Index

A

ACKNOWLEDGEMENT_CUES (in module rag.core.passage) answer (rag.core.query.QueryResult attribute) ask() (in module rag.core.query) (rag.core.backends.langchain.backend.LangChainBackend method) (rag.core.backends.llamaindex.backend.LlamaIndexBackend method) (rag.core.backends.protocol.RagBackend method)

ask_llamaindex() (in module rag.core.query) AUTHOR_QUESTION_CUES (in module rag.core.passage) AUTHOR_SYNTHESIS_PROMPT (in module rag.core.query)

B

build_embed_model() (in module rag.core.ingest)

build_llm() (in module rag.core.query) build_query_engine() (in module rag.core.query)

C

CATALOG_FILE_NAME (in module rag.core.catalog) catalog_from_documents() (in module rag.core.catalog) catalog_path() (in module rag.core.catalog) CatalogJsonKey (class in rag.core.catalog) CHROMA_COLLECTION_ENV (in module rag.core.config) chroma_dir (rag.core.config.RagConfig attribute) (rag.core.ingest.IngestResult attribute) CHROMA_DIR_ENV (in module rag.core.config) chunk_overlap (rag.core.config.RagConfig attribute) CHUNK_OVERLAP_ENV (in module rag.core.config)

chunk_size (rag.core.config.RagConfig attribute) CHUNK_SIZE_ENV (in module rag.core.config) Citation (class in rag.core.citations) citations (rag.core.query.QueryResult attribute) citations_from_nodes() (in module rag.core.citations) citations_markdown (rag.core.query.QueryResult attribute) classify_query_scope() (in module rag.core.catalog) collection_name (rag.core.config.RagConfig attribute) (rag.core.ingest.IngestResult attribute) CORPUS (rag.core.catalog.QueryScope attribute) CORPUS_SYNTHESIS_PROMPT (in module rag.core.query)

D

DEFAULT_CHUNK_OVERLAP (in module rag.core.config) DEFAULT_CHUNK_SIZE (in module rag.core.config) DEFAULT_COLLECTION_NAME (in module rag.core.config) DEFAULT_EMBED_MODEL (in module rag.core.config)

DEFAULT_LLM_MODEL (in module rag.core.config) DEFAULT_SIMILARITY_TOP_K (in module rag.core.config) DEFAULT_SNIPPET_CHARS (in module rag.core.citations) documents (rag.core.ingest.IngestResult attribute)

E

EMBED_MODEL_ENV (in module rag.core.config)

embed_model_name (rag.core.config.RagConfig attribute) ensure_dirs() (rag.core.config.RagConfig method)

F

FILE_NAME (rag.core.catalog.CatalogJsonKey attribute) file_name (rag.core.catalog.PaperCatalogEntry attribute) (rag.core.citations.Citation attribute) FILE_NAME (rag.core.citations.FileMetadataKey attribute) FILE_PATH (rag.core.citations.FileMetadataKey attribute)

FileMetadataKey (class in rag.core.citations) FILENAME (rag.core.citations.FileMetadataKey attribute) find_repo_root() (in module rag.core.config) format_citations() (in module rag.core.citations) from_env() (rag.core.config.RagConfig class method) from_json() (rag.core.catalog.PaperCatalog class method)

G

get_backend() (in module rag.core.backends.registry)

I

index_dir_for() (in module rag.core.strategy) INDEXES_DIR_NAME (in module rag.core.strategy) IndexMissingError ingest() (rag.core.backends.langchain.backend.LangChainBackend method) (rag.core.backends.llamaindex.backend.LlamaIndexBackend method) (rag.core.backends.protocol.RagBackend method) INGEST_ALL (in module rag.core.strategy) ingest_llamaindex() (in module rag.core.ingest) ingest_papers() (in module rag.core.ingest)

ingestible() (rag.core.strategy.RagStrategy class method) IngestNotSupportedError IngestResult (class in rag.core.ingest) is_acknowledgement_text() (in module rag.core.passage) is_author_question() (in module rag.core.passage) is_proceedings_boilerplate_text() (in module rag.core.passage) is_prose_text() (in module rag.core.text_quality) is_ready() (rag.core.backends.langchain.backend.LangChainBackend method) (rag.core.backends.llamaindex.backend.LlamaIndexBackend method) (rag.core.backends.protocol.RagBackend method)

K

keep (rag.core.query.ProseNodePostprocessor attribute)

L

LANGCHAIN (rag.core.strategy.RagStrategy attribute) LangChainBackend (class in rag.core.backends.langchain.backend) list_pdf_paths() (in module rag.core.loaders) LLAMAINDEX (rag.core.strategy.RagStrategy attribute) LlamaIndexBackend (class in rag.core.backends.llamaindex.backend)

llm_model_name (rag.core.config.RagConfig attribute) load_paper_catalog() (in module rag.core.catalog) load_pdf_documents() (in module rag.core.ingest) load_pdf_pages() (in module rag.core.loaders) load_repo_dotenv() (in module rag.core.config) load_vector_index() (in module rag.core.index)

M

MAX_GLUED_RATIO (in module rag.core.text_quality) MAX_SALAD_RATIO (in module rag.core.text_quality) MAX_SINGLE_LETTER_RATIO (in module rag.core.text_quality) MAX_SNAKE_RATIO (in module rag.core.text_quality) metadata (rag.core.loaders.PageDocument attribute) MIN_PROSE_CHARS (in module rag.core.text_quality) MIN_REAL_WORDS (in module rag.core.text_quality) MIN_STUTTER_TOKEN_CHARS (in module rag.core.text_quality) MIN_WORDS_IF_SPARSE (in module rag.core.text_quality) module rag rag.core rag.core.backends rag.core.backends.langchain rag.core.backends.langchain.backend rag.core.backends.llamaindex rag.core.backends.llamaindex.backend rag.core.backends.protocol rag.core.backends.registry rag.core.catalog rag.core.citations rag.core.config rag.core.index rag.core.ingest rag.core.loaders rag.core.passage rag.core.query rag.core.strategy rag.core.text_quality

N

nodes (rag.core.ingest.IngestResult attribute)

O

open_chroma_collection() (in module rag.core.index)

OPENAI_API_KEY_ENV (in module rag.core.config) OPENAI_MODEL_ENV (in module rag.core.config)

P

page (rag.core.citations.Citation attribute) PAGE (rag.core.citations.PageMetadataKey attribute) page_from_mapping() (in module rag.core.passage) PAGE_LABEL (rag.core.citations.PageMetadataKey attribute) PAGE_NUMBER (rag.core.citations.PageMetadataKey attribute) PageDocument (class in rag.core.loaders) PageMetadataKey (class in rag.core.citations) PAPER (rag.core.catalog.QueryScope attribute) PaperCatalog (class in rag.core.catalog)

PaperCatalogEntry (class in rag.core.catalog) PAPERS (rag.core.catalog.CatalogJsonKey attribute) papers (rag.core.catalog.PaperCatalog attribute) papers_dir (rag.core.config.RagConfig attribute) PAPERS_DIR_ENV (in module rag.core.config) PAPERS_REFINE_TEMPLATE (in module rag.core.query) PAPERS_TEXT_QA_TEMPLATE (in module rag.core.query) PROCEEDINGS_BOILERPLATE_CUES (in module rag.core.passage) ProseNodePostprocessor (class in rag.core.query)

Q

QA_GROUNDING_RULES (in module rag.core.passage)

QueryResult (class in rag.core.query) QueryScope (class in rag.core.catalog)

R

rag module rag.core module rag.core.backends module rag.core.backends.langchain module rag.core.backends.langchain.backend module rag.core.backends.llamaindex module rag.core.backends.llamaindex.backend module rag.core.backends.protocol module rag.core.backends.registry module rag.core.catalog module rag.core.citations module

rag.core.config module rag.core.index module rag.core.ingest module rag.core.loaders module rag.core.passage module rag.core.query module rag.core.strategy module rag.core.text_quality module RAG_STRATEGY_ENV (in module rag.core.strategy) RagBackend (class in rag.core.backends.protocol) RagConfig (class in rag.core.config) RagStrategy (class in rag.core.strategy) require_openai_api_key() (in module rag.core.query) RETRIEVAL_OVERFETCH (in module rag.core.query)

S

score (rag.core.citations.Citation attribute) similarity_top_k (rag.core.config.RagConfig attribute) SIMILARITY_TOP_K_ENV (in module rag.core.config)

snippet (rag.core.citations.Citation attribute) SOURCE (rag.core.citations.FileMetadataKey attribute) strategy (rag.core.config.RagConfig attribute)

T

text (rag.core.loaders.PageDocument attribute) TITLE (rag.core.catalog.CatalogJsonKey attribute) title (rag.core.catalog.PaperCatalogEntry attribute)

title_from_text() (in module rag.core.catalog) TITLE_SNIPPET_CHARS (in module rag.core.catalog) to_json() (rag.core.catalog.PaperCatalog method) to_markdown() (rag.core.catalog.PaperCatalog method)

U

UNKNOWN_FILE_NAME (in module rag.core.citations)

W

write_paper_catalog() (in module rag.core.catalog)