Skip to content

neurotrace.core.hippocampus.memory_orchestrator

neurotrace.core.hippocampus.memory_orchestrator

MemoryOrchestrator

Manages both short-term and long-term memory for Neurotrace agents.

Source code in neurotrace/core/hippocampus/memory_orchestrator.py 
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
class MemoryOrchestrator:
    """Manages both short-term and long-term memory for Neurotrace agents."""

    def __init__(
        self,
        llm: Union[BaseLLM, BaseChatModel],
        graph_store: GraphStore,
        vector_store: VectorStore,
    ):
        self.llm = llm
        self._graph_indexer = GraphTripletIndexer(llm)
        self._graph_memory_adapter = GraphMemoryAdapter(llm, graph_store, self._graph_indexer)
        self._vector_memory_adapter = VectorMemoryAdapter(vector_store)

    def save_in_graph_memory(self, summary: str, tags: List[str] = None) -> str:
        """Saves a summary in graph memory."""
        # todo: interface this with Message class
        self._graph_memory_adapter.add_conversation(summary, tags=tags)
        return "Graph memory saved."

    def save_in_vector_memory(self, summary: str, tags: List[str] = None) -> str:
        """Saves a summary in vector memory."""
        message = Message(role=Role.AI.value, content=summary, metadata=MessageMetadata(tags=tags))
        self._vector_memory_adapter.add_messages([message])
        return "Vector memory saved."

    def search_vector_memory(self, query: str, k: int = 5) -> List[Message]:
        return self._vector_memory_adapter.search(query, k)

    def search_graph_memory(self, query: str) -> str:
        return self._graph_memory_adapter.ask_graph(query)["result"]

save_in_graph_memory(summary, tags=None)

Saves a summary in graph memory.

Source code in neurotrace/core/hippocampus/memory_orchestrator.py 
28
29
30
31
32
def save_in_graph_memory(self, summary: str, tags: List[str] = None) -> str:
    """Saves a summary in graph memory."""
    # todo: interface this with Message class
    self._graph_memory_adapter.add_conversation(summary, tags=tags)
    return "Graph memory saved."

save_in_vector_memory(summary, tags=None)

Saves a summary in vector memory.

Source code in neurotrace/core/hippocampus/memory_orchestrator.py 
34
35
36
37
38
def save_in_vector_memory(self, summary: str, tags: List[str] = None) -> str:
    """Saves a summary in vector memory."""
    message = Message(role=Role.AI.value, content=summary, metadata=MessageMetadata(tags=tags))
    self._vector_memory_adapter.add_messages([message])
    return "Vector memory saved."