Create a RAG Module

Factory function to create a Retrieval-Augmented Generation module. RAG modules automatically retrieve relevant context from a document store before generating responses.

Usage

rag_module(
  signature,
  store = NULL,
  retriever = NULL,
  k = 5L,
  context_format = "relevant_context",
  config = list(),
  chat = NULL
)

Arguments

signature

A signature string or Signature object defining inputs/outputs.

store

Optional ragnar store for document retrieval.

retriever

Optional custom retriever function. Should accept (query, k) and return a character vector of documents.

k

Number of documents to retrieve (default 5).

context_format

Field name for the context in the prompt (default "relevant_context"). This field is automatically added to inputs.

config

Optional configuration list.

chat

Optional ellmer Chat object.

Value

A RAGModule object.

Examples

if (FALSE) { # \dontrun{
# With ragnar store
store <- ragnar::ragnar_store_create(documents)
mod <- rag_module(
  "question, relevant_context -> answer",
  store = store,
  k = 3
)

result <- run(mod, question = "What is the capital of France?")

# With custom retriever
my_retriever <- function(query, k) {
  # Custom retrieval logic
  c("Document 1 content", "Document 2 content")
}

mod <- rag_module(
  "query, relevant_context -> response",
  retriever = my_retriever
)
} # }