liuyu/qlib
1
0
Fork 0
mirror of https://github.com/microsoft/qlib.git synced 2026-07-01 18:11:18 +08:00
Code Activity
1,960 Commits 46 Branches 26 Tags
4fccf8112d890ca7fd0344425bfc666f776fdb9d
Commit Graph

1960 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Xu Yang
4fccf8112d fix one workflow 2023-06-30 14:33:41 +08:00
Xu Yang
73bd79ca1a merge into one commit 2023-06-30 14:23:40 +08:00
Fivele-Li
7e84f3aae2 Add backtest and backforward task (#1568) 
* * add TrainTask & BacktestTask;
* add BackForwardTask;
* adjust prompt_template.yaml which default config failed to backtest;
* run workflow in loop
* add update method to prompt_template.py

* remove debug code

* Adjust Learn Process
* add LearnManager class & use LearnManager to update system prompt;
* use qrun to replace recorder for training and backtesting;

* Adjust analyser
* analyser independent of recorder;
* rename analyser's workspace attribution;
* analyser load variable by recorder.

---------

Co-authored-by: Cadenza-Li <362237642@qq.com>
 2023-06-30 10:04:43 +08:00
Fivele-Li
1326ac614d Add docs to context and retrieve (#1566) 
* add analyser docstring to context;
* add retrieve method to context manager;

* add notes to retrieve
 2023-06-24 21:47:27 +08:00
Fivele-Li
f12184cc0f Add analyser task and optimize interact (#1552) 
* * optimize interact
* add AnalyserTask
* optimize logger format and add render feature

* format optimize
 2023-06-16 11:42:45 +08:00
Xu Yang
a70386ad52 Merge pull request #1550 from microsoft/xuyang1/refine_task_prompts 
add datahandler and design action task according to component
 2023-06-14 14:52:42 +08:00
Xu Yang
74619ed8d8 fix using defaut in record strategy and backtest 2023-06-14 14:52:16 +08:00
Fivele-Li
1a523df007 Optimize log and interact of FinCo (#1549) 
* use FinCoLog for a better interact experience

* addition file changes

* optimize format

* optimize format
 2023-06-14 14:48:17 +08:00
Xu Yang
f9cc8a5aaa remove useless prompt 2023-06-14 10:46:38 +08:00
Xu Yang
7762c5a1fd add datahandler and design action task according to component 2023-06-13 23:28:27 +08:00
Xu Yang
fa7ef29281 Merge pull request #1548 from microsoft/xuyang1/add_dump_to_file_task 
add simple readme & move prompt templates to outer yaml file to make the code clean
 2023-06-13 15:29:13 +08:00
Xu Yang
429c9a7c66 format 2023-06-13 15:27:59 +08:00
Xu Yang
80fbc00792 move prompt templates to yaml file to make code clean 2023-06-13 15:21:19 +08:00
Xu Yang
01accec24c update code 2023-06-12 16:25:16 +08:00
Fivele-Li
1d88830b0d Add recorder task and visualize (#1542) 
* add recorder task

* add batch generate summarize report unittest.

* * add recorder to RecorderTask;
* add matplot figure to analyzer.py

* add image to markdown;

* Add some log

* update figure path.

---------

Co-authored-by: Young <afe.young@gmail.com>
Co-authored-by: Cadenza-Li <362237642@qq.com>
 2023-06-12 15:48:00 +08:00
you-n-g
ad7498e287 Edit yaml task (#1538) 
* Edit yaml task

* update comments
 2023-06-02 00:44:41 +08:00
you-n-g
73d51f05b4 Init workspace and CMDTask (#1537) 
* Update setup.py and config

* WIP

* init_workspace and CMDTask

* Delete test_sumarize.py
 2023-06-01 23:32:35 +08:00
Fivele-Li
3b56b8e6c0 Optimize summarize task prompt and others (#1533) 
* 1.update prompt;
2.update fetch information method.

* 1.update prompt;
2.save result to markdown;

* 1.get context info from context_manager;
2.run the entire process successfully.
 2023-06-01 21:22:24 +08:00
you-n-g
40e0c329ba Add configurable dataset (#1535) 2023-06-01 20:05:02 +08:00
Xu Yang
e376648860 Merge pull request #1536 from microsoft/xuyang1/add_debug_mode_to_save_cache 
add a debug mode to speed up debug process
 2023-06-01 19:44:17 +08:00
Xu Yang
5f37f32184 update code 2023-06-01 19:38:26 +08:00
Xu Yang
d46b4c1ebf Merge pull request #1534 from microsoft/xuyang1/add_code_implementation_task 
add code implementation task
 2023-06-01 18:13:05 +08:00
Xu Yang
0515524b51 add code implementation code 2023-06-01 18:04:31 +08:00
Xu Yang
cda32d5703 Merge pull request #1532 from microsoft/xuyang1/add-plan-and-config-task-implementation 
add the initial version of plan and config task implementation
 2023-06-01 11:20:04 +08:00
Xu Yang
e2332a004b imporove some words in prompt 2023-06-01 01:09:14 +08:00
Xu Yang
08d9dbccc9 update v1 code containing SLplan and config action 2023-06-01 00:36:04 +08:00
Fivele-Li
e7cd93a36d add base method for summarization; (#1530) 2023-05-31 15:50:34 +08:00
Xu Yang
3919678028 split task into workflow and task to make the strcture more clear 2023-05-31 11:45:25 +08:00
Xu Yang
421b1403b2 Merge pull request #1528 from microsoft/xuyang1/refine_task_and_implement_workflow_task_as_example 
Xuyang1/refine task and implement workflow task as example
 2023-05-31 11:36:36 +08:00
Xu Yang
94102fb742 remove tasktype variable 2023-05-31 11:35:54 +08:00
Cadenza-Li
74a5d7c8af add parse method for summarization; 2023-05-31 00:08:21 +08:00
Xu Yang
ce39b4b6f8 add qlib auto init so logger can display info 2023-05-30 21:52:35 +08:00
Xu Yang
2af35d9c89 second commit 2023-05-30 20:20:16 +08:00
Xu Yang
f37643550b first round 2023-05-30 20:19:58 +08:00
Xu Yang
55611aa43e Merge pull request #1527 from microsoft/xuyang1/add_openai_api_support 
add openai interface support
 2023-05-30 13:44:10 +08:00
Xu Yang
f24253efd2 add openai interface support 2023-05-30 13:42:01 +08:00
Young
7c4f3b8a7d Initial interface for discussion 2023-05-24 12:18:31 +08:00
you-n-g
94268619c4 Update README.md 2023-05-23 09:50:00 +08:00
Huoran Li
8d60a6a02b Resolve RL FIXMES (#1503) 
* Solve several small FIXMEs left in RL

* Add TODO in example

* Minor bugfix

* black
 2023-05-17 16:57:08 +08:00
Fivele-Li
7234308651 Add base config in yml (#1500) 
* path on Windows contains double '/' which may cause open file failed.

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* add baseConfig in yml,user can add new keys or update/drop keys in baseConfig;

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* pip release version 23.1 on Apr.15 2023, CI failed to run, Please refer to #1495 ofr detailed logs. The pip version has been temporarily fixed to 23.0.1.

* 1.Search for baseConfig in multiple directories;
2.Add user instructions in qrun;

* fix format with black

* 1.modify baseConfig key to BASE_CONFIG_PATH;
2.only find config file in absolute path and relative path;

* load BASE_CONFIG_PATH on absolute path & relative path;

* fix Lint with black

---------

Co-authored-by: lijinhui <362237642@qq.com>
 2023-05-12 17:35:37 +08:00
Chaoying
acf5df27ce Add support for redis password (#1508) 2023-05-08 16:17:15 +08:00
Chaoying
37a59f28d3 Fix deprecated syntax in numpy (#1507) 
* Fix deprecated syntax in numpy

* Replace np.bool with bool
 2023-05-08 16:17:02 +08:00
YQ Tsui
b084c352f5 provide dtype to empty series to surpress warning; fix type (#1449) 2023-05-05 17:47:44 +08:00
Maksim Zayakin
9e22e5168b Remove unused DNNModelPytorch params (#1470) 
* Remove lr_decay and lr_decay_steps params

More flexible way to pass a scheduler (via callable function) is already
supported

* remove lr_decay and lr_decay_steps from mlp workflow configs
 2023-04-28 17:48:40 +08:00
Fivele-Li
dceff7b471 Specify the tianshou version to match the dev environment to avoid the error in issue #1477. (#1502) 2023-04-28 13:50:25 +08:00
Huoran Li
7f1e8c5206 Refine Qlib RL data format (#1480) 
* wip

* wip

* wip

* Fix naming errors

* Backtest test passed

* Why training stuck?

* Minor

* Refine train configs

* Use dummy in training

* Remove pickle_dataframe

* CI

* CI

* Add more strict condition to filter orders

* Pass test

* Add TODO in example

---------

Co-authored-by: Young <afe.young@gmail.com>
 2023-04-26 21:14:30 +08:00
Fivele-Li
46264dfec9 normpath for Windows (#1495) 
* path on Windows contains double '/' which may cause open file failed.

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* locate import numpy error

* pip release version 23.1 on Apr.15 2023, CI failed to run, Please refer to #1495 ofr detailed logs. The pip version has been temporarily fixed to 23.0.1.

---------

Co-authored-by: lijinhui <362237642@qq.com>
 2023-04-26 16:26:12 +08:00
Fivele-Li
754799ab05 update ubuntu CI version; (#1488) 
* update ubuntu CI version;
(End of standard support for 18.04 LTS - 31 May 2023)

* update ubuntu CI version;

---------

Co-authored-by: lijinhui <362237642@qq.com>
 2023-04-10 17:06:48 +08:00
you-n-g
32c3070b73 Refine DDG-DA (#1472) 
* Run ddg-da successfully

* Support include valid; More parameters

* Support L2 reg & visualization

* Blackformat

* Enable fill_method

* Support specify handler & optim dataset

* Fix Pylint
 2023-04-07 15:00:21 +08:00
you-n-g
40de67265a Update Docs about some concepts in DataHandler (#1485) 2023-04-07 10:02:16 +08:00