Yoann Poupart

For the record, I began to refactor in `dynamicloader+`.

You can use the `from_embeddings` method. ```python texts = [d.page_content for d in docs] doc_embeddings = embeddings.embed_documents(docs) text_embedding_pairs = zip(texts, doc_embedding_pairs) vectorstore = FAISS.from_embeddings(text_embedding_pairs, embeddings) ```

The langchain logic is not there yet. To make it right, I think this would require quite some effort. Yet, as trancethehuman said, you can work this out directly with...

You are right my code fails to update docstore ids. It shouldn't be that hard to do btw. But I think yours won't work since Faiss translates the indices when...

@Pixcoder Beware of Faiss overwriting indices. Removing the biggest indices first seems superfluous since you can remove multiple indices simultaneously. Re-indexing `index_to_docstore_id` is always necessary. For the record, here is...

@Pixcoder This should never happen, as the only purpose of the docstore is to store documents indexed by the vectorstore index. I think you could raise an error if you...

Both errors come from your environment and not the code in itself. The first is related to your `playwright` configuration. The second is related to the iPython version since inline...

Not necessarily, in notebooks (recent enough) inline await is possible.

If by local you mean directly on your machine, the is `SelfHostedHuggingFaceLLM`. If you mean on a local server, maybe this PR #2830 solves your problem. (Feel free to review,...

Unsure if this would be that great. Up to you to manage your own ids in the metadata.