add Weaviate memory adapter #95
Conversation
zainhas
requested review from
ashwinb,
yanxi0830,
hardikjshah,
dltn and
raghotham
as code owners
|
Hi @zainhas! Thank you for your pull request and welcome to our community. Action RequiredIn order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at [email protected]. Thanks! |
facebook-github-bot
added
the
CLA Signed
|
Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks! |
llama_stack/apis/memory/client.py
Outdated
| @@ -42,7 +42,11 @@ async def get_memory_bank(self, bank_id: str) -> Optional[MemoryBank]: | |||
| params={ | |||
| "bank_id": bank_id, | |||
| }, | |||
| headers={"Content-Type": "application/json"}, | |||
| headers={ | |||
There was a problem hiding this comment.
can we revert these changes from here, since this is a pretty generic client -- we will figure out how to get a bunch of these examples situated.
| async def get_adapter_impl(config: RemoteProviderConfig, _deps): | ||
| from .weaviate import WeaviateMemoryAdapter | ||
|
|
||
| impl = WeaviateMemoryAdapter(config.url, config.username, config.password) |
There was a problem hiding this comment.
the config does not define url / username / password but instead only defines a collection?
There was a problem hiding this comment.
the cluster url and api_key for weaviate will come in via the WeaviateRequestProviderData class, in this case should I just get rid of the WeaviateConfig class?
| async def initialize(self) -> None: | ||
| try: | ||
|
|
||
| request_provider_data = get_request_provider_data() |
There was a problem hiding this comment.
provider data isn't present at this point -- it is only provided at the time of the request. you should initialize the client on every client call and if we need a cache of clients then, we'd need to build that.
There was a problem hiding this comment.
modified the method to return an initialized client. following this I initialize the client on every client call using the method.
|
@ashwinb addressed comments. Please let me know how we want to test this end to end. I can provide a weaviate cloud url and api_key to you. |
|
|
||
| data_objects = [] | ||
| for i, chunk in enumerate(chunks): | ||
| print(f"Adding chunk #{i} tokens={chunk.token_count}") |
There was a problem hiding this comment.
curious: are we ok with print logging?
| if request_provider_data is not None: | ||
| assert isinstance(request_provider_data, WeaviateRequestProviderData) | ||
|
|
||
| print(f"WEAVIATE API KEY: {request_provider_data.weaviate_api_key}") |
There was a problem hiding this comment.
do we really want to print the api key into the console?
Why this PR
We want to setup Weaviate as a remote vector db provider for llama-stack.
What is in the PR