phoenix-oss/llama-stack-mirror
0
0
Fork 1
mirror of https://github.com/meta-llama/llama-stack.git synced 2025-10-16 14:57:20 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

migrate memory banks to Resource and new registration

Browse source
This commit is contained in:
Dinesh Yeduguru 2024-11-08 15:45:26 -08:00
parent b4416b72fd
commit c82f13bf9e
16 changed files with 178 additions and 104 deletions
Download patch file Download diff file Expand all files Collapse all files

9
llama_stack/providers/remote/memory/qdrant/qdrant.py
View file

@ -12,6 +12,7 @@ from numpy.typing import NDArray
from qdrant_client import AsyncQdrantClient, models
from qdrant_client.models import PointStruct
from llama_stack.apis.memory_banks import * # noqa: F403
from llama_stack.providers.datatypes import MemoryBanksProtocolPrivate
from llama_stack.apis.memory import * # noqa: F403
@ -112,11 +113,11 @@ class QdrantVectorMemoryAdapter(Memory, MemoryBanksProtocolPrivate):
async def register_memory_bank(
self,
memory_bank: MemoryBankDef,
memory_bank: MemoryBank,
) -> None:
assert (
memory_bank.type == MemoryBankType.vector.value
), f"Only vector banks are supported {memory_bank.type}"
memory_bank.memory_bank_type == MemoryBankType.vector
), f"Only vector banks are supported {memory_bank.memory_bank_type}"
index = BankWithIndex(
bank=memory_bank,
@ -125,7 +126,7 @@ class QdrantVectorMemoryAdapter(Memory, MemoryBanksProtocolPrivate):
self.cache[memory_bank.identifier] = index
async def list_memory_banks(self) -> List[MemoryBankDef]:
async def list_memory_banks(self) -> List[MemoryBank]:
# Qdrant doesn't have collection level metadata to store the bank properties
# So we only return from the cache value
return [i.bank for i in self.cache.values()]

12
llama_stack/providers/remote/memory/weaviate/weaviate.py
View file

@ -114,11 +114,11 @@ class WeaviateMemoryAdapter(
async def register_memory_bank(
self,
memory_bank: MemoryBankDef,
memory_bank: MemoryBank,
) -> None:
assert (
memory_bank.type == MemoryBankType.vector.value
), f"Only vector banks are supported {memory_bank.type}"
memory_bank.memory_bank_type == MemoryBankType.vector
), f"Only vector banks are supported {memory_bank.memory_bank_type}"
client = self._get_client()
@ -141,7 +141,7 @@ class WeaviateMemoryAdapter(
)
self.cache[memory_bank.identifier] = index
async def list_memory_banks(self) -> List[MemoryBankDef]:
async def list_memory_banks(self) -> List[MemoryBank]:
# TODO: right now the Llama Stack is the source of truth for these banks. That is
# not ideal. It should be Weaviate which is the source of truth. Unfortunately,
# list() happens at Stack startup when the Weaviate client (credentials) is not
@ -157,8 +157,8 @@ class WeaviateMemoryAdapter(
raise ValueError(f"Bank {bank_id} not found")
client = self._get_client()
if not client.collections.exists(bank_id):
raise ValueError(f"Collection with name `{bank_id}` not found")
if not client.collections.exists(bank.identifier):
raise ValueError(f"Collection with name `{bank.identifier}` not found")
index = BankWithIndex(
bank=bank,