Commit Graph

22 Commits

Author SHA1 Message Date
Timothy Carambat
dc4ad6b5a9
[BETA] Live document sync (#1719)
* wip bg workers for live document sync

* Add ability to re-embed specific documents across many workspaces via background queue
bgworkser is gated behind expieremental system setting flag that needs to be explictly enabled
UI for watching/unwatching docments that are embedded.
TODO: UI to easily manage all bg tasks and see run results
TODO: UI to enable this feature and background endpoints to manage it

* create frontend views and paths
Move elements to correct experimental scope

* update migration to delete runs on removal of watched document

* Add watch support to YouTube transcripts (#1716)

* Add watch support to YouTube transcripts
refactor how sync is done for supported types

* Watch specific files in Confluence space (#1718)

Add failure-prune check for runs

* create tmp workflow modifications for beta image

* create tmp workflow modifications for beta image

* create tmp workflow modifications for beta image

* dual build
update copy of alert modals

* update job interval

* Add support for live-sync of Github files

* update copy for document sync feature

* hide Experimental features from UI

* update docs links

* [FEAT] Implement new settings menu for experimental features (#1735)

* implement new settings menu for experimental features

* remove unused context save bar

---------

Co-authored-by: timothycarambat <rambat1010@gmail.com>

* dont run job on boot

* unset workflow changes

* Add persistent encryption service
Relay key to collector so persistent encryption can be used
Encrypt any private data in chunkSources used for replay during resync jobs

* update jsDOC

* Linting and organization

* update modal copy for feature

---------

Co-authored-by: Sean Hatfield <seanhatfield5@gmail.com>
2024-06-21 13:38:50 -07:00
Anush
771889ad7f
[FIX] Incorrect vectors count with Qdrant (#1561)
Co-authored-by: Timothy Carambat <rambat1010@gmail.com>
2024-06-06 13:18:01 -07:00
Timothy Carambat
b23cb1a90f
Improve RAG results via chunkHeader append (#1473) 2024-05-21 14:43:39 -05:00
Timothy Carambat
cae6cee1b5
Do not go through LLM to embed when embedding documents (#1428) 2024-05-16 17:51:04 -07:00
Timothy Carambat
9655880cf0
Update all vector dbs to filter duplicate source documents that may be pinned (#1122)
* Update all vector dbs to filter duplicate parents

* cleanup
2024-04-17 18:04:39 -07:00
Timothy Carambat
24b523d5eb
append missing import for some vectordb providers (#1066) 2024-04-07 14:40:23 -07:00
Timothy Carambat
ce98ff4653
Enable customization of chunk length and overlap (#1059)
* Enable customization of chunk length and overlap

* fix onboarding link
show max limit in UI and prevent overlap >= chunk size
2024-04-06 16:38:07 -07:00
Sean Hatfield
56fa17caf2
create configurable topN per workspace (#616)
* create configurable topN per workspace

* Update TopN UI text
Fix fallbacks for all providers
Add SQLite CHECK to TOPN value

* merge with master
Update zilliz provider for variable TopN

---------

Co-authored-by: timothycarambat <rambat1010@gmail.com>
2024-01-18 12:34:20 -08:00
Timothy Carambat
658e7fa390
chore: Better VectorDb and Embedder error messages (#620)
* chore: propogate embedder and vectordb errors during document mutations

* add default value for errors on addDocuments
2024-01-18 11:40:48 -08:00
Timothy Carambat
d0a3f1e3e1
Fix present diminsions on vectorDBs to be inferred for providers who require it (#605) 2024-01-16 13:41:01 -08:00
Shuyoou
6faa0efaa8
Issue #543 support milvus vector db (#579)
* issue #543 support milvus vector db

* migrate Milvus to use MilvusClient instead of ORM
normalize env setup for docs/implementation
feat: embedder model dimension added

* update comments

---------

Co-authored-by: timothycarambat <rambat1010@gmail.com>
2024-01-12 13:23:57 -08:00
Sayan Gupta
b7d2756754
Issue #204 Added a check to ensure that 'chunk.payload' exists and contains the 'id' property (#526)
* Issue #204 Added a check to ensure that 'chunk.payload' exists and contains the 'id' property before attempting to destructure it

* run linter

* simplify condition and comment

---------

Co-authored-by: timothycarambat <rambat1010@gmail.com>
2024-01-04 16:39:43 -08:00
Timothy Carambat
8cc1455b72
feat: add support for variable chunk length (#415)
fix: cleanup code for embedding length clarify
resolves #388
2023-12-07 16:27:36 -08:00
Timothy Carambat
88d4808c52
315 show citations based on relevancy score (#316)
* settings for similarity score threshold and prisma schema updated

* prisma schema migration for adding similarityScore setting

* WIP

* Min score default change

* added similarityThreshold checking for all vectordb providers

* linting

---------

Co-authored-by: shatfield4 <seanhatfield5@gmail.com>
2023-11-06 16:49:29 -08:00
Timothy Carambat
be9d8b0397
Infinite prompt input and compression implementation (#332)
* WIP on continuous prompt window summary

* wip

* Move chat out of VDB
simplify chat interface
normalize LLM model interface
have compression abstraction
Cleanup compressor
TODO: Anthropic stuff

* Implement compression for Anythropic
Fix lancedb sources

* cleanup vectorDBs and check that lance, chroma, and pinecone are returning valid metadata sources

* Resolve Weaviate citation sources not working with schema

* comment cleanup
2023-11-06 13:13:53 -08:00
Timothy Carambat
5d56ab623b
Anthropic claude 2 support (#305)
* WIP Anythropic support for chat, chat and query w/context

* Add onboarding support for Anthropic

* cleanup

* fix Anthropic answer parsing
move embedding selector to general util
2023-10-30 15:44:03 -07:00
Sean Hatfield
669d7a396d
282 return relevancy score with similarityresponse (#304)
* include score value in similarityResponse for weaviate

* include score value in si
milarityResponse for qdrant

* include score value in si
milarityResponse for pinecone

* include score value in similarityResponse for chroma

* include score value in similarityResponse for lancedb

* distance to similarity

---------

Co-authored-by: timothycarambat <rambat1010@gmail.com>
2023-10-30 12:46:38 -07:00
Timothy Carambat
a8ec0d9584
Compensate for upper OpenAI emedding limit chunk size (#292)
Limit is due to POST body max size. Sufficiently large requests will abort automatically
We should report that error back on the frontend during embedding
Update vectordb providers to return on failed
2023-10-26 10:57:37 -07:00
Sean Hatfield
a126b5f5aa
Replace custom sqlite dbms with prisma (#239)
* WIP converted all sqlite models into prisma calls

* modify db setup and fix ApiKey model calls in admin.js

* renaming function params to be consistent

* converted adminEndpoints to utilize prisma orm

* converted chatEndpoints to utilize prisma orm

* converted inviteEndpoints to utilize prisma orm

* converted systemEndpoints to utilize prisma orm

* converted workspaceEndpoints to utilize prisma orm

* converting sql queries to prisma calls

* fixed default param bug for orderBy and limit

* fixed typo for workspace chats

* fixed order of deletion to account for sql relations

* fix invite CRUD and workspace management CRUD

* fixed CRUD for api keys

* created prisma setup scripts/docs for understanding how to use prisma

* prisma dependency change

* removing unneeded console.logs

* removing unneeded sql escape function

* linting and creating migration script

* migration from depreciated sqlite script update

* removing unneeded migrations in prisma folder

* create backup of old sqlite db and use transactions to ensure all operations complete successfully

* adding migrations to gitignore

* updated PRISMA.md docs for info on how to use sqlite migration script

* comment changes

* adding back migrations folder to repo

* Reviewing SQL and prisma integraiton on fresh repo

* update inline key replacement

* ensure migration script executes and maps foreign_keys regardless of db ordering

* run migration endpoint

* support new prisma backend

* bump version

* change migration call

---------

Co-authored-by: timothycarambat <rambat1010@gmail.com>
2023-09-28 14:00:03 -07:00
Sean Hatfield
ce6951b21f
Renamed all indicies to vectors to avoid confusion of vocab (#246)
* renamed all indicies to vectors to avoid confusion of vocab

* removing unneeded files

* changed indicies on frontend sidebar to vectors
2023-09-21 12:04:17 -07:00
timothycarambat
4f8abeb7fc better loggin on addDocumentToWorkspace and add Qdrant setup doc 2023-08-22 10:30:01 -07:00
Timothy Carambat
cf0b24af02
Add Qdrant support for embedding, chat, and conversation (#192)
* Add Qdrant support for embedding, chat, and conversation

* Change comments
2023-08-15 15:26:44 -07:00