Jeff Huber
Jeff Huber
The latest version now wipes access-token and client for subsequent API requests: https://github.com/lynndylanhurley/devise_token_auth/blob/00fa5f4507da806f66664787745086ce404d95b4/app/controllers/devise_token_auth/concerns/set_user_by_token.rb#L18 ```ruby # keep track of request duration def set_request_start @request_started_at = Time.zone.now @used_auth_by_token = true # initialize...
Works with Chroma's dockerized backend. Feedback welcome! I tried to stay very similar to the `hnsw-node` implementation. The PR is open on chroma's repo to add this JS SDK -...
There are few things that can be improved in the integration: - [ ] Add tests - this will require mocking the chroma backend responses to have any value and...
The basic vector store example started breaking because `Document` required `not None` for metadata, but Chroma stores metadata as `None` if none is provided. This creates a fallback which fixes...
This PR adds full JS client parity with the python client and also breaks out tests into individual files. Should likely be landed after https://github.com/chroma-core/chroma/pull/297 - so that CI can...
Currently we use `sentence-transformers` as the default embedding model. However this means that it, and a lot of it's deps are included in the project. Additionally it downloads the model...
Many users look for the chroma container inside docker and can't find it. We should push releases there as well as `ghcr` Additionally- the docs need to be updated to...
Improved docstrings for https://github.com/chroma-core/docs/pull/28
We need to handle - easy getting started with batteries included - real production deployments with client and backend separation, with concerns around package size and build times. We've thought...
Chroma currently is column-oriented for storage and retrieval across the `python` and `js` APIs. For example: ```python collection.add( documents=["This is a document", "This is another document"], metadatas=[{"source": "my_source"}, {"source": "my_source"}],...