pablodanswer
7bfa99766d
Add support for google slides ( #2083 )
...
* add support for google slides
* remove log + account for dead code
* squash
2024-08-09 17:12:51 +00:00
hagen-danswer
b230082891
Openai encoding temp hotfix ( #2094 )
2024-08-09 08:17:31 -07:00
Yuhong Sun
8cd1eda8b1
Rework Rerankers ( #2093 )
2024-08-08 21:33:49 -07:00
Yuhong Sun
7dcc42aa95
Intent Model Update ( #2069 )
2024-08-08 20:45:53 -07:00
pablodanswer
384e61f4b0
add new gpt-4o model
2024-08-08 16:32:57 -07:00
pablodanswer
c1e3a1b3e7
Select proper assistant override ( #2068 )
...
* encode images properly
* proper assistant default model updates
* remove now unneeded image encoding update
* update naming of persona llm option gathering
2024-08-08 21:02:11 +00:00
rkuo-danswer
be9ed319d5
add unit test for quotes ( #2085 )
...
* add unit test for quotes
* test answer and quotes together
2024-08-08 18:20:07 +00:00
Richard Kuo (Danswer)
bdaaebe955
use re.search instead of re.match (which searches from start of string only)
2024-08-07 20:55:18 -07:00
pablodanswer
9eb48ca2c3
account for empty links + fix quote processing
2024-08-07 20:55:18 -07:00
rkuo-danswer
509fa3a994
add postgres configuration ( #2076 )
2024-08-08 00:13:59 +00:00
pablodanswer
5097c7f284
Handle saved search docs in eval flow ( #2075 )
2024-08-07 16:18:34 -07:00
pablodanswer
c4e1c62c00
Admin UX updates ( #2057 )
2024-08-07 14:55:16 -07:00
pablodanswer
44d8e34b5a
Improve seeding (includes all enterprise features) ( #2065 )
2024-08-07 10:44:33 -07:00
pablodanswer
d2e16a599d
Improve shared chat page ( #2066 )
...
* improve look of shared chat page
* remove log
* cleaner display
* add initializing loader to shared chat page
* updated danswer loaders (for prism)
* remove default share
2024-08-07 16:13:55 +00:00
pablodanswer
291e6c4198
somewhat clearer API errors ( #2064 )
2024-08-07 03:04:26 +00:00
Chris Weaver
bb7e1d6e55
Add integration tests for document set syncing ( #1904 )
2024-08-06 18:00:19 -07:00
rkuo-danswer
fcc4c30ead
don't skip the start of the json answer value ( #2067 )
2024-08-06 23:59:13 +00:00
pablodanswer
e0f0cfd92e
Ensure relevance functions for selected docs ( #2063 )
...
* ensure relevance functions for selected docs
* remove logs
* remove log
2024-08-06 21:06:44 +00:00
pablodanswer
ab564a9ec8
Add cleaner loading / streaming for image loading ( #2055 )
...
* add image loading
* clean
* add loading skeleton
* clean up
* clearer comments
2024-08-06 19:28:48 +00:00
rkuo-danswer
cc3856ef6d
enforce index attempt deduping on secondary indexing. ( #2054 )
...
* enforce index attempt deduping on secondary indexing.
* black fix
* typo fixes
---------
Co-authored-by: Richard Kuo <rkuo@rkuo.com>
2024-08-06 17:45:16 +00:00
Yuhong Sun
a8a4ad9546
Chunk Filter Metadata Format ( #2053 )
2024-08-05 15:12:36 -07:00
pablodanswer
5825d01d53
Better assistant interactions + UI ( #2029 )
...
* add assistnat re-ordering, selections, etc.
* squash
* remove unnecessary comment
* squash
* adapt dragging for all IDs + smoother animation + consistency
* fix minor typing issue
* fix minor typing issue
* remove logs
2024-08-05 18:22:57 +00:00
pablodanswer
a3ea217f40
ensure consistency of answers + update llm relevance prompting ( #2045 )
2024-08-05 08:27:15 -07:00
pablodanswer
6d67d472cd
Add answers to search ( #2020 )
2024-08-04 23:02:55 -07:00
Weves
76b7792e69
Harden embedding calls
2024-08-04 15:11:45 -07:00
Chris Weaver
9d7100a287
Fix secondary index attempts showing up as the primary index status + scheduling while in-progress ( #2039 )
2024-08-04 13:29:44 -07:00
pablodanswer
0261d689dc
Various Admin Page + User Flow Improvements ( #1987 )
2024-08-03 18:09:46 -07:00
hagen-danswer
57cc53ab94
Added content tags to zendesk connector ( #2017 )
2024-08-02 10:09:53 -07:00
rkuo-danswer
6a61331cba
Feature/log despam ( #2022 )
...
* move a lot of log spam to debug level. Consolidate some info level logging
* reformat more indexing logging
2024-08-02 15:28:53 +00:00
Weves
51731ad0dd
Fix issue where large docs/batches break openai embedding
2024-08-02 01:07:09 -07:00
rkuo-danswer
f280586e68
pass function to Process correctly instead of running it inline ( #2018 )
...
* pass function to Process correctly instead of running it inline
* mypy fixes and pass back return result (even tho we don't use it right now)
2024-08-02 00:06:35 +00:00
hagen-danswer
e6a92aa936
support confluence single page only indexing ( #2008 )
...
* added index recursively checkbox
* mypy fixes
* added migration to not break existing connectors
2024-08-01 20:32:46 +00:00
pablodanswer
a54ea9f9fa
Fix cartesian issue with index attempts ( #2015 )
2024-08-01 10:25:25 -07:00
Yuhong Sun
73a92c046d
Fix chunker ( #2014 )
2024-08-01 10:18:02 -07:00
pablodanswer
459bd46846
Add Prompt library ( #1990 )
2024-08-01 08:40:35 -07:00
Chris Weaver
445f7e70ba
Fix image generation ( #2009 )
2024-08-01 00:27:02 -07:00
Yuhong Sun
ca893f9918
Rerank Handle Null ( #2010 )
2024-07-31 22:59:02 -07:00
hagen-danswer
1be1959d80
Changed default local model to nomic ( #1943 )
2024-07-31 18:54:02 -07:00
rkuo-danswer
7c283b090d
Feature/postgres connection names ( #1998 )
...
* avoid reindexing secondary indexes after they succeed
* use postgres application names to facilitate connection debugging
* centralize all postgres application_name constants in the constants file
* missed a couple of files
* mypy fixes
* update dev background script
2024-07-31 20:36:30 +00:00
rkuo-danswer
288e6fa606
Bugfix/pg connections ( #2002 )
...
* increase max_connections to 150 in all docker files
* lower celery worker concurrency to 6
2024-07-31 19:49:20 +00:00
hagen-danswer
5307d38472
Fixed tokenizer logic ( #1986 )
2024-07-31 09:59:45 -07:00
Yuhong Sun
d619602a6f
Skip shortcut docs ( #1999 )
2024-07-31 09:51:01 -07:00
Yuhong Sun
348a2176f0
Fix Dropped Documents ( #1997 )
2024-07-31 09:33:36 -07:00
pablodanswer
89b6da36a6
process files with null title ( #1989 )
2024-07-31 08:18:50 -07:00
Yuhong Sun
036d5c737e
No Null Embeddings ( #1982 )
2024-07-30 19:54:49 -07:00
hagen-danswer
d151082871
Moved warmup_encoders into scope ( #1978 )
2024-07-30 16:37:32 +00:00
pablodanswer
e4b1f5b963
fix index attempt migration where no credential ID
2024-07-30 08:57:57 -07:00
hagen-danswer
3938a053aa
Rework tokenizer ( #1957 )
2024-07-29 23:01:49 -07:00
Chris Weaver
fb6695a983
Fix flow where oidc_expiry is different from token expiry ( #1974 )
2024-07-30 03:17:08 +00:00
rkuo-danswer
015f415b71
avoid reindexing secondary indexes after they succeed ( #1971 )
2024-07-30 03:12:58 +00:00
