Paper QA Emulation Tool

We here emulate the workflow of the PaperQA system by Andrew White (https://thewhitelab.org/). This is a Retrieval Augmented Generation (RAG) application using a Map-Reduce model where we query our embedded index for based on a question, we then write summaries for each returned document based on relevance to the underlying question. Finally, we synthesize each summary into an essay presented as an answer to the original question.

---

source

PaperQAEmulationTool

 PaperQAEmulationTool (db:alhazen.utils.ceifns_db.Ceifns_LiteratureDb, llm
                       :Optional[langchain_core.language_models.chat_model
                       s.BaseChatModel]=None, slm:Optional[langchain_core.
                       language_models.chat_models.BaseChatModel]=None,
                       name:str='simple_qa_over_papers',
                       description:str='Runs a Map-Reduce model where we
                       write a short essay to answer a scientific question
                       based on a set of supporting documents.', args_sche
                       ma:Optional[Type[pydantic.v1.main.BaseModel]]=None,
                       return_direct:bool=True, verbose:bool=False, callba
                       cks:Union[List[langchain_core.callbacks.base.BaseCa
                       llbackHandler],langchain_core.callbacks.base.BaseCa
                       llbackManager,NoneType]=None, callback_manager:Opti
                       onal[langchain_core.callbacks.base.BaseCallbackMana
                       ger]=None, tags:Optional[List[str]]=None,
                       metadata:Optional[Dict[str,Any]]=None, handle_tool_
                       error:Union[bool,str,Callable[[langchain_core.tools
                       .ToolException],str],NoneType]=False, handle_valida
                       tion_error:Union[bool,str,Callable[[pydantic.v1.err
                       or_wrappers.ValidationError],str],NoneType]=False)

Write a short essay to answer a scientific question based documents from a preset collection.

---

source

PaperQAEmulationToolSchema

 PaperQAEmulationToolSchema (question:str,
                             n_sample_size:Optional[int]=None,
                             n_summary_size:Optional[int]=None,
                             collection_id:Optional[int]=None)

Create a new model by parsing and validating input data from keyword arguments.

Raises ValidationError if the input data cannot be parsed to form a valid model.