"""
Farnsworth RAG Module
Self-evolving retrieval-augmented generation with:
- Hybrid semantic + keyword search with RRF fusion
- RL-optimized retrieval strategy learning
- Adaptive chunking and reranking
- Query success feedback loops
Q1 2025 (v0.2.0) Features:
- Hybrid Search v2 with intent classification
- Multi-hop retrieval for complex questions
- Context compression for token efficiency
"""
from farnsworth.rag.embeddings import EmbeddingManager
from farnsworth.rag.hybrid_retriever import HybridRetriever
from farnsworth.rag.document_processor import DocumentProcessor
from farnsworth.rag.self_refining_rag import SelfRefiningRAG
# Q1 2025 Features
from farnsworth.rag.hybrid_search_v2 import (
HybridSearchV2,
QueryIntent,
QueryAnalysis,
SearchHop,
AttributedResult,
HybridSearchResult,
)
from farnsworth.rag.context_compression import (
ContextCompressor,
CompressionLevel,
ContentPriority,
ContextBlock,
CompressionResult,
ContextBudget,
)
__all__ = [
# Core RAG components
"EmbeddingManager",
"HybridRetriever",
"DocumentProcessor",
"SelfRefiningRAG",
# Q1 2025: Hybrid Search v2
"HybridSearchV2",
"QueryIntent",
"QueryAnalysis",
"SearchHop",
"AttributedResult",
"HybridSearchResult",
# Q1 2025: Context Compression
"ContextCompressor",
"CompressionLevel",
"ContentPriority",
"ContextBlock",
"CompressionResult",
"ContextBudget",
]
