New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Interface for remote search index #1524
Comments
cc @ellisonbg |
@hwchase17 It is not clear which interface to use for building applications with langchain and Kendra. Kendra is mainly a search index, so applications might want to integrate it in data augmented retrieval and question answering use cases using Langchain. It seems like there are 2 interfaces of interest here.
We might need a common interface that has these 2 interfaces, but need your advice on the next steps. Thanks for your help. |
This **partially** addresses #1524, but it's also useful for some of our use cases. This `DocstoreFn` allows to lookup a document given a function that accepts the `search` string without the need to implement a custom `Docstore`. This could be useful when: * you don't want to implement a `Docstore` just to provide a custom `search` * it's expensive to construct an `InMemoryDocstore`/dict * you retrieve documents from remote sources * you just want to reuse existing objects
This **partially** addresses langchain-ai#1524, but it's also useful for some of our use cases. This `DocstoreFn` allows to lookup a document given a function that accepts the `search` string without the need to implement a custom `Docstore`. This could be useful when: * you don't want to implement a `Docstore` just to provide a custom `search` * it's expensive to construct an `InMemoryDocstore`/dict * you retrieve documents from remote sources * you just want to reuse existing objects
Closing, as completed with #6616 |
Documentation for Indexes covers local indexes that use local loaders and indexes.
https://langchain.readthedocs.io/en/latest/modules/indexes/getting_started.html
Is there an appropriate interface for connecting with a remote search index? Is
DocStore
the correct interface to implement?https://github.com/hwchase17/langchain/blob/4f41e20f0970df42d40907ed91f1c6d58a613541/langchain/docstore/base.py#L8
Also, DocStore is missing the top-k search parameter which is needed for it to be usable. Is this something we can add to this interface?
The text was updated successfully, but these errors were encountered: