h,dZddlmZddlZddlZddlZddlmZmZddl m Z ddl m Z m Z ddlmZddlmZdd lmZmZdd lmZdd lmZdd lmZe d ddGddeZy)zCChain that interprets a prompt and executes python code to do math.) annotationsN)AnyOptional) deprecated)AsyncCallbackManagerForChainRunCallbackManagerForChainRun)BaseLanguageModel)BasePromptTemplate) ConfigDictmodel_validator)ChainLLMChain)PROMPTz0.2.13zThis class is deprecated and will be removed in langchain 1.0. See API reference for replacement: https://api.python.langchain.com/en/latest/chains/langchain.chains.llm_math.base.LLMMathChain.htmlz1.0)sincemessageremovalcLeZdZUdZded<dZded< eZded< d Zd ed <d Z d ed <e ddZ e de ddZeddZeddZddZ d dZ d!dZ d" d#dZ d" d$dZed%dZe ef d&dZy)' LLMMathChainaChain that interprets a prompt and executes python code to do math. Note: this class is deprecated. See below for a replacement implementation using LangGraph. The benefits of this implementation are: - Uses LLM tool calling features; - Support for both token-by-token and step-by-step streaming; - Support for checkpointing and memory of chat history; - Easier to modify or extend (e.g., with additional tools, structured responses, etc.) Install LangGraph with: .. code-block:: bash pip install -U langgraph .. code-block:: python import math from typing import Annotated, Sequence from langchain_core.messages import BaseMessage from langchain_core.runnables import RunnableConfig from langchain_core.tools import tool from langchain_openai import ChatOpenAI from langgraph.graph import END, StateGraph from langgraph.graph.message import add_messages from langgraph.prebuilt.tool_node import ToolNode import numexpr from typing_extensions import TypedDict @tool def calculator(expression: str) -> str: """Calculate expression using Python's numexpr library. Expression should be a single line mathematical expression that solves the problem. Examples: "37593 * 67" for "37593 times 67" "37593**(1/5)" for "37593^(1/5)" """ local_dict = {"pi": math.pi, "e": math.e} return str( numexpr.evaluate( expression.strip(), global_dict={}, # restrict access to globals local_dict=local_dict, # add common mathematical functions ) ) llm = ChatOpenAI(model="gpt-4o-mini", temperature=0) tools = [calculator] llm_with_tools = llm.bind_tools(tools, tool_choice="any") class ChainState(TypedDict): """LangGraph state.""" messages: Annotated[Sequence[BaseMessage], add_messages] async def acall_chain(state: ChainState, config: RunnableConfig): last_message = state["messages"][-1] response = await llm_with_tools.ainvoke(state["messages"], config) return {"messages": [response]} async def acall_model(state: ChainState, config: RunnableConfig): response = await llm.ainvoke(state["messages"], config) return {"messages": [response]} graph_builder = StateGraph(ChainState) graph_builder.add_node("call_tool", acall_chain) graph_builder.add_node("execute_tool", ToolNode(tools)) graph_builder.add_node("call_model", acall_model) graph_builder.set_entry_point("call_tool") graph_builder.add_edge("call_tool", "execute_tool") graph_builder.add_edge("execute_tool", "call_model") graph_builder.add_edge("call_model", END) chain = graph_builder.compile() .. code-block:: python example_query = "What is 551368 divided by 82" events = chain.astream( {"messages": [("user", example_query)]}, stream_mode="values", ) async for event in events: event["messages"][-1].pretty_print() .. code-block:: none ================================ Human Message ================================= What is 551368 divided by 82 ================================== Ai Message ================================== Tool Calls: calculator (call_MEiGXuJjJ7wGU4aOT86QuGJS) Call ID: call_MEiGXuJjJ7wGU4aOT86QuGJS Args: expression: 551368 / 82 ================================= Tool Message ================================= Name: calculator 6724.0 ================================== Ai Message ================================== 551368 divided by 82 equals 6724. Example: .. code-block:: python from langchain.chains import LLMMathChain from langchain_community.llms import OpenAI llm_math = LLMMathChain.from_llm(OpenAI()) r llm_chainNzOptional[BaseLanguageModel]llmr promptquestionstr input_keyanswer output_keyTforbid)arbitrary_types_allowedextrabefore)modec ddl}d|vrItjddd|vr.|d)|j dt }t |d| |d<|S#t$r}d}t||d}~wwxYw) NrzXLLMMathChain requires the numexpr package. Please install it with `pip install numexpr`.rzDirectly instantiating an LLMMathChain with an llm is deprecated. Please instantiate with llm_chain argument or using the from_llm class method.) stacklevelrrrr)numexpr ImportErrorwarningswarngetrr)clsvaluesr'emsgrs \/var/www/html/eduruby.in/venv/lib/python3.12/site-packages/langchain/chains/llm_math/base.pyraise_deprecationzLLMMathChain.raise_deprecations *  F? MM    &(VE]-FHf5&.6%=&P{# ! *@ c" )  *sA A1A,,A1c|jgS)z2Expect input key. :meta private: )rselfs r0 input_keyszLLMMathChain.input_keyss c|jgS)z3Expect output key. :meta private: )rr3s r0 output_keyszLLMMathChain.output_keyss   r6cddl} tjtjd}t |j |j i|}tjdd|S#t$r}d|d|d}t||d}~wwxYw) Nr)pir.) global_dict local_dictzLLMMathChain._evaluate("z") raised error: z4. Please try again with a valid numerical expressionz^\[|\]$) r'mathr:r.revaluatestrip Exception ValueErrorresub)r4 expressionr'r<outputr.r/s r0_evaluate_expressionz!LLMMathChain._evaluate_expressions ) $dff5J  $$& ")!Fvvj"f-- )*:,6GsKFF S/q (  )sA A)) B 2BB c|j|d|j|j}tjd|tj }|rc|j d}|j|}|jd|j|j|d|jd|z}n@|jd r|}n,d |vrd|jd d z}nd |}t||j|iS Ngreen)colorverbosez^```text(.*?)```z Answer: )rLyellowzAnswer: zAnswer:zunknown format from LLM: on_textrLr@rCsearchDOTALLgrouprG startswithsplitrBrr4 llm_output run_manager text_matchrErFrr/s r0_process_llm_resultz LLMMathChain._process_llm_results Jgt||L%%' YY2J J #))!,J..z:F    dll  C   h  M&(F  " "9 -F * $*"2"29"=b"AAF-j\:CS/ !((r6cRK|j|d|jd{|j}tjd|tj }|rs|j d}|j|}|jd|jd{|j|d|jd{d|z}n@|jd r|}n,d |vrd|jd d z}nd |}t||j|iS777\wrIrPrWs r0_aprocess_llm_resultz!LLMMathChain._aprocess_llm_results  !!*GT\\!RRR%%' YY2J J #))!,J..z:F%%lDLL%I I I%%fHdll%S S S&(F  " "9 -F * $*"2"29"=b"AAF-j\:CS/ !((! S J Ss5"D'D!A=D'"D##%D'D% AD'#D'%D'c|xstj}|j||j|jj ||jdg|j }|j||SNz ```output)rstop callbacks)rget_noop_managerrQrrpredict get_childr[r4inputsrY _run_managerrXs r0_callzLLMMathChain._callsz #S&@&Q&Q&S VDNN34^^++DNN+",,., '' LAAr6cJK|xstj}|j||jd{|jj ||jdg|j d{}|j||d{S7`7 7wr_)rrbrQrrapredictrdr]res r0_acallzLLMMathChain._acalls #X&E&V&V&X ""6$..#9:::>>22DNN+",,.3  ..z<HHH ; Is4:B#BAB#>B?B#B!B#B#!B#cy)Nllm_math_chainr3s r0 _chain_typezLLMMathChain._chain_type+sr6c 0t||}|dd|i|S)Nr&rrnr)r,rrkwargsrs r0from_llmzLLMMathChain.from_llm/s#V4 1Y1&11r6)r-dictreturnr)rtz list[str])rErrtr)rXrrYrrtdict[str, str])rXrrYrrtru)N)rfrurYz$Optional[CallbackManagerForChainRun]rtru)rfrurYz)Optional[AsyncCallbackManagerForChainRun]rtru)rtr)rr rr rqrrtr)__name__ __module__ __qualname____doc____annotations__rrrrrr model_configr classmethodr1propertyr5r8rGr[r]rhrkrorrrnr6r0rrssj'+C $+*!'F 'IIsJ $L (#$*  !!.,))0)  ).))5)  )4=A B B: B  B"BF I I? I  I  &,2 2#2 2  22r6r)ry __future__rr>rCr)typingrrlangchain_core._apirlangchain_core.callbacksrrlangchain_core.language_modelsr langchain_core.promptsr pydanticr r langchain.chains.baser langchain.chains.llmr langchain.chains.llm_math.promptrrrnr6r0rsdI"  *=50')3  m V25V2V2r6