b6ad5a6ed3
avoid selecting the same components too many times ( #654 )
2025-03-05 13:12:08 +08:00
b49505be8e
feat: variable printing tool of data_science coder testing ( #658 )
...
* update debug_info printing
* refine infomation printing
* update ensemble evaluator prompt
* substitute repr with reprlib.repr
* strip comments
2025-03-05 13:08:30 +08:00
2ed6ec53d9
feat: add do_truncate control for the load function ( #656 )
...
feat: add do_truncate control for the load function (#656 )
2025-03-04 15:18:36 +08:00
30b52d7d57
scores.csv should not be rewritten ( #653 )
2025-02-28 19:56:20 +08:00
d753daf758
update Wechat group QR code ( #652 )
2025-02-28 15:39:52 +08:00
74cfc5b906
feat: add type checker to api backend & align litellm and old backend ( #647 )
...
* move cache auto continue and retry to all api backend
* add type checker to json mode output
* fix CI
* feat: Add json_mode handling and streaming support in chat completion function
* lint
* fix a bug when returning a dict which value could contain int or bool
* remove litellm
---------
Co-authored-by: Xu Yang <xuyang1@microsoft.com >
Co-authored-by: Young <afe.young@gmail.com >
2025-02-28 15:13:43 +08:00
58910d2ce6
docs: Add guidelines for competition participation in prompts.yaml ( #651 )
2025-02-28 10:40:14 +08:00
bd258fe970
fix workflow eval logic ( #650 )
2025-02-27 22:07:20 +08:00
68bf5665f0
fix: fix some bugs (ensemble output, HPO, model tuning) ( #648 )
2025-02-27 18:28:13 +08:00
a57e625b17
feat: add eda to data science scenario ( #639 )
...
* add eda to data science scenario
* fix CI
2025-02-26 22:35:44 +08:00
d7da15c826
feat: add max time config to costeer in data science ( #645 )
...
* add max time config to costeer
* fix a small bug
---------
Co-authored-by: Xu Yang <xuyang1@microsoft.com >
2025-02-26 18:36:51 +08:00
ac7c2155e4
fix: clarify cross_validation ( #644 )
2025-02-26 17:52:30 +08:00
bccf5aed8a
fix: Set default value for 'entry' parameter in Env.run method ( #643 )
2025-02-26 16:51:38 +08:00
5bb73ff6cb
chore: sample unique labels ( #642 )
...
* sample unique labels
* reformat
2025-02-26 15:24:39 +08:00
b772fe2191
load code from file dict instead of folder ( #641 )
...
Co-authored-by: Xu <v-xuminrui@microsoft.com >
2025-02-26 14:21:28 +08:00
28f072da31
change ensemble metric prompt's code tips ( #640 )
2025-02-26 11:40:25 +08:00
34e143cd66
fix: avoid try-except in ensemble eval prompts ( #637 )
...
* avoid try-except in prompts
* remove try-except limit from coder prompt
* add no try-except criteria in ensemble eval
2025-02-25 23:06:41 +08:00
e73db7e0d7
ignore new numeric type ( #638 )
2025-02-25 18:20:14 +08:00
f120341c9c
fix: qlib and other scenario bugs ( #636 )
...
* fix qlib and other scenario bugs
* bug fix
2025-02-25 12:06:07 +08:00
54cd73eb85
fix: add ensemble test, change to "use cross-validation if possible" in workflow spec ( #634 )
...
* change to "use cross-validation if possible" in workflow spec
* Limit the evaluation indicator to only one
* add metric tips
* string change
2025-02-24 20:28:32 +08:00
20bd55c3a5
fix: fix a bug in initial tasks ( #635 )
...
* fix a bug in initial tasks
* fix ci
2025-02-24 20:23:54 +08:00
403d7fcf80
feat: Make system prompt role customizable in LLM settings ( #632 )
2025-02-23 00:52:30 +08:00
f5761b1f2d
feat: improve the framework's ability to adaptively adjust the model ( #629 )
2025-02-21 22:46:14 +08:00
d4ade36184
refine component choosing part ( #630 )
2025-02-21 18:35:36 +08:00
982b43e4fd
feat: add output_path to load function of LoopBase ( #628 )
...
* feat: add output_path to load from checkpoint function
* add default value to output_path
* sort import
* sort imports
---------
Co-authored-by: Xu <v-xuminrui@microsoft.com >
2025-02-21 16:46:58 +08:00
b64ab18caa
fix: fix a bug in cross validation ( #618 )
...
* fix a bug in cross_validation
* fix cross-validation prompt position
* fix
* fix
* append all error messages and full traceback details in execution
---------
Co-authored-by: Bowen Xian <xianbowen@outlook.com >
2025-02-20 16:18:40 +08:00
433594297a
fix: fix some bugs in workflow unit test ( #624 )
...
* fix some bugs in workflow unit test
* fix ci
2025-02-20 15:42:17 +08:00
dff89d2950
refactor: add run_ret_code method and update run method to use it ( #623 )
...
* refactor: Add run_ret_code method and update run method to use it
* feat: Add kwargs support to run methods and test for run_ret_code
* fix: preserve exit code after chmod in DockerEnv entry command
* chore: Change file permissions from 755 to 644 in env_tpl directory
* refactor: Return execution code and update evaluator logic
* lint
* refactor: Use MappingProxyType for running_extra_volume in DockerEnv methods
* lint
2025-02-20 00:42:10 +08:00
0bd8366254
several bug fix ( #621 )
2025-02-19 22:03:39 +08:00
2222897b3f
fix: change ensemble test ( #622 )
...
* change ensemble test
* remove redundant info
* fix bug
* add model name duplication check in ensemble test
* change text
2025-02-19 21:26:09 +08:00
fc5f295de7
fix: move docker timeout message to __run() ( #620 )
...
* move docker timeout msg to __run()
* fix CI
2025-02-19 19:09:52 +08:00
05c7712770
update qrcode ( #616 )
2025-02-19 13:44:40 +08:00
1737aff7c9
fix: Runnable on first complete & Rename method to next_incomplete_component for clarity ( #615 )
2025-02-19 01:00:25 +08:00
464222094f
add a check for the name of ensemble's score in scores.csv ( #614 )
2025-02-18 22:24:11 +08:00
42d15623ca
fix: move next_component_required logic to DSTrace class and accurate implement ( #612 )
...
* refactor: Move next_component_required logic to DSTrace class
* lint
2025-02-18 21:56:32 +08:00
6c20263e2e
fix: sota comparison logic ( #608 )
...
* concat sota and current result
* fix the bug in current_exp's score vs sota's score
* fix ci
---------
Co-authored-by: yuanteli <1957922024@qq.com >
2025-02-18 20:43:28 +08:00
50e637f9cd
add PYTHONUNBUFFERED=1 for docker env ( #613 )
2025-02-18 20:07:49 +08:00
1605cddb40
feat: add loop_n parameter to the main loop ( #611 )
...
* add loop_n parameter to main loop
* complete the loop_n
---------
Co-authored-by: Xu <v-xuminrui@microsoft.com >
2025-02-18 19:22:52 +08:00
75e496d461
fix restart bug ( #609 )
2025-02-18 15:39:34 +08:00
9fcd5d473e
refactor: replace Evaluator with CoSTEEREvaluator in eva_utils.py ( #607 )
...
* refactor: Replace Evaluator with CoSTEEREvaluator in eva_utils.py
* fix bug
---------
Co-authored-by: Linlang <Lv.Linlang@hotmail.com >
2025-02-18 13:38:13 +08:00
5f1c814355
fix: a small bug in exp_gen ( #606 )
2025-02-17 22:40:05 +08:00
21d88f38d0
fix: fix some minor bugs and add AutoML & cross-validation ( #604 )
...
* fix some complex sentences
* add AutoML in model and cross-validation in workflow
2025-02-17 19:14:50 +08:00
fd27842a98
print head for txt files and shuffle when sample ( #603 )
2025-02-17 18:44:26 +08:00
bddb5eb935
fix: docker error will trigger retry and data science runner loop set to 3 ( #602 )
...
* docker error will trigger retry
data science runner loop set to 3
* fix CI
2025-02-17 17:36:05 +08:00
ffc85936f1
refactor: refactor core framework to better propogate feedbacks ( #599 )
...
* refactor: Update type annotations and remove unused class in evolving modules
* refactor: Simplify evolving agent and feedback handling in CoSTEER module
* lint & CI
* mypy
* ruff for core
* mypy
* refactor: remove unnecessary comments and update feedback handling logic
* refactor: Add prev_task_feedback parameter to evolving strategies
* feat: Clear folder before extracting zip file in DockerEnv
* fix: Correct retrieval of last experiment from history
2025-02-16 01:40:44 +08:00
241c05ccac
chore: print txt info ( #600 )
...
* print head for txt files
* update sample
2025-02-14 19:16:28 +08:00
50e020b75d
fix: auto continue small bug ( #598 )
2025-02-14 17:03:57 +08:00
d1c005ff6d
fix: fix a bug in proposal (add last loop's exception to last task desc) ( #596 )
...
* fix a bug in proposal (add last loop's exception to last task desc)
* fix
2025-02-14 16:36:49 +08:00
96ee62c7a2
small bugs in cache rerun ( #597 )
2025-02-14 16:24:26 +08:00
c321ebfbcb
fix: fix some bugs in the ensemble component ( #595 )
...
* allow the LLM in the ensemble to focus on the significance of metrics
* fix a bug in feedback
* prun spec
* fix a bug in ensemble
* delete unnecessary prompts
* fix the logic in spec
* generalize dataset split
2025-02-14 15:32:40 +08:00
Haoran Pan