Anton Troynikov
Anton Troynikov
This PR adds Chroma to the vector database examples in the OpenAI cookbook. [Chroma](docs.trychroma.com) is a lightweight, developer-friendly embeddings store which aims to get A.I application developers up and running...
Because `add` allows duplicate / redundant IDs. When `update` gets called and finds redundant IDs, it fails with an obtuse error. This is an artifact of allowing redundant IDs, we...
## Description of changes *Summarize the changes made by this PR.* Adds `upsert` as a method on `Collection`. Upsert updates entries in the embeddings store if their ids already exist,...
In several places, we convert `UUID` objects back and forth from and to strings. This isn't great, and it's hard to keep track of what they should be, when. We...
## TODO: - Blog post - Documentation - ~~Resolve merge conflicts~~ ~~This change requires a change to the index metric (from `l2` to `ip`). I'll wait for that to be...
Up-to-date version of #72 This PR replaces Pinecone with Chroma. This considerably simplifies getting babyagi up and running, since it now requires one less external API. Users no longer need...
## Description of changes This PR adds embedding functions to the test suite. This allows us to test the by-document testing flow, throughout the test suite. It introduces a simple...
### Describe the problem The embedding functions module is a single file which is getting really unwieldy. https://github.com/chroma-core/chroma/blob/main/chromadb/utils/embedding_functions.py For example, it's difficult to land PR's like https://github.com/chroma-core/chroma/pull/1447 because they need...
### Describe the problem Currently we only set the `ef_search` parameter when we init the index, usually on creation of a `Collection`. However, this parameter is actually per-query at the...