lexluo09
954c67c947
[improvement][Headless] Add unit test cases and remove unnecessary logs. ( #698 )
2024-01-30 16:46:06 +08:00
mainmain
c398ac1a84
[improvement][supersonic] add text-to-sql evaluation ( #696 )
...
* [improvement] llm supports all models
* [improvement] alias convert to SemanticParseInfo
* [improvement] support join
* [improvement] add evaluation.py
* [improvement] add text2sql_evalution.py
* [improvement] add text2sql_evalution.py
* [improvement] add evalution
* [improvement] add evalution
* [improvement] add evalution
---------
Co-authored-by: zuopengge <hwzuopengge@tencent.com >
2024-01-30 10:46:45 +08:00
lexluo09
841db25198
[improvement][Headless] Fix the caching issue in struct queries and add headless test cases. ( #689 )
2024-01-25 11:31:15 +08:00
lexluo09
922201c181
[improvement][Headless] Simplify the QueryService interface, optimize Query permissions, and add integration testing. ( #687 )
2024-01-24 17:33:12 +08:00
Scott
48fb01f6bc
(chat)(fix):not add group by if already has distinct ( #685 )
2024-01-24 09:48:21 +08:00
lexluo09
9d6f96e6d4
[improvement][project] Keep the style consistent when displaying various SQL queries on the Chat page. ( #684 )
2024-01-23 15:10:48 +08:00
LXW
42a6f61456
(improvement)(chat) Provide agent tool type list ( #683 )
...
Co-authored-by: jolunoluo
2024-01-23 11:45:15 +08:00
lexluo09
163e782f51
[improvement][headless] Merge struct and SQL query code, unify the queries. ( #680 )
2024-01-22 22:54:07 +08:00
LXW
b28eb637c8
(improvement)(chat) Put queryText to PluginParseResult ( #673 )
...
Co-authored-by: jolunoluo
2024-01-21 16:26:49 +08:00
lexluo09
9c6bd7cf19
(improvement)(Headless) Support optimizing and generating SQL based on the engine type. ( #666 )
2024-01-19 22:32:28 +08:00
lexluo09
36edc0c1b4
(improvement)(test) Fix Python exemplar file name error and fix chat unit tests. ( #661 )
2024-01-19 18:12:01 +08:00
lexluo09
cbf38ed785
(improvement)(chat) Remove QueryReq parameter from QueryContext. ( #656 )
2024-01-19 16:17:31 +08:00
lexluo09
3e77fc3069
(improvement)(headless) Headless has preliminarily completed the abstraction of QueryCache, QueryParser, QueryPlanner, and QueryExecutor. ( #651 )
2024-01-18 22:39:58 +08:00
Scott
90f9da162e
(fix)(chat):skip null values to avoid NPE ( #645 )
2024-01-18 19:56:03 +08:00
lexluo09
a06a1fa898
(improvement)(headless) Remove ItemDateResp and databaseResp from SemanticModel. ( #648 )
2024-01-18 16:30:18 +08:00
Scott
dcc1f26542
(fix)(chat):queryByStruct use default metrics period ( #621 )
2024-01-12 17:00:53 +08:00
mainmain
1e93282c9f
[improvement] llm supports all models ( #615 )
2024-01-10 22:01:21 +08:00
lexluo09
bbad302efd
(improvement)(chat) Add the pluginList parameter to QueryContext. ( #609 )
2024-01-09 15:23:07 +08:00
lexluo09
ad8079e058
(improvement)(chat) The system settings module supports setting the number of returned parses. ( #606 )
2024-01-08 19:35:28 +08:00
lexluo09
8eef11f342
(improvement)(chat) Fix the thread safety issue in parallel insertion of nature. ( #603 )
2024-01-08 17:08:38 +08:00
lexluo09
b55b4c130e
(improvement)(chat) Add queryFilters setting to queryContext. ( #601 )
2024-01-05 12:04:55 +08:00
lexluo09
602b9547b8
(improvement)(chat) Make it compatible with the case when the agent is empty. ( #596 )
2024-01-04 21:39:56 +08:00
lexluo09
ade96c3adc
(improvement)(chat) Clean up unused code and optimize some code and make the tests pass ( #595 )
2024-01-04 18:22:40 +08:00
lexluo09
023e84c420
(improvement)(chat) Split chat into three modules: server, api, and core. ( #594 )
2024-01-04 16:56:49 +08:00
LXW
0858c13365
(improvement)(headless) Update server and core, server calls core one-way ( #592 )
...
* (improvement)(headless) Update server and core, server calls core one-way
* (improvement)(Auth) When obtaining the user information, determine whether the user is a system admin.
---------
Co-authored-by: jolunoluo
2024-01-04 12:08:12 +08:00
LXW
52fea5311d
(improvement)(headless) Merge server and core in headless-api ( #590 )
...
(improvement)(headless) Merge server and core in headless-api
---------
Co-authored-by: jolunoluo
2024-01-02 20:08:23 +08:00
lexluo09
d72166944c
(improvement)(chat) Add 'Few-shot Examples' display to the Chat chart. ( #589 )
2024-01-02 18:07:15 +08:00
LXW
e7f13572d7
(improvement)(headless) Modify the headless module structure to api, core and server ( #588 )
...
Co-authored-by: jolunoluo
2024-01-02 16:43:28 +08:00
codescracker
af1c560cc4
(fix)(chat):fix typo in s2sql and add prompt to output. ( #581 )
...
* 1.refactor the retrieval module. 2.refactor the http service module. 3.upgrade text2sql output format the parse for absolute time related expression in query.
* fix bug.
* upgrade the config module, now support config llm suppoted by langchain.
* fix conflicts.
* update text2sql config reload to be compitable with new config format.
* modify default config.
* 1.add self-consistency feature for text2sql. 2.upgrade llm api call from sync to async. 3.refactor text2sql module. 4. refactor semantical retriever modules.
* merege with upstream master
* add general retrieve service.
* add api service for sql_agent for crud opereations of few-shots examples.
* modify requirements
* add auto-cot feature
* 1. output log to a fixed log file. 2.allow few-shots examples tied to data model, and add strategy that extend examples when retrieved examples tied to a data model is not enough. 3. fix misformat in s2ql args.
* add prior_ext to output.
* fix type in in s2sql
* add prompt to output.
---------
Co-authored-by: shaoweigong <shaoweigong@tencent.com >
2024-01-02 16:35:33 +08:00
LXW
af6c8cdbda
(feature)(headless) Add a new headless-api module and Change old headless-api module to headless-common ( #582 )
...
* (improvement)(headless) Add headless-api module
* (improvement)(headless) Change old headless-api module to headless-common
---------
Co-authored-by: jolunoluo <jolunoluo@tencent.com >
2023-12-28 22:14:57 +08:00
lexluo09
6c8ebdfe1a
(improvement)(chat) Modify the default configuration related to text2sql numbers. ( #575 )
2023-12-27 20:39:08 +08:00
codescracker
b706c4efb4
Add feature to s2sql that allow few-shots example tied to data model. ( #571 )
...
* 1.refactor the retrieval module. 2.refactor the http service module. 3.upgrade text2sql output format the parse for absolute time related expression in query.
* fix bug.
* upgrade the config module, now support config llm suppoted by langchain.
* fix conflicts.
* update text2sql config reload to be compitable with new config format.
* modify default config.
* 1.add self-consistency feature for text2sql. 2.upgrade llm api call from sync to async. 3.refactor text2sql module. 4. refactor semantical retriever modules.
* merege with upstream master
* add general retrieve service.
* add api service for sql_agent for crud opereations of few-shots examples.
* modify requirements
* add auto-cot feature
* 1. output log to a fixed log file. 2.allow few-shots examples tied to data model, and add strategy that extend examples when retrieved examples tied to a data model is not enough. 3. fix misformat in s2ql args.
* add prior_ext to output.
---------
Co-authored-by: shaoweigong <shaoweigong@tencent.com >
2023-12-27 19:39:50 +08:00
LXW
40c86810bb
(improvement)(chat) Modify query type rules in QueryTypeParser ( #570 )
...
Co-authored-by: jolunoluo
2023-12-25 20:54:04 +08:00
LXW
e0955c0618
(improvement)(chat) Replace the table name in correct sql with a string concatenated by the model name in FromCorrector ( #566 )
...
Co-authored-by: jolunoluo
2023-12-22 19:24:57 +08:00
lexluo09
0a1f446fb8
(improvement)(chat) After a failed query, the absence of a returned queryId causes the recommendation question call to fail. ( #562 )
2023-12-21 22:10:48 +08:00
lexluo09
7b580b7c94
(improvement)(chat) Introduce a separate log file to record key outputs in the pipeline. ( #559 )
2023-12-20 22:45:20 +08:00
LXW
2631352c30
(improvement)(common) Add JsonUtil configuration to avoid deserialization failure when enumeration values change ( #558 )
...
Co-authored-by: jolunoluo
2023-12-20 22:11:36 +08:00
Scott
f7914ff6f4
(fix):fill values into param options in webservice query, some refactor ( #557 )
2023-12-20 19:56:06 +08:00
lexluo09
150d67f903
(improvement)(semantic) Change 'semantic' to 'headless' ( #555 )
2023-12-20 17:08:16 +08:00
lexluo09
1d9b6d6877
[improvement](chat) optimize functionCall result format ( #549 )
2023-12-19 15:46:17 +08:00
LXW
d8930e8906
(improvement)(chat) Optimize the conditions for chatDemo loading ( #548 )
...
Co-authored-by: jolunoluo
2023-12-19 14:57:30 +08:00
LXW
7cb7697353
(improvement)(chat) Modify the order of demo loaders ( #542 )
...
Co-authored-by: jolunoluo
2023-12-18 23:15:23 +08:00
LXW
e7d52f87f0
(improvement)(python) run.py read files in utf-8 format ( #540 )
...
Co-authored-by: jolunoluo
2023-12-18 22:07:27 +08:00
lexluo09
e08435902a
[improvement](chat) fix SqlGeneration format issues ( #538 )
2023-12-18 22:04:40 +08:00
LXW
d7f1f06daf
(improvement)(chat) Remove candidateParses in ParseResp ( #536 )
...
Co-authored-by: jolunoluo
2023-12-18 21:10:53 +08:00
LXW
d7fafa361d
(improvement)(chat) Switching metric supports default aggregation method of metric ( #534 )
...
Co-authored-by: jolunoluo
2023-12-18 18:51:21 +08:00
Scott
0c69651ef3
(fix):remove parantheses in generated sql ( #532 )
2023-12-18 17:57:06 +08:00
LXW
61641ecb00
(improvement)(chat) The history record only retains the query result of the first parse ( #526 )
...
Co-authored-by: jolunoluo
2023-12-18 10:20:50 +08:00
lexluo09
fe75b3e393
[improvement](chat) Add persistence to InMemoryEmbeddingStore, fix the issue of PythonServiceS2EmbeddingStore being empty. ( #524 )
2023-12-17 11:04:29 +08:00
lexluo09
59c21ea19a
[improvement](chat) support filterCondition in InMemoryS2EmbeddingStore ( #523 )
2023-12-16 20:04:01 +08:00
