o 粪g$@sdZddlmZddlmZmZmZmZddlm Z ddl m Z ddl m Z ddlmZmZddlmZdd lmZmZmZdd lmZdd lmZdddZe ddddGdddeZdS)zLCombine documents by doing a first pass and then refining on more documents.) annotations)AnyDictListTuple) deprecated) Callbacks)Document)BasePromptTemplateformat_documentPromptTemplate) ConfigDictFieldmodel_validator)BaseCombineDocumentsChain)LLMChainreturnr cCstdgddS)N page_contentz{page_content})input_variablestemplater rre/var/www/html/chatdoc2/venv/lib/python3.10/site-packages/langchain/chains/combine_documents/refine.py_get_default_document_promptsrz0.3.1z1.0zThis class is deprecated. Please see the migration guide here for a recommended replacement: https://python.langchain.com/docs/versions/migrating_chains/refine_docs_chain/)sinceremovalmessagecseZdZUdZded< ded< ded< ded< eedZd ed < d Zd ed < e d6fdd Z e dddZ e dded7ddZe dded7ddZ d8d9d%d&Z d8d9d'd(Zd:d+d,Zd;d0d1ZdRefineDocumentsChaina& Combine documents by doing a first pass and then refining on more documents. This algorithm first calls `initial_llm_chain` on the first document, passing that first document in with the variable name `document_variable_name`, and produces a new variable with the variable name `initial_response_name`. Then, it loops over every remaining document. This is called the "refine" step. It calls `refine_llm_chain`, passing in that document with the variable name `document_variable_name` as well as the previous response with the variable name `initial_response_name`. Example: .. code-block:: python from langchain.chains import RefineDocumentsChain, LLMChain from langchain_core.prompts import PromptTemplate from langchain_community.llms import OpenAI # This controls how each document will be formatted. Specifically, # it will be passed to `format_document` - see that function for more # details. document_prompt = PromptTemplate( input_variables=["page_content"], template="{page_content}" ) document_variable_name = "context" llm = OpenAI() # The prompt here should take as an input variable the # `document_variable_name` prompt = PromptTemplate.from_template( "Summarize this content: {context}" ) initial_llm_chain = LLMChain(llm=llm, prompt=prompt) initial_response_name = "prev_response" # The prompt here should take as an input variable the # `document_variable_name` as well as `initial_response_name` prompt_refine = PromptTemplate.from_template( "Here's your first summary: {prev_response}. " "Now add to it based on the following context: {context}" ) refine_llm_chain = LLMChain(llm=llm, prompt=prompt_refine) chain = RefineDocumentsChain( initial_llm_chain=initial_llm_chain, refine_llm_chain=refine_llm_chain, document_prompt=document_prompt, document_variable_name=document_variable_name, initial_response_name=initial_response_name, ) rinitial_llm_chainrefine_llm_chainstrdocument_variable_nameinitial_response_name)default_factoryr document_promptFboolreturn_intermediate_stepsr List[str]cstj}|jr |dg}|S)z2Expect input key. :meta private: intermediate_steps)super output_keysr&)self _output_keys __class__rrr*ds z RefineDocumentsChain.output_keysTforbid)arbitrary_types_allowedextrabefore)modevaluesrrcCsd|vr |d|d<|d=|S)zFor backwards compatibility.return_refine_stepsr&r)clsr4rrrget_return_intermediate_stepsts z2RefineDocumentsChain.get_return_intermediate_stepscCspd|vrtd|djj}d|vr$t|dkr |d|d<|Std|d|vr6td|dd||S) z4Get default document variable name, if not provided.rz"initial_llm_chain must be providedr!rzWdocument_variable_name must be provided if there are multiple llm_chain input_variableszdocument_variable_name z- was not found in llm_chain input_variables: ) ValueErrorpromptrlen)r6r4llm_chain_variablesrrr"get_default_document_variable_name}s"      z7RefineDocumentsChain.get_default_document_variable_nameNdocsList[Document] callbacksrkwargsTuple[str, dict]c Ks|j|fi|}|jjdd|i|}|g}|ddD]}|||}i||}|jjdd|i|}||q|||S)aCombine by mapping first chain over all, then stuffing into final chain. Args: docs: List of documents to combine callbacks: Callbacks to be passed through **kwargs: additional parameters to be passed to LLM calls (like other input variables besides the documents) Returns: The first element returned is the single string output. The second element returned is a dictionary of other keys to return. r@r8Nr)_construct_initial_inputsrpredict_construct_refine_inputsrappend_construct_result r+r>r@rAinputsres refine_stepsdoc base_inputsrrr combine_docss    z!RefineDocumentsChain.combine_docsc s|j|fi|}|jjdd|i|IdH}|g}|ddD]!}|||}i||}|jjdd|i|IdH}||q!|||S)aAsync combine by mapping a first chain over all, then stuffing into a final chain. Args: docs: List of documents to combine callbacks: Callbacks to be passed through **kwargs: additional parameters to be passed to LLM calls (like other input variables besides the documents) Returns: The first element returned is the single string output. The second element returned is a dictionary of other keys to return. r@Nr8r)rCrapredictrErrFrGrHrrr acombine_docss    z"RefineDocumentsChain.acombine_docsrKrJcCs"|jr d|i}||fSi}||fS)Nr()r&)r+rKrJextra_return_dictrrrrGs z&RefineDocumentsChain._construct_resultrLr Dict[str, Any]cCs|jt||j|j|iSN)r!r r$r")r+rLrJrrrrEsz-RefineDocumentsChain._construct_refine_inputsc s\d|dji|djfdd|jjD}|j|jjdi|i}i||}|S)Nrrcsi|]}||qSrr).0k base_inforr szBRefineDocumentsChain._construct_initial_inputs..r)rupdatemetadatar$rr!format)r+r>rA document_inforMrIrrVrrCs z.RefineDocumentsChain._construct_initial_inputscCsdS)Nrefine_documents_chainr)r+rrr _chain_typesz RefineDocumentsChain._chain_type)rr')r4rrrrS)r>r?r@rrArrrB)rKr'rJr rrB)rLr rJr rrR)r>r?rArrrR)rr )__name__ __module__ __qualname____doc____annotations__rrr$r&propertyr*r model_configr classmethodr7r=rNrPrGrErCr^ __classcell__rrr-rrsJ  2       rN)rr )rb __future__rtypingrrrrlangchain_core._apirlangchain_core.callbacksrlangchain_core.documentsr langchain_core.promptsr r langchain_core.prompts.promptr pydanticrrr'langchain.chains.combine_documents.baserlangchain.chains.llmrrrrrrrs$