Chunk

datapizza.type.Chunk dataclass

A class for storing the chunk response from a client.

__init__

__init__(id, text, embeddings=None, metadata=None)

Initialize a Chunk object.

Parameters:

Name	Type	Description	Default
id	str	The id of the chunk.	required
text	str	The text of the chunk.	required
embeddings	list[Embedding]	The embeddings of the chunk. Defaults to [].	None
metadata	dict	The metadata of the chunk. Defaults to {}.	None

Overview

The Chunk class represents a unit of text content that has been segmented from a larger document. It's a fundamental data structure in datapizza-ai used throughout the RAG pipeline for text processing, embedding, and retrieval operations. Serializable: Can be easily stored and retrieved from databases