phoenix-oss/llama-stack
1
0
Fork 0
forked from phoenix-oss/llama-stack-mirror
Code Pull requests Releases Packages 2 Activity Actions

migrate memory banks to Resource and new registration (#411)

Browse source 
* migrate memory banks to Resource and new registration

* address feedback

* address feedback

* fix tests

* pgvector fix

* pgvector fix v2

* remove auto discovery

* change register signature to make params required

* update client

* client fix

* use annotated union to parse

* remove base MemoryBank inheritence

---------

Co-authored-by: Dinesh Yeduguru <dineshyv@fb.com>
This commit is contained in:
Dinesh Yeduguru 2024-11-11 17:10:44 -08:00 • committed by GitHub
parent 6b9850e11b
commit 38cce97597
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
19 changed files with 240 additions and 129 deletions
Download patch file Download diff file Expand all files Collapse all files

21
llama_stack/providers/tests/memory/fixtures.py
View file

@ -10,11 +10,10 @@ import tempfile
import pytest
import pytest_asyncio
from llama_stack.distribution.datatypes import Api, Provider
from llama_stack.distribution.datatypes import Api, Provider, RemoteProviderConfig
from llama_stack.providers.inline.memory.faiss import FaissImplConfig
from llama_stack.providers.remote.memory.pgvector import PGVectorConfig
from llama_stack.providers.remote.memory.weaviate import WeaviateConfig
from llama_stack.providers.tests.resolver import resolve_impls_for_test_v2
from llama_stack.providers.utils.kvstore import SqliteKVStoreConfig
from ..conftest import ProviderFixture, remote_stack_fixture
@ -78,7 +77,23 @@ def memory_weaviate() -> ProviderFixture:
)
MEMORY_FIXTURES = ["meta_reference", "pgvector", "weaviate", "remote"]
@pytest.fixture(scope="session")
def memory_chroma() -> ProviderFixture:
return ProviderFixture(
providers=[
Provider(
provider_id="chroma",
provider_type="remote::chromadb",
config=RemoteProviderConfig(
host=get_env_or_fail("CHROMA_HOST"),
port=get_env_or_fail("CHROMA_PORT"),
).model_dump(),
)
]
)
MEMORY_FIXTURES = ["meta_reference", "pgvector", "weaviate", "remote", "chroma"]
@pytest_asyncio.fixture(scope="session")

40
llama_stack/providers/tests/memory/test_memory.py
View file

@ -8,6 +8,7 @@ import pytest
from llama_stack.apis.memory import * # noqa: F403
from llama_stack.distribution.datatypes import * # noqa: F403
from llama_stack.apis.memory_banks.memory_banks import VectorMemoryBankParams
# How to run this test:
#
@ -43,14 +44,15 @@ def sample_documents():
async def register_memory_bank(banks_impl: MemoryBanks):
bank = VectorMemoryBankDef(
identifier="test_bank",
embedding_model="all-MiniLM-L6-v2",
chunk_size_in_tokens=512,
overlap_size_in_tokens=64,
)
await banks_impl.register_memory_bank(bank)
return await banks_impl.register_memory_bank(
memory_bank_id="test_bank",
params=VectorMemoryBankParams(
embedding_model="all-MiniLM-L6-v2",
chunk_size_in_tokens=512,
overlap_size_in_tokens=64,
),
)
class TestMemory:
@ -68,20 +70,28 @@ class TestMemory:
# NOTE: this needs you to ensure that you are starting from a clean state
# but so far we don't have an unregister API unfortunately, so be careful
_, banks_impl = memory_stack
bank = VectorMemoryBankDef(
identifier="test_bank_no_provider",
embedding_model="all-MiniLM-L6-v2",
chunk_size_in_tokens=512,
overlap_size_in_tokens=64,
)
await banks_impl.register_memory_bank(bank)
bank = await banks_impl.register_memory_bank(
memory_bank_id="test_bank_no_provider",
params=VectorMemoryBankParams(
embedding_model="all-MiniLM-L6-v2",
chunk_size_in_tokens=512,
overlap_size_in_tokens=64,
),
)
response = await banks_impl.list_memory_banks()
assert isinstance(response, list)
assert len(response) == 1
# register same memory bank with same id again will fail
await banks_impl.register_memory_bank(bank)
await banks_impl.register_memory_bank(
memory_bank_id="test_bank_no_provider",
params=VectorMemoryBankParams(
embedding_model="all-MiniLM-L6-v2",
chunk_size_in_tokens=512,
overlap_size_in_tokens=64,
),
)
response = await banks_impl.list_memory_banks()
assert isinstance(response, list)
assert len(response) == 1