microsoft
/
LightGBM
зеркало из https://github.com/microsoft/LightGBM.git
1
0
Форкнуть 0
Код Задачи Пакеты Проекты Релизы Вики Активность
1 087 коммитов 24 Ветки 39 Теги 80 MiB
Граф коммитов

83 Коммитов

Автор SHA1 Сообщение Дата
wxchan a4a0235d17 use json instead of repr/eval for pandas_categorical (#247) 
* use json instead of repr/eval for pandas_categorical

* fix json dumps with numpy data

* add more test cases
 2017-01-23 19:06:36 +08:00
wxchan abaefb54ce [python-package] add plot importance (#237) 
* add plot importance

* add plot example
 2017-01-20 13:53:18 +08:00
wxchan 57d552726f fix bug for pandas auto categorical_feature (#218) 
* fix bug for categorical_feature

* add test on load model with categorical feature

* add unseen category in test dataset

* save/load pandas_categorical to model

* fix logic

* cast pandas columns to string

* add load pandas_categorical from file to _InnerPredictor init
 2017-01-16 17:01:48 +08:00
ClimbsRocks b0f7aa508a sklearn compatibility update- renames .feature_importance_ to .feature_importances_ 2017-01-12 00:01:37 -08:00
wxchan 6c248d37c3 suppprt pandas categorical (#193) 
* suppprt pandas categorical

* refine logic

* make default=auto

* fix train/valid categorical codes

* add test

* unify set _predictor

* fix tests

* fix warning

* support feature_name=int
 2017-01-12 15:27:35 +08:00
wxchan 7f4610a8ad refine pmml.py (#179) 
* add pmml to test

* refine pmml.py

* use ~n instead of -n-1

* change map to list comprehension

* fix check

* fix 'use ~n instead of -n-1'

* fix exception
 2017-01-10 00:27:52 +08:00
wxchan 1b7643ba60 `_is_constructed` -> `handle is not None`; add FAQ for docs (#173) 
* use handle is not None for _is_constructed

* sort imports; clean code; move FAQ to docs
 2017-01-08 16:51:04 +08:00
Guolin Ke 551d59ca71 R package (#168) 
* finish R's c_api

* clean code

* fix sizeof pointer in 32bit system.

* add predictor class

* add Dataset class

* format code

* add booster

* add type check for expose function

* add a simple callback

* add all callbacks

* finish the basic training logic

* update docs

* add an simple training interface

* add basic test

* adapt the changes in c_api

* add test for Dataset

* add test for custom obj/eval functions

* fix python test

* fix bug in metadata init

* fix R CMD check
 2017-01-08 09:13:36 +08:00
wxchan e29ab9f682 fix reset parameter; re-define CVBooster (#166) 
* fix reset parameter

* redefine CVBooster

* env.model won't be None

* update env.params
 2017-01-06 21:37:35 +08:00
Guolin Ke 21ee59476e fix double to string precision (std::numeric_limits<double>::digits10 + 2) 2017-01-04 16:26:39 +08:00
wxchan dd425973a5 python code style with pep8 (#161) 
* format python code with pep8

* **DO NOT MERGE** deliberately break rules to see what will happen during check

* Revert "**DO NOT MERGE** deliberately break rules to see what will happen during check"

This reverts commit 0db93cd7a4.

* fix format in test.py

* add docs for pep-8
 2017-01-04 15:19:12 +08:00
Guolin Ke cf4edf0e39 add test for check model file persistence 2017-01-04 09:31:04 +08:00
wxchan 1c6c7046f0 add @property to sklearn interface (#155) 
* add @property to sklearn interface

* add deprecated; fix binary_metric
 2017-01-03 13:44:26 +08:00
Guolin Ke f6024c8bd3 fix test for continued train, due to default saved number of model is best_iteration now 2017-01-02 12:38:10 +08:00
Guolin Ke 28972b8667 [python-package] fix tmp file access problem in windows 2017-01-02 10:56:46 +08:00
wxchan a034ceeb3a support pickle (#151) 
* support pickle

* add pickle/joblib test; change test_basic to unittest

* remove file for deepcopy

* fix tests

* test basic predict from file

* Revert "test basic predict from file"

This reverts commit 60d2c31585.

* test predict from file

* use tempfile for copy & pickle

* use tempfile w/o binary mode

* clean test
 2017-01-01 21:05:44 +08:00
wxchan bd7274baee add callbacks to sklearn interface (#150) 2016-12-31 11:12:00 +08:00
wxchan 49178defdc add dart parameter to sklearn interface (#145) 
* remove other parameters in sklearn fit

* add dart parameters to sklearn init
 2016-12-28 21:39:50 +08:00
wxchan db18367e32 rewrite test with unittest (#142) 
* add troubleshooting

* use unittest

* update unittest version

* fix test_engine.py

* fix test_sklearn.py

* default eval_metric by subclass

* add test grid search

* remove verbose_eval
 2016-12-23 17:43:53 +08:00
wxchan 99b483dd8d add learning rate to dart (#133) 
add learning rate to dart (#133)
 2016-12-19 23:50:15 +08:00
Guolin Ke b51c7be43e refine Dataset class (#113) 
Provide a high level Dataset class for easy use.
 2016-12-08 21:36:11 +08:00
wxchan 2cd024e9c6 add feature importance in python (#109) 
* add feature importances in python; add pandas support

* solve best_iteration issue
 2016-12-06 14:35:54 +08:00
Guolin Ke 1466f90781 Categorical feature support (#108) 
Categorical feature support (#108)
 2016-12-05 12:51:00 +08:00
wxchan 6911452588 solve conflicts; add README for python examples 2016-12-02 17:35:39 +08:00
wxchan eba6d200a8 Squash into one commit: 
1. merge python-package
2. add dump model to json
3. fix bugs
4. clean code with pylint
5. update python examples
 2016-12-02 14:34:52 +08:00
Guolin Ke 164524d8fb weighted objective function 2016-11-30 15:20:06 +08:00
Guolin Ke b59a5a4c11 test for early_stopping 2016-11-30 13:39:19 +08:00
Guolin Ke f8267a5044 add min_data, fix test 2016-11-30 13:19:27 +08:00
Guolin Ke f65164f67e less verbose in test 2016-11-30 12:46:30 +08:00
Guolin Ke c67d289086 move pandas support into basic.py 2016-11-30 12:24:32 +08:00
Guolin Ke 452b41f015 add more tests 2016-11-30 11:40:31 +08:00
Guolin Ke 44fcf16c6f fix travis 2016-11-30 10:27:27 +08:00
Guolin Ke b8d9372efd Add simple test 2016-11-24 18:21:22 +08:00