Johannes Vass

Automatically remove documents that are not present in the source any more

2

I am wondering whether there is any built-in way of automatically removing documents from Danswer that have been deleted in the source (say for example Google Drive). Looking at your...

When enabling the semantic re-ranker I get the following error in the chat mode: ``` File "/app/danswer/chat/process_message.py", line 330, in stream_chat_message reference_db_search_docs = [ ^ File "/app/danswer/chat/process_message.py", line 331, in...

Remove documents from index which are not returned by connector

1

Closes #938 This implements deletion of documents which are not returned from a connector anymore. Known Limitations: - only works for `LoadConnector`, not `PollConnector` - does not batch deletions

Hello, I am looking for a way how to prevent the plugin from overwriting my state from the local storage if some condition matches. I think I found a solution...

Relates to #120 : adds further information to the error message so that the mentioned issue can be more easily debugged (e.g. on CI environments)

multilingual-e5-large exported by recent sentence-transformers version cannot be loaded

2

### System Info Tested TEI versions: - v1.2.0 (official Docker) - v1.2.3 (official Docker) - [cc1c510](https://github.com/huggingface/text-embeddings-inference/commit/cc1c510e8d8af8447c01e6b14c417473cf2dfda9) (current main, built on Ubuntu 23.10, cargo 1.75.0) As it already fails during model...

This is necessary in order to load models whose tokenizers have been created by a version after the breaking change https://github.com/huggingface/tokenizers/pull/1476 (i.e. >= v0.19.0) Fixes #265 ## Before submitting -...

runtime error: invalid memory address or nil pointer dereference in service.go

2

Since a few days I am running servidor on my raspberrypi behind an nginx reverse proxy in a systemd service. For few days it worked like a charm but yesterday...