autorag.nodes.passageaugmenter package

Submodules

autorag.nodes.passageaugmenter.base module

autorag.nodes.passageaugmenter.base.passage_augmenter_node(func)

autorag.nodes.passageaugmenter.pass_passage_augmenter module

autorag.nodes.passageaugmenter.pass_passage_augmenter.pass_passage_augmenter(ids_list: List[List[str]], contents_list: List[List[str]], scores_list: List[List[float]], **kwargs)

Do not perform augmentation. Return given passages, scores, and ids as is.

autorag.nodes.passageaugmenter.prev_next_augmenter module

autorag.nodes.passageaugmenter.prev_next_augmenter.prev_next_augmenter(ids_list: List[List[str]], corpus_df: DataFrame, num_passages: int = 1, mode: str = 'both') → List[List[str]]

Add passages before and/or after the retrieved passage. For more information, visit https://docs.llamaindex.ai/en/stable/examples/node_postprocessor/PrevNextPostprocessorDemo/.

Parameters:

• ids_list – The list of lists of ids retrieved

• corpus_df – The corpus dataframe

• num_passages – The number of passages to add before and after the retrieved passage Default is 1.

• mode – The mode of augmentation ‘prev’: add passages before the retrieved passage ‘next’: add passages after the retrieved passage ‘both’: add passages before and after the retrieved passage Default is ‘next’.

Returns:

The list of lists of augmented ids

autorag.nodes.passageaugmenter.prev_next_augmenter.prev_next_augmenter_pure(ids: List[str], corpus_df: DataFrame, mode: str, num_passages: int)

autorag.nodes.passageaugmenter.run module

autorag.nodes.passageaugmenter.run.run_passage_augmenter_node(modules: List[Callable], module_params: List[Dict], previous_result: DataFrame, node_line_dir: str, strategies: Dict) → DataFrame

Module contents