Title / Abstract Mapping Tool

Langchain tools that execute zero-shot classification tasks over a local database of title/abstracts from papers previously imported into our database.

source

BaseTitleAbstractMappingTool

 BaseTitleAbstractMappingTool
                               (db:alhazen.utils.ceifns_db.Ceifns_Literatu
                               reDb, llm:Optional[langchain_core.language_
                               models.chat_models.BaseChatModel]=None, slm
                               :Optional[langchain_core.language_models.ch
                               at_models.BaseChatModel]=None,
                               name:str='tiab_mapping',
                               description:str='Runs a specified document
                               mapping pipeline over papers in a
                               collection.', args_schema:Optional[Type[pyd
                               antic.v1.main.BaseModel]]=None,
                               return_direct:bool=True,
                               verbose:bool=False, callbacks:Union[List[la
                               ngchain_core.callbacks.base.BaseCallbackHan
                               dler],langchain_core.callbacks.base.BaseCal
                               lbackManager,NoneType]=None, callback_manag
                               er:Optional[langchain_core.callbacks.base.B
                               aseCallbackManager]=None,
                               tags:Optional[List[str]]=None,
                               metadata:Optional[Dict[str,Any]]=None, hand
                               le_tool_error:Union[bool,str,Callable[[lang
                               chain_core.tools.ToolException],str],NoneTy
                               pe]=False, handle_validation_error:Union[bo
                               ol,str,Callable[[pydantic.v1.error_wrappers
                               .ValidationError],str],NoneType]=False)

Runs a specified document mapping pipeline over papers in a collection.


source

TitleAbstractMappingToolSchema

 TitleAbstractMappingToolSchema (collection_id:str,
                                 repeat_run:Optional[bool]=None,
                                 run_label:Optional[str]=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.


source

TitleAbstractDiscourseMappingTool

 TitleAbstractDiscourseMappingTool
                                    (db:alhazen.utils.ceifns_db.Ceifns_Lit
                                    eratureDb, llm:Optional[langchain_core
                                    .language_models.chat_models.BaseChatM
                                    odel]=None, slm:Optional[langchain_cor
                                    e.language_models.chat_models.BaseChat
                                    Model]=None, name:str='tiab_one_doc_cl
                                    assification', description:str='Runs
                                    through the text of each title +
                                    abstract and split them based on
                                    discourse.', args_schema:Optional[Type
                                    [pydantic.v1.main.BaseModel]]=None,
                                    return_direct:bool=True,
                                    verbose:bool=False, callbacks:Union[Li
                                    st[langchain_core.callbacks.base.BaseC
                                    allbackHandler],langchain_core.callbac
                                    ks.base.BaseCallbackManager,NoneType]=
                                    None, callback_manager:Optional[langch
                                    ain_core.callbacks.base.BaseCallbackMa
                                    nager]=None,
                                    tags:Optional[List[str]]=None,
                                    metadata:Optional[Dict[str,Any]]=None,
                                    handle_tool_error:Union[bool,str,Calla
                                    ble[[langchain_core.tools.ToolExceptio
                                    n],str],NoneType]=False, handle_valida
                                    tion_error:Union[bool,str,Callable[[py
                                    dantic.v1.error_wrappers.ValidationErr
                                    or],str],NoneType]=False)

Runs through the text of each title + abstract and split them based on discourse.