Retrieval-Augmented Generation and RAG Evaluation Tools

---

Documentation for package ‘ragR’ version 0.1.0

• DESCRIPTION file
• Package README

Help Pages

ragR-package	ragR: Retrieval-Augmented Generation and Evaluation Tools
api_chat_handler	API handler: chat with the RAG pipeline
api_clear_all_handler	API handler: clear the entire vector store
api_clear_handler	API handler: clear a vector-store collection
api_ragas_clear_handler	API handler: clear RAG evaluation files
api_ragas_handler	API handler: compute RAG evaluation metrics
api_ragas_report_handler	API handler: generate a RAG evaluation performance report
chunk_text_sentence	Chunk text into sentences (strict)
clear_qa_log	Clear the QA log on disk
clear_qa_metrics	Clear QA metrics on disk
compute_ragas_metrics	Compute RAGAS metrics
compute_ragas_metrics_llm	Compute RAGAS-style metrics (LLM scored)
generate_openai_chat	Generate a chat completion from OpenAI
generate_ragas_report	Generate a RAG evaluation performance report
generate_ragas_report_llm	Generate a RAG evaluation report using large language model scoring
get_openai_embeddings	Get embeddings from OpenAI
ingest_documents	Ingest documents into the vector store
load_qa_log	Load a QA log from disk
load_qa_metrics	Load QA metrics from disk
load_rag_config	Load RAG configuration
log_rag_interaction	Append one RAG interaction to the QA log
plot_ragas_means	Plot mean RAG evaluation metrics to a PNG
qa_log_empty	Create an empty QA log tibble
qa_metrics_empty	Create an empty QA metrics tibble
query_rag	Query the RAG pipeline
ragR	ragR: Retrieval-Augmented Generation and Evaluation Tools
save_qa_log	Save a QA log to disk
save_qa_metrics	Save QA metrics to disk
summarize_ragas	Summarize RAGAS metrics across QA pairs
vectorstore_clear_all	Delete all collections from the R-native vector store
vectorstore_delete_collection	Delete a collection from the R-native vector store
vectorstore_load	Load the R-native vector store
vectorstore_query	Query the R-native vector store for nearest neighbors
vectorstore_upsert	Upsert embeddings into the R-native vector store