agents.rag.multi_agent_rag.simple_enhanced_workflows¶
Simple Enhanced Multi-Agent RAG Workflows.
Clean implementation of advanced RAG patterns without complex dependencies.
Classes¶
Agent that grades retrieved documents for relevance. |
|
Agent that decides if requerying is needed based on document grades. |
|
Simple Corrective RAG implementation using sequential processing. |
|
Simple HYDE RAG agent that generates hypothetical documents before retrieval. |
Functions¶
|
Factory function to create simple RAG workflows. |
Module Contents¶
- class agents.rag.multi_agent_rag.simple_enhanced_workflows.DocumentGradingAgent¶
Bases:
haive.agents.base.agent.Agent
Agent that grades retrieved documents for relevance.
- build_graph()¶
Build graph that grades each retrieved document.
- Return type:
haive.core.graph.state_graph.base_graph2.BaseGraph
- class agents.rag.multi_agent_rag.simple_enhanced_workflows.RequeryDecisionAgent¶
Bases:
haive.agents.base.agent.Agent
Agent that decides if requerying is needed based on document grades.
- build_graph()¶
Build graph that analyzes grades and decides on requerying.
- Return type:
haive.core.graph.state_graph.base_graph2.BaseGraph
- class agents.rag.multi_agent_rag.simple_enhanced_workflows.SimpleCorrectiveRAGAgent(documents=None, **kwargs)¶
Bases:
haive.agents.multi.base.SequentialAgent
Simple Corrective RAG implementation using sequential processing.
Init .
- Parameters:
config – [TODO: Add description]
documents (list[langchain_core.documents.Document] | None)
- class agents.rag.multi_agent_rag.simple_enhanced_workflows.SimpleHYDERAGAgent(documents=None, **kwargs)¶
Bases:
haive.agents.multi.base.SequentialAgent
Simple HYDE RAG agent that generates hypothetical documents before retrieval.
Init .
- Parameters:
config – [TODO: Add description]
documents (list[langchain_core.documents.Document] | None)
- agents.rag.multi_agent_rag.simple_enhanced_workflows.create_simple_rag_workflow(workflow_type='crag', documents=None, **kwargs)¶
Factory function to create simple RAG workflows.