Commit Graph
108 Commits
Author SHA1 Message Date
onlyjackfrost 16c175f972 add langfuse related env for local debugging 2024-11-18 16:08:57 +08:00
Pao Sheng 75910d2923 chore(wren-ai-service): separate the api key for each LLM and embedder (#917) 2024-11-15 20:45:28 +08:00
Chih-Yu Yeh d6e3bd3d34 chore(wren-ai-service): remove ollama in config (#914) 2024-11-15 13:17:00 +08:00
0f1d8d857c feat(wren-ai-service): multi-llm deployment and setting loading mechanism (#792)
Co-authored-by: ChihYu Yeh <chihyu.jimmy.yeh@gmail.com>
Co-authored-by: Ted Yan <tedyyan@hotmail.com>
Co-authored-by: Teddy Yan <xuan.yan@nokia.com>
2024-11-15 11:11:22 +08:00
william chang(張仲威) dff425a87b release 0.10.0 (#881) 2024-11-07 17:21:20 +08:00
william chang(張仲威) 0772c081eb update AI version (#878) 2024-11-07 14:26:48 +08:00
Chih-Yu Yeh 359e964ae4 chore(wren-ai-service): optimize retrieval process (#871) 2024-11-06 15:55:08 +08:00
Andy Yen 1a42964274 Release 0.10.0-rc.1 (#865) 2024-11-05 17:19:43 +08:00
Freda Lai c98e7deef1 chore(wren-ui): remove connectSampleNBA e2e test and update readme file (#840) 2024-10-30 13:36:21 +08:00
Andy Yen a80efa4a61 feature(wren-ui): integrate with rust version Wren Engine (#770)
* add env config to accept EXPERIMENTAL_ENGINE_RUST_VERSION

* make mdl changes and ibis changes to integrate with rust version wren engine

* support add experimental flag in wren launcher

* update wren-ui readme

* add logger for debugging

* use boolean instead of 1/0 in MDL

* remove null value key in MDL and remove express if it's empty string

* should throw error after sending events

* fix typescript error

* update ui image for testing

* update platform

* add platform flag

* revert version for testing
2024-10-28 11:29:52 +08:00
Shimin 95518aa78e release 0.9.1 (#824) 2024-10-24 22:21:43 +08:00
Shimin 8d0cb958b0 release 0.9.1-rc.1 (#820) 2024-10-24 17:19:55 +08:00
Andy Yen a41829c215 release 0.9.0 (#803) 2024-10-23 11:09:58 +08:00
Piotr IdzikandPao Sheng c9706b52c4 docs: update dead links in docker/README.md (#769)
* docs: update dead links in `docker/README.md`

* docs: update the description of wren-ai-service

Co-authored-by: Piotr Idzik <65706193+vil02@users.noreply.github.com>

---------

Co-authored-by: Pao Sheng <paooap.oappao@gmail.com>
2024-10-22 23:25:48 +08:00
Andy Yen 7a1c502f96 release rc (#794) 2024-10-22 17:29:39 +08:00
Andy Yen e88ca57d8c release 0.9.0-rc.5 (#772) 2024-10-18 16:08:02 +08:00
Andy Yen 34c454ad1d release 0.9.0-rc.4 (#764) 2024-10-16 15:59:45 +08:00
Andy Yen 3f16b53f2f release 0.9.0-rc.3 (#758) 2024-10-15 09:50:22 +08:00
Andy Yen f2ea08525d release 0.9.0-rc.2 (#744) 2024-10-09 17:55:45 +09:00
Andy Yen 6f445634d8 release 0.9.0-rc.1 (#738) 2024-10-08 23:07:36 +08:00
william chang(張仲威) d5514b435d release 0.8.1 (#714) 2024-10-03 15:47:43 +08:00
william chang(張仲威) 8b439a8f19 release 0.8.1-rc.1 2024-10-03 12:04:15 +08:00
Chih-Yu Yeh ac65f31d9d chore(wren-ai-service): minor updates (#673)
* fix conflict

* update

* add notebook for analysis

* change title

* update

* add Qdrant timeout

* clear output

* update
2024-09-20 11:53:20 +08:00
Chih-Yu Yeh c428bb0f5e feat(wren-ai-service): sql-expansion (#667) 2024-09-19 12:19:17 +08:00
Chih-Yu Yeh b5e269ec38 chore(wren-ai-service): add LLM_TIMEOUT, EMBEDDER_TIMEOUT (#660) 2024-09-11 17:13:27 +08:00
Pao Sheng 11dd3ebd93 feat(wren-ai-service): Support TTL for query cache (#661)
* chore: add cachetools into dependencies

* feat: using TTL LRU cache to store task results

* chore: expose environment variables to control maxsize and ttl for query cache

* fix: unsupported operand type(s) for ** or pow(): 'dict' and 'dict'

* chore: add mock pipe for testing and test case for query cache ttl

* chore: set a large number for maxsize and dont expose it to control via env var
2024-09-11 15:17:43 +08:00
Andy Yen dda3417e19 release 0.8.0 (#651) 2024-09-05 17:46:36 +08:00
Andy Yen 66a471b490 release 0.8.0-rc.3 (#649) 2024-09-05 16:59:18 +08:00
Andy Yen be7bb837d1 release 0.8.0-rc.2 (#647) 2024-09-05 14:04:32 +08:00
Chih-Yu Yeh 322c34aa43 chore(wren-ai-service): minor refactor (#643)
* fix conflict

* update

* add sql summary

* fix bug

* fix conflict

* remove pandasai

* update

* fix bug

* refactor pipeline

* refactor

* update

* refactor

* fix langfuse metadata

* update

* add pandas
2024-09-04 16:35:06 +08:00
Andy Yen 2e25e69a15 release 0.8.0-rc.1 (#633) 2024-08-30 16:15:25 +08:00
Chih-Yu Yeh 4e0acc908f chore(wren-ai-service): retrieval improvement (#599)
* refactor

* add table description to indexing

* remove print

* rename

* fix conflict

* refine indexing to have columns per batch

* update

* fix

* fix but

* fix bug

* fix

* fix

* update

* update

* fix bug

* update top k

* fix eval

* fix bug

* update

* update

* update

* skip if 0 documents

* fix top k

* delete documents by filters instead of ids

* fix bug

* update

* update

* update

* fix test

* add COLUMN_INDEXING_BATCH_SIZE

* update

* fix bug

* update

* fix bug
2024-08-30 14:35:26 +08:00
Chih-Yu Yeh 7a04977fb8 chore(wren-ai-service): update deps (#615)
* update deps

* upgrade qdrant version to 1.11.0
2024-08-20 17:53:50 +08:00
Chih-Yu Yeh aa8ff445c3 remove wren-ui (#608) 2024-08-19 10:20:16 +08:00
Chih-Yu Yeh f493519459 disable SHOULD_FORCE_DEPLOY by default and enabled only using launcher (#600) 2024-08-14 16:34:15 +08:00
Chih-Yu Yeh e5f552b4a4 chore(wren-ai-service): minor update (#592)
* adjust

* update

* add env

* fix bug

* add qdrant_host

* update kustomizations configs

* change AI_SERVICE_ENV to SHOULD_FORCE_DEPLOY

* remove

* update

* update.gitignore

* remove file

* adjust entrypoint
2024-08-13 14:27:44 +08:00
Chih-Yu Yeh af556b0261 feat(wren-ai-service): optimize qdrant for multi users setting (#558)
* add binary quantization

* allow multi users setting

* user_id backward compatible

* fix test

* fix conflict

* add user_id to indexing

* remove print

* rename user_id to id

* upgrade qdrant to v1.10.1

* update

* allow using qdrant cloud

* update

* enable binary quantization only if embedding model dim >= 1024

* fix tests

* fix test

* revert

* fix mkdir

* simple refactor

* fix
2024-08-05 13:43:01 +08:00
Andy Yen d14a17d025 release 0.7.5 (#573) 2024-08-01 15:47:38 +08:00
Andy Yen d8a16033dd release 0.7.5-rc.2 (#570) 2024-07-31 16:50:46 +08:00
Chih-Yu Yeh 679e610c2f chore(wren-ai-service): fix entrypoint (#568)
* fix entrypoint

* add timeout

* fix

* update

* fix
2024-07-31 16:40:35 +08:00
Andy Yen 04955bb488 release 0.7.5-rc.1 (#564) 2024-07-30 16:04:26 +08:00
Freda Lai 6de58371bb chore(wren-ui): remove unused code (#555)
* chore(wren-ui): remove dead CSS style

* chore(wren-ui): remove unused component

* chore(wren-ui): remove connection info GraphQL API

* chore(wren-ui): update bootstrap and launcher
2024-07-29 22:04:01 +08:00
Andy Yen d023b75735 release 0.8.0-rc.1 (#545) 2024-07-19 18:03:56 +08:00
Andy Yen 970232dff7 release 0.7.4 (#542) 2024-07-19 14:54:16 +08:00
Chih-Yu Yeh 0fa6f625a0 feat(wren-ai-service): add gpt-4o-mini (#541)
* add gpt-4o-mini

* fix docker setup
2024-07-19 10:20:37 +08:00
Andy Yen 021c0873c5 release 0.7.3 (#529) 2024-07-17 17:14:57 +08:00
Andy Yen 6e52c2a6df Release 0.7.2 (#524)
* release 0.7.2

* update env.example
2024-07-16 16:03:36 +08:00
Andy Yen c23ae65f20 release 0.7.1 (#502) 2024-07-10 16:11:50 +08:00
Andy Yen 5b314b5a53 release 0.7.1-rc1 (#499) 2024-07-10 12:22:31 +08:00
Andy Yen b8b866b2ab release 0.7.0 (#484) 2024-07-05 18:07:02 +08:00