Closed walbercardoso closed 1 week ago
For some functions it is not necessary to use run_in_executor, since there are async functions in langchain for qdrant.
@danny-avila I would appreciate it if you could take a look at the implementation and let me know if everything is ok.
New variables need to be add at the .env
QDRANT_API_KEY=
QDRANT_HOST=
VECTOR_DB= #qdrant or pgvector (default if not `provided)`
COLLECTION_NAME=
EMBEDDINGS_DIMENSION=
Can you please check this if it need some change? If it's ok I'm planning to add other vector stores
there are some conflicts due to the mongodb atlas merge
No problem. I can fix that.
@danny-avila Conflicts removed. Can you check it again?
@danny-avila Conflicts removed. Can you check it again?
hi @walbercardoso there are some new conflicts. I've added enums to help manage the addition of vector stores. Also, I would like to see all routes working with screenshots and configuration settings for your setup, because the last addition of a vector store does not seem fully functional.
Thanks
hi @danny-avila . Conflicts have been fixed.
Some screenshots:
Embed:
Route
Qdrant dash:
Delete attach file before starting query:
Route:
Query doc:
Route:
Delete file:
Route:
Configuration params at .env
QDRANT_API_KEY=
QDRANT_HOST=""
VECTOR_DB_TYPE="qdrant"
QDRANT_VECTOR_COLLECTION="test"
QDRANT_EMBEDDINGS_DIMENSION=768
Please let me know if this is enough
Replacing with https://github.com/danny-avila/rag_api/pull/73
Add qdrant vector store to improve vectors database