INNER CODE UNIT · Python

split_doc

junruxiong/IncarnaMind · docs2db.py:145

def split_doc(
    doc: List[Document], chunk_size: int, chunk_overlap: int, chunk_idx_name: str
):
    """Splits a document into smaller chunks based on the provided size and overlap.

    Args:
        doc (Document): Document to be split.
        chunk_size (int): Size of each chunk.
        chunk_overlap (int): Overlap between adjacent chunks.
        chunk_idx_name (str): Metadata key for storing chunk indices.

    Returns:
        list: List of Document objects representing the chunks.
    """
    data_splitter = RecursiveCharacterTextSplitter(
        chunk_size=chunk_size,
        chunk_overlap=chunk_overlap,
        length_function=tiktoken_len,

View source record →

📰 Research Paper
Loading…
⏳ Fetching content…