Introduce incremental updates for embeddings in document stores #812
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
tanaysoni
force-pushed
the
refactor-update-embedding
branch
from
3f1793b
to
69d8986
Compare
tanaysoni
force-pushed
the
refactor-update-embedding
branch
from
95d7cd2
to
ec9687f
Compare
tanaysoni
force-pushed
the
refactor-update-embedding
branch
from
e9f5ef4
to
64d4d9b
Compare
tanaysoni
force-pushed
the
refactor-update-embedding
branch
from
ff80592
to
3c6879a
Compare
tanaysoni
changed the title
tanaysoni
changed the title
lalitpagaria
reviewed
Feb 8, 2021
lalitpagaria
reviewed
Feb 8, 2021
tholor
reviewed
Feb 9, 2021
tanaysoni
force-pushed
the
refactor-update-embedding
branch
from
3d73f4d
to
5ab1da2
Compare
tanaysoni
force-pushed
the
refactor-update-embedding
branch
from
a4128ea
to
3beffb2
Compare
tanaysoni
force-pushed
the
refactor-update-embedding
branch
from
18addba
to
6060fcd
Compare
tanaysoni
changed the title
Merged
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Outdated suggestions cannot be applied.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.
The document stores provide a
update_embedding()method that takes a retriever instance as an argument to create/update embeddings of all indexed documents. This works well for the creation of new document stores. However, this approach is inefficient when a large document collection is already present and you want to update only the newly added documents that don't have embeddings yet.This PR introduces a new parameter
update_existing_embeddingsforupdate_embeddings(), which when set toFalse, will only generate embeddings for documents without embeddings. Additionally,filtersare now also added to update embeddings for a subset of documents.Resolves #806