messages (List[BaseMessage]): The list of messages. **kwargs (Any): Optional arguments to add additional parameters to the resulting dictionary. Returns: Dict[str, Any]: A dictionary containing the message content and default parameters. rcVg|]&}t|tt|'Sr)r@rrL)rms rK zBQianfanChatEndpoint._convert_prompt_msg_params..s@!!]33'**rMcBg|]\}}t|t|Sr)r@r)rirs rKrzBQianfanChatEndpoint._convert_prompt_msg_params..s+TTT1z!]7S7ST!TTTrMsystemrW ) enumeraterrPr7r)rrr messages_dictrs rK_convert_prompt_msg_paramsz.QianfanChatEndpoint._convert_prompt_msg_paramss( !) UT ( 3 3TTT M MA},,*, h' ( # # #tC!1D'E'E'L L # # # #  "   rMstop run_managerc z|jrd}i}d}|j|||fi|D]B}|j|jn|}||jz }t |jt r |jj}Ct|i|} t| td} t| g|pi|j dS|j |fi|} || d<|jjd i| } t!| } t| d di| d i} | d i} | |j d}t| g|S) aCall out to an qianfan models endpoint for each generation with a prompt. Args: messages: The messages to pass into the model. stop: Optional list of stop words to use when generating. Returns: The string generated by the model. Example: .. code-block:: python response = qianfan_model.invoke("Tell me a joke.") rWNr7rFrirrZr3generation_info token_usage model_name generations llm_outputrZrYr`r)ry_streamrtextr@r3rrirr!rRr#rrrrdorlrG)rrrrr completionchat_generation_inforichunklc_msggenrresponse_payloadrrs rK _generatezQianfanChatEndpoint._generates$ > J)+ 6:N%hkLLVLL B B,8))-% ej( em^<<B%*]%AN""$-F ! $6 : : :C E#1#7R"&* 10DDVDDv)4;>33F33)*:;;"&&vr22   '**7B77 %0 KK se CCCCrMcK|jrd}i}d}|j|||fi|23d{V}|j|jn|}||jz }t |jt r |jj}H6t|i|} t| td} t| g|pi|j dS|j |fi|} || d<|jjd i| d{V} t!| } g} t| ddi| d i} | | | d i}||j d}t| |S) NrWrrrrrrrZrYr`r)ry_astreamrrr@r3rrirr!rRr#rrrradorlrGappend)rrrrrrrrirrrrrrrrs rK _ageneratezQianfanChatEndpoint._agenerate3s  > J)+ 6:N,t}Xt[SSFSS B B B B B B Be,8))-% ej( em^<<B%*]%AN T""$-F ! $6 : : :C E#1#7R"&* 10DDVDDv!0!:!:6!:!:::::::)*:;; "&&vr22    3&**7B77 %0 KK kjIIIIsA(c +K|j|fi|}||d<d|d<|jjd i|D]}|rt|}|jdi}t |dt|jd||j d|j D|j } |r| | j | | VdS) NrTrr>rVr9c |g|]9}t|dtj|d|dd:Sr:r;r[N)r:r;r[indexrrCrDrtcs rKrz/QianfanChatEndpoint._stream..[***!# ,%'Z%)Z6 %;%;#%d8&* ***rMr7r6rFritool_call_chunksrr3rrr) rrrrrlrFrGr"rr7rirBon_llm_new_tokenr rrrrrrresmsgrFrs rKrzQianfanChatEndpoint._streamjs&10DDVDDvx!4;>++F++  C .s33$'$9$=$=or$R$R!+X* # (*;'*'9**'*n*** %($9#&J0050III 3  rMc K|j|fi|}||d<d|d<|jjd i|d{V23d{V}|rt|}|jdi}t |dt|jd||j d|j D|j } |r"| | j | d{V| WV6dS) NrTrr>rVr9c |g|]9}t|dtj|d|dd:Srrrs rKrz0QianfanChatEndpoint._astream..rrMrrrr) rrrrrlrFrGr"rr7rirBrrrs rKrzQianfanChatEndpoint._astreamsf10DDVDDvx.t{8888888888       # .s33$'$9$=$=or$R$R!+X* # (*;'*'9**'*n*** %($9#&P%66uz6OOOOOOOOO 3988sCtoolsc Rd|D}tjdd|i|S)a3Bind tool-like objects to this chat model. Assumes model is compatible with OpenAI tool-calling API. Args: tools: A list of tool definitions to bind to this chat model. Can be a dictionary, pydantic model, callable, or BaseTool. Pydantic models, callables, and BaseTools will be automatically converted to their schema dictionary representation. **kwargs: Any additional parameters to pass to the :class:`~langchain.runnable.Runnable` constructor. c8g|]}t|dS)r?r*)rtools rKrz2QianfanChatEndpoint.bind_tools..s&VVV1$77 CVVVrM functionsr)rbind)rrrformatted_toolsrs rK bind_toolszQianfanChatEndpoint.bind_toolss;$WVPUVVVuww|@@o@@@@rM) include_rawschemarc |rtd|t|tot|}||g}|rt |gd}n,t |dd}t|d}|rhtj td|zd }tj d } | | gd } t|| zS||zS)a*Model wrapper that returns outputs formatted to match the given schema. Args: schema: The output schema as a dict or a Pydantic class. If a Pydantic class then the model output will be an object of that class. If a dict then the model output will be a dict. With a Pydantic class the returned attributes will be validated, whereas with a dict they will not be. If `method` is "function_calling" and `schema` is a dict, then the dict must match the OpenAI function-calling spec. include_raw: If False then only the parsed structured output is returned. If an error occurs during model output parsing it will be raised. If True then both the raw model response (a BaseMessage) and the parsed model response will be returned. If an error occurs during output parsing it will be caught and returned as well. The final output is always a dict with keys "raw", "parsed", and "parsing_error". Returns: A Runnable that takes any ChatModel input and returns as output: If include_raw is True then a dict with keys: raw: BaseMessage parsed: Optional[_DictOrPydantic] parsing_error: Optional[BaseException] If include_raw is False then just _DictOrPydantic is returned, where _DictOrPydantic depends on the schema: If schema is a Pydantic class then _DictOrPydantic is the Pydantic class. If schema is a dict then _DictOrPydantic is a dict. Example: Function-calling, Pydantic schema (method="function_calling", include_raw=False): .. code-block:: python from langchain_mistralai import QianfanChatEndpoint from pydantic import BaseModel class AnswerWithJustification(BaseModel): '''An answer to the user question along with justification for the answer.''' answer: str justification: str llm = QianfanChatEndpoint(endpoint="ernie-3.5-8k-0329") structured_llm = llm.with_structured_output(AnswerWithJustification) structured_llm.invoke("What weighs more a pound of bricks or a pound of feathers") # -> AnswerWithJustification( # answer='They weigh the same', # justification='Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ.' # ) Example: Function-calling, Pydantic schema (method="function_calling", include_raw=True): .. code-block:: python from langchain_mistralai import QianfanChatEndpoint from pydantic import BaseModel class AnswerWithJustification(BaseModel): '''An answer to the user question along with justification for the answer.''' answer: str justification: str llm = QianfanChatEndpoint(endpoint="ernie-3.5-8k-0329") structured_llm = llm.with_structured_output(AnswerWithJustification, include_raw=True) structured_llm.invoke("What weighs more a pound of bricks or a pound of feathers") # -> { # 'raw': AIMessage(content='', additional_kwargs={'tool_calls': [{'id': 'call_Ao02pnFYXD6GN1yzc0uXPsvF', 'function': {'arguments': '{"answer":"They weigh the same.","justification":"Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ."}', 'name': 'AnswerWithJustification'}, 'type': 'function'}]}), # 'parsed': AnswerWithJustification(answer='They weigh the same.', justification='Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume or density of the objects may differ.'), # 'parsing_error': None # } Example: Function-calling, dict schema (method="function_calling", include_raw=False): .. code-block:: python from langchain_mistralai import QianfanChatEndpoint from pydantic import BaseModel from langchain_core.utils.function_calling import convert_to_openai_tool class AnswerWithJustification(BaseModel): '''An answer to the user question along with justification for the answer.''' answer: str justification: str dict_schema = convert_to_openai_tool(AnswerWithJustification) llm = QianfanChatEndpoint(endpoint="ernie-3.5-8k-0329") structured_llm = llm.with_structured_output(dict_schema) structured_llm.invoke("What weighs more a pound of bricks or a pound of feathers") # -> { # 'answer': 'They weigh the same', # 'justification': 'Both a pound of bricks and a pound of feathers weigh one pound. The weight is the same, but the volume and density of the two substances differ.' # } zReceived unsupported arguments T)rfirst_tool_onlyr?r:)key_namerrawcdSrr_s rKz.BsRVrM)parsed parsing_errorcdSrrrs rKrz.DsdrM)rr) exception_key)r) ValueErrorr@r_r-rr r+rr&assignrwith_fallbacksr%) rrrris_pydantic_schemallm output_parserr parser_assign parser_noneparser_with_fallbacks rKwith_structured_outputz*QianfanChatEndpoint.with_structured_outputsBP  IGvGGHH H'55W:OPV:W:Woovh''  .Ah $///MM .f55jA&IH4!4M  '/6!%((=8M.4NNKKKK#0#?#? _$@$$ 3'''*>> >& &rM)NN)<__name__ __module__ __qualname____doc__r0rRrprrPr__annotations__rqrrrvr r1rxryboolr|intr}floatr~rrrr/ model_configr2 classmethodrpropertyrrrr rrrr#rrrrr"rrrr rr r.rr'r$rrr __classcell__)rs@rKrnrnsu}}~#(%"="="=Kc3h===@$)5#>#>#>L$sCx.>>>8FC',eD &J&J&JJ#JJJ&+eD &M&M&MJ#MMM %Ix~%%%/%*U2Y%?%?%?OXc]???0 E8E?   '#'K%'''+%&M8E?&&&!5...E8C=...#Hhsm"""E:L_(###)$)3)))[$#)V T#s(^     X $3$$$X$ 6c3h 6 6 6X 6# {## #  c3h # # # # P%):> =D=D{#=DtCy!=D67 =D  =D  =D=D=D=DD%)?C 5J5J{#5JtCy!5J;< 5J  5J  5J5J5J5Jt%):> ##{##tCy!#67 #  # % & ####P%)?C ##{##tCy!#;< #  # * + ####JAd38nd9oxQRSAA $k1 2 AAAAAA2" A'A'A'dDO+,A' A'  A' $eD)O&<< = A'A'A'A'A'A'A'A'rMrn)LrCloggingrgoperatorrtypingrrrrrr r r r r rrlangchain_core.callbacksrrlangchain_core.language_modelsr*langchain_core.language_models.chat_modelsrlangchain_core.messagesrrrrrrrrlangchain_core.messages.airlangchain_core.messages.toolr"langchain_core.output_parsers.baser*langchain_core.output_parsers.openai_toolsrr langchain_core.outputsr!r"r#langchain_core.runnablesr$r%r&langchain_core.toolsr'langchain_core.utilsr(r)%langchain_core.utils.function_callingr+langchain_core.utils.pydanticr,r-pydanticr.r/r0r1r2 getLoggerrloggerrRrLrPrErlrnrrMrKr#s                              >=====DDDDDD                    544444888888??????SRRRRRRRRROOOOOOOOOO))))))LLLLLLLLHHHHHHKKKKKKKK  8 $ $[T: 4%T%T#s(^8K2L-M(M"N 4SV 4 4 4 4/GCH$5/)////dq 'q 'q 'q 'q '-q 'q 'q 'q 'q 'rM