New Retriever API #619
Merged
New Retriever API #619
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Thank you @AyushExel , I agree with the design you proposed. Users will be able to use different vector dbs by simply giving different keys. Should we put "retriever" folder in "contrib" as well? Maybe even move What do you think? @sonichi |
jackgerrits
pushed a commit
that referenced
this pull request
Oct 2, 2024
* Fix outstanding_tasks and cancellation bugs * formatting * Use separate except blocks --------- Co-authored-by: Eric Zhu <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I communicated these changes over discord
Why are these changes needed?
This PR creates a new Retriever API that makes plugging different retrievers simpler. It also removes hardcoding of particular vectorDBs and makes a centralized module that deal with optional vectordb dependencies
Retrieval functions were hardcoded. This PR introduces a new
Retrieverabstract interface that simply has 3 functions.Retriever.init_db()- To initialize connection/client or nothing if in serverless settingRetriever.ingest_data()- To upsert data in the dbRetriever.query()- To run queriesAny DB provider can simply implement these functions and register as a supported retriever for Autogen.
Note: This PR will not break the existing examples/notebooks.
This is still WIP but I tested the current chromadb implementation and it works well with the existing examples.
Related issue number
Closes #586 #416 and a few others
Checks