MCPcopy Create free account

hub / github.com/SaiAkhil066/CORTEX-AI-SUPER-RAG / functions

Functions21 in github.com/SaiAkhil066/CORTEX-AI-SUPER-RAG

⨍ Functions 21◇ Types & classes 0
↓ 4 callersFunction_thinking_html
(text: str, live: bool)
app.py:455
↓ 3 callersFunction_ollama_generate
Single non-streamed Ollama completion. Returns '' on failure.
utils/advanced_rag.py:12
↓ 2 callersFunction_source_html
(sources: list)
app.py:469
↓ 1 callersFunction_apply_contextual_retrieval
Prepend an LLM-generated situating sentence to each chunk (Anthropic technique).
utils/doc_handler.py:26
↓ 1 callersFunctionbuild_knowledge_graph
(docs)
utils/build_graph.py:5
↓ 1 callersFunctioncheck_semantic_cache
Return a cached entry if a semantically similar question was asked before.
app.py:703
↓ 1 callersFunctioncontextualize_chunk
Generate a short situating context for a chunk given its document summary.
utils/advanced_rag.py:31
↓ 1 callersFunctioncosine_similarity
Cosine similarity between two equal-length float lists.
utils/advanced_rag.py:92
↓ 1 callersFunctionexpand_query
HyDE: append a hypothetical answer to improve recall.
utils/retriever_pipeline.py:12
↓ 1 callersFunctiongenerate_query_variants
Return [original] + up to n reworded search queries.
utils/advanced_rag.py:48
↓ 1 callersFunctiongenerate_response
(prompt_text: str, model: str)
app.py:535
↓ 1 callersFunctionget_ollama_models
()
app.py:63
↓ 1 callersFunctiongrade_document
Return True if the document is relevant to the query (LLM judge).
utils/advanced_rag.py:78
↓ 1 callersFunctionload_cache_embeddings
Lightweight embeddings client used only for semantic caching.
app.py:54
↓ 1 callersFunctionload_reranker
()
app.py:43
↓ 1 callersFunctionprocess_documents
(uploaded_files, reranker, embedding_model, base_url, llm_model=None, enable_contextual=
utils/doc_handler.py:52
↓ 1 callersFunctionreciprocal_rank_fusion
Merge multiple ranked Document lists into one via RRF.
utils/advanced_rag.py:64
↓ 1 callersFunctionreset_documents
()
utils/doc_handler.py:18
↓ 1 callersFunctionretrieve_documents
(query, uri, model, chat_history="")
utils/retriever_pipeline.py:26
↓ 1 callersFunctionretrieve_from_graph
Return actual document text chunks related to graph nodes matching the query.
utils/build_graph.py:15
↓ 1 callersFunctionstore_in_cache
(query_emb, query, answer, thinking, sources)
app.py:724