yujunjun/autogen
1
0
Fork 0
mirror of https://github.com/microsoft/autogen.git synced 2025-07-23 00:42:54 +00:00
Code Issues Packages Projects Releases Wiki Activity
887 Commits 201 Branches 121 Tags
Commit Graph

887 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Chi Wang
c35b844030
simplify warmstart in blendsearch (#558) 
* simplify warmstart in blendsearch

* bump version to 1.0.3
 2022-05-28 19:17:40 -07:00
Chi Wang
49e8f7f028
use zeroshot when no budget is given; custom_hp (#563) 
* use zeroshot when no budget is given; custom_hp

* update Getting-Started

* protobuf version

* X_val
 2022-05-28 17:22:09 -07:00
Chi Wang
7748e0ff49
Update Tune-User-Defined-Function.md (#562) 
Added one more use case of tune.
 2022-05-26 11:13:08 -07:00
Prajwal Borkar
f8babac21b
Removed cat_hp_host (#559) 2022-05-26 10:21:51 -07:00
Qingyun Wu
dcc640c1a3
Documentation, test and bugfix (#556) 
* add bs restore test

* use default metric when not provided

* update documentation

* remove print

* period

* remove bs restore test

* Update website/docs/Use-Cases/Task-Oriented-AutoML.md
 2022-05-26 07:12:30 -07:00
LinWencong
515a77ac71
solve issue #542. fix pickle.UnpickingError while blendsearch warm start (#554) 
Issue I encountered:

#542 run test_restore.py and got _pickle.UnpicklingError: state is not a dictionary

I observed:

1. numpy version
  i. When numpy==1.16*, np.random.RandomState.__getstate__() returns a tuple, not a dict.
    _pickle.UnpicklingError occurs
  ii. When numpy>1.17.0rc1, it returns a dict;
    _pickle.UnpicklingError does not occur
  iii. When numpy>1.17.0rc1, flaml uses np_random_generator = np.random.Generator,
    _pickle.UnpicklingError does not occur

2. class _BackwardsCompatibleNumpyRng
  When I remove func _BackwardsCompatibleNumpyRng.__getattr__() , _pickle.UnpicklingError doesn't occur (regardless of numpy version == 1.16* or 1.17*)

To sum up,
I think making modifications to class _BackwardsCompatibleNumpyRng is not a good choice (_BackwardsCompatibleNumpyRng came from ray）and we still need to learn more about the operation mechanism of pickle.

So I upgraded the numpy version that flaml requires：
  setup.py："NumPy>=1.17.0rc1"
 2022-05-23 11:23:00 -07:00
Chi Wang
d402c63312
align indent and add missing quotation (#555) 
* align indent and add missing quotation
 2022-05-20 10:49:39 -07:00
harish445
992ea3416c
fix indentation in automl.py (#553) 
* fix indentation in flaml.py
 2022-05-19 16:29:45 -07:00
Qiaochu Song
2851134052
Quick-fix (#539) 
* fix doc string; enable label transform in automl.score
 2022-05-19 11:43:34 -04:00
Chi Wang
7126b69ce0
choose n_jobs for ensemble according to n_jobs per learner (#551) 
* set n_jobs in ensemble dict

* catch the ensemble error

* choose n_jobs for stacker

* clarify
 2022-05-18 21:01:51 -07:00
Xueqing Liu
2ca9e41e4b
fixing roberta add_prefix_space bug (#546) 
* fixing roberta add_prefix_space bug
 2022-05-12 10:57:25 -04:00
Xueqing Liu
2a8decdc50
fix the post-processing bug in NER (#534) 
* fix conll bug

* update DataCollatorForAuto

* adding label_list comments
 2022-05-10 17:22:57 -04:00
Chi Wang
c1bb66980c
test reproducibility from retrain (#533) 2022-05-07 09:13:17 -07:00
Qingyun Wu
bcdfdc8735
handle non-flaml scheduler in flaml.tune (#532) 
* handle non-flaml scheduler in flaml.tune

* revise time budget

* Update website/docs/Use-Cases/Tune-User-Defined-Function.md

Co-authored-by: Chi Wang <wang.chi@microsoft.com>

* Update website/docs/Use-Cases/Tune-User-Defined-Function.md

Co-authored-by: Chi Wang <wang.chi@microsoft.com>

* Update flaml/tune/tune.py

Co-authored-by: Chi Wang <wang.chi@microsoft.com>

* add docstr

* remove random seed

* StopIteration

* StopIteration format

* format

* Update flaml/tune/tune.py

Co-authored-by: Chi Wang <wang.chi@microsoft.com>

* revise docstr

Co-authored-by: Chi Wang <wang.chi@microsoft.com>
 2022-05-06 14:09:35 -04:00
Xueqing Liu
c1e1299855
fixing use_ray in automl.py (#531) 
* fixing use_ray
 2022-05-02 08:05:23 -07:00
Chi Wang
dab0a3f6e5
Bump cross-fetch from 3.1.4 to 3.1.5 in /website (#529) 
* Bump cross-fetch from 3.1.4 to 3.1.5 in /website

* Bump async from 2.6.3 to 2.6.4 in /website

* cleanup
 2022-04-30 11:54:43 -07:00
Xueqing Liu
ca35fa969f
refactoring TransformersEstimator to support default and custom_hp (#511) 
* refactoring TransformersEstimator to support default and custom_hp

* handling starting_points not in search space

* addressing starting point more than max_iter

* fixing upper < lower bug
 2022-04-28 14:06:29 -04:00
Jaden Kropp
d03038bfcb
docstr cleanup #523: removed lines 259 to 260 in a1c49ca (#524) 2022-04-27 07:50:38 -07:00
Chi Wang
a1c49ca27b
allow evaluated_rewards shorter than points_to_evaluate (#522) 
* allow evaluated_rewards shorter than points_to_evaluate

* docstr update
 2022-04-23 16:22:34 -07:00
Chi Wang
f301bf8d8f
max choice is n-1 (#521) 
close #520
 2022-04-23 15:33:35 -07:00
Jay Shankar
46f80dfa16 Fix AttributeError: readonly attribute for Python 3.10.4 2022-04-15 15:01:06 -07:00
Xueqing Liu
cfed657812
Handling fractional gpu_per_trial for NLP (#513) 
* handling fractional gpu_per_trial
 2022-04-12 14:46:14 -04:00
Chi Wang
e877de6414 use ffill in forecasting example 2022-04-01 09:23:23 -07:00
Chi Wang
84f1ae7424
Bump minimist from 1.2.5 to 1.2.6 in /website (#502) 
* Bump minimist from 1.2.5 to 1.2.6 in /website

* check best_config in test
 2022-03-30 22:19:47 -07:00
Qingyun Wu
2cdc08a75a update notebook and test 2022-03-30 19:11:10 -07:00
Chi Wang
9128c8811a
handle failing trials (#505) 
* handle failing trials

* clarify when to return {}

* skip ensemble in accuracy check
 2022-03-28 16:57:52 -07:00
liususan091219
0bcf618fea fixing bug 2022-03-28 11:53:58 -07:00
Qingyun Wu
6c16e47e42
Bug fix and add documentation for metric_constraints (#498) 
* metric constraint documentation

* update link

* update notebook

* fix a bug in adding 'time_total_s' to result

* use the default multiple factor from config file

* update notebook

* format

* improve test

* revise test budget for macos

* bug fix in adding time_total_s

* increase performance check budget

* revise test

* update notebook

* uncomment test

* remove redundancy

* clear output

* remove n_jobs

* remove constraint in notebook

* increase budget

* revise test

* add python version

* use getattr

* improve code robustness

Co-authored-by: Qingyun Wu <qxw5138@psu.edu>
 2022-03-26 21:11:45 -04:00
Xueqing Liu
72301b8568
fixing a few bugs in nlp (#503) 
* fixing bugs in nlp
 2022-03-26 14:08:51 -04:00
Chi Wang
7eb7b46ea9
version number and doc (#497) 
* version number

* add missing tasks in documentation

* update node-forge version
 2022-03-25 17:32:37 -07:00
Xueqing Liu
5f97532986
adding evaluation (#495) 
* adding automl.score

* fixing the metric name in train_with_config

* adding pickle after score

* fixing a bug in automl.pickle
 2022-03-25 17:00:08 -04:00
Chi Wang
1d029436e7 doc update (#490) 2022-03-21 12:31:36 -07:00
Xueqing Liu
af423463c3
fixing bug for ner (#463) 
* fixing bug for ner

* removing global var

* adding class for trial counter

* adding notebook

* adding use_ray dict

* updating documentation for nlp
 2022-03-20 22:03:02 -04:00
Chi Wang
5f688c1662 lightgbm 2022-03-20 16:27:47 -07:00
Qingyun Wu
f6ae1331f5
metric constraints in flaml.automl (#479) 
* metric constraints

* revise docstr

* fix docstr

* improve docstr

* Update flaml/automl.py

Co-authored-by: Chi Wang <wang.chi@microsoft.com>

* Update flaml/automl.py

Co-authored-by: Chi Wang <wang.chi@microsoft.com>

* Update flaml/automl.py

Co-authored-by: Chi Wang <wang.chi@microsoft.com>

* docstr

Co-authored-by: Qingyun Wu <qxw5138@psu.edu>
Co-authored-by: Chi Wang <wang.chi@microsoft.com>
 2022-03-12 00:39:35 -05:00
Kevin Chen
f9eda0cc40
update documentation for time series forecasting (#472) 
* update automl.py
- documentation update

* update test_forecast.py

* update model.py

* update automl_time_series_forecast.ipynb

* update time series forecast website examples

Signed-off-by: Kevin Chen <chenkevin.8787@gmail.com>
 2022-03-08 11:21:18 -08:00
Chi Wang
f0b0cae682
zero-shot AutoML in readme (#474) 
* zero-shot AutoML in readme

* use pydoc-markdown 4.5.0 to avoid error in 4.6.0
 2022-03-05 11:49:39 -08:00
Chi Wang
31ac984c4b
don't init global search with points_to_evaluate unless evaluated_rewards is provided; handle callbacks in fit kwargs (#469) 2022-03-01 18:39:16 -08:00
Chi Wang
df01031cfe
Zero-shot AutoML (#468) 
* Prepare for release

Co-authored-by: Moe Kayali <t-moekayali@microsoft.com>

* bug fix

* improve doc and code quality

Co-authored-by: Qingyun Wu
 2022-03-01 15:39:09 -08:00
Chi Wang
01ca0422ea Bump prismjs from 1.25.0 to 1.27.0 in /website 2022-02-26 16:09:43 -08:00
Chi Wang
e3e737c71a
make AutoML.classes_ an array (#467) 
* remove .tolist()

* docstr
 2022-02-25 22:13:41 -08:00
Chi Wang
6095be039f
update regression metrics in notebooks (#454) 2022-02-15 13:17:43 -08:00
Qingyun Wu
05f9065ade
Docstr update (#460) 
* parallel tuning docstr

* update n_concurrent_trials docstr

* n_jobs default

* parallel tuning in tune docstr
 2022-02-15 09:41:53 -08:00
Chi Wang
393106d531 Bump follow-redirects from 1.14.7 to 1.14.8 in /website 2022-02-15 08:24:37 -08:00
Camila Martinez Bedoya
dea2beec34 Update Getting-Started.md 
Solving issue #456 related to a typo on the webpage.
 2022-02-12 20:19:33 -08:00
Chi Wang
9e88f22167
fix a bug when using ray & update ray on aml (#455) 
* fix a bug when using ray & update ray on aml
When using with_parameters(), the config argument must be the first argument in the trainable function.
* make training function runnable standalone
 2022-02-11 20:14:10 -08:00
Chi Wang
b4d312412a
bump ray version to 1.10 (#450) 
* bump ray version to 1.10

* init ray in test

* Update setup.py to include hotfixes

Co-authored-by: Antoni Baum <antoni.baum@protonmail.com>
 2022-02-09 15:04:29 -08:00
MichaelMarien
a4970d279c
corrected typo in example xgboost documentation (#449) 2022-02-04 13:03:09 -08:00
vvijayalakshmi21
b930de0e1f
Update Tune-User-Defined-Function.md (#448) 
Fix for #445
 2022-02-03 09:40:57 -08:00
vvijayalakshmi21
34557de30c
Update Task-Oriented-AutoML.md (#447) 
Fix for issue #444
 2022-02-03 09:04:49 -08:00