Parent Document Retriver

Parent Document retriever is a technique where large documents are split into smaller pieces, called "child chunks." These chunks are stored in a way that lets the system find and compare specific parts of a document with a user’s query. The large document, or "parent," is still kept but is only retrieved if one of its child chunks is relevant to the query.

Reference: Parent Document Retriver

Initial Setup

Indexing

Parent Child Text Spliting

Retriever

RAG Chain

Preparing Data for Evaluation

Evaluation in Athina AI

We will use Context Recall eval here. It Measures the extent to which the retrieved context aligns with the expected response. Please refer to our documentation for further details