Commits · b7ed7395c1eb555bd3e5462a43adaa172460615c · tianlh / LightGBM-DCU

02 Jan, 2017 3 commits
- Fix the initializer list of GBDT (#156) · b7ed7395
  Tsukasa OMOTO authored Jan 02, 2017
```
This commit fixes https://github.com/Microsoft/LightGBM/issues/154#issuecomment-269908741.
```
  b7ed7395
- fix test for continued train, due to default saved number of model is best_iteration now · f6024c8b
  Guolin Ke authored Jan 02, 2017
  
  f6024c8b
- [python-package] fix tmp file access problem in windows · 28972b86
  Guolin Ke authored Jan 02, 2017
  
  28972b86
01 Jan, 2017 1 commit
- change output model to double precision. · 1d7acf57
  Guolin Ke authored Jan 01, 2017
  
  1d7acf57
31 Dec, 2016 2 commits
- fix bug for save binary for non-filename dataset · 487bd835
  Guolin Ke authored Dec 31, 2016
  
  487bd835
- some refine for c_api (#152) · 72c2d790
  Guolin Ke authored Dec 31, 2016
```
1. add csc support
2. some data type from float to double
```
  72c2d790
28 Dec, 2016 3 commits
- [python-package] pass params of engine.train and engine.cv to Dataset. · 616388e0
  Guolin Ke authored Dec 28, 2016
  
  616388e0
- decouple num_class in Dataset class · 292f972e
  Guolin Ke authored Dec 28, 2016
  
  292f972e
- refine predictor logic in c_api · 45e0da2c
  Guolin Ke authored Dec 28, 2016
  
  45e0da2c
27 Dec, 2016 1 commit
- quick fix for #144 · 728e50a9
  Guolin Ke authored Dec 27, 2016
  
  728e50a9
26 Dec, 2016 1 commit
- Adapt learning rate to DART and xgboost dart mode(#139) · 96cba416
  Guolin Ke authored Dec 26, 2016
  
  96cba416
24 Dec, 2016 2 commits
- use int64_t for num_class * num_data · 1cacaef9
  Guolin Ke authored Dec 24, 2016
  
  1cacaef9
- reduce memory cost for multi classification · f449a45b
  Guolin Ke authored Dec 24, 2016
  
  f449a45b
23 Dec, 2016 5 commits
- fix bug in GetPredictAt for multi class · 381a945d
  Guolin Ke authored Dec 23, 2016
  
  381a945d
- fix bug in multi_error · 8aa15e88
  Guolin Ke authored Dec 23, 2016
  
  8aa15e88
- add get/set c_api for leaf value · 233f1b71
  Guolin Ke authored Dec 23, 2016
  
  233f1b71
- replace const handle* to const handle · d38ada73
  Guolin Ke authored Dec 23, 2016
  
  d38ada73
- typo · 26b7e818
  Guolin Ke authored Dec 23, 2016
  
  26b7e818
20 Dec, 2016 2 commits
- fix size of feature_name error when tail features are all zeros · da031707
  Guolin Ke authored Dec 20, 2016
  
  da031707
- fix bug in ResetTrainingData · 9db054cf
  Guolin Ke authored Dec 20, 2016
  
  9db054cf
19 Dec, 2016 4 commits
- add learning rate to dart (#133) · 99b483dd
  wxchan authored Dec 19, 2016
```
add learning rate to dart (#133)
```
  99b483dd
- better performance for reset parameters · 574d5342
  Guolin Ke authored Dec 19, 2016
  
  574d5342
- fix #132 · 25200c3a
  Guolin Ke authored Dec 19, 2016
  
  25200c3a
- cannot reset random object in reset_parameters · fca93b78
  Guolin Ke authored Dec 19, 2016
  
  fca93b78
18 Dec, 2016 4 commits
- refine reset_parameters logic · c2e94f17
  Guolin Ke authored Dec 18, 2016
  
  c2e94f17
- fix bugs in reset parameters · 714c6732
  Guolin Ke authored Dec 18, 2016
  
  714c6732
- bug fix in get_group · 97df0684
  Guolin Ke authored Dec 18, 2016
  
  97df0684
- fix bug in set_query · cbc56c74
  Guolin Ke authored Dec 18, 2016
  
  cbc56c74
15 Dec, 2016 2 commits
- quick fix for #125 · c7ef8322
  Guolin Ke authored Dec 15, 2016
  
  c7ef8322
- fix bug in save binary · 4e402370
  Guolin Ke authored Dec 15, 2016
  
  4e402370
13 Dec, 2016 2 commits
- add voting based parallel · 762b5707
  Guolin Ke authored Dec 13, 2016
  
  762b5707
- bug fixed in save binary file · 26d33232
  Guolin Ke authored Dec 13, 2016
  
  26d33232
08 Dec, 2016 1 commit
- refine Dataset class (#113) · b51c7be4
  Guolin Ke authored Dec 08, 2016
```
Provide a high level Dataset class for easy use.
```
  b51c7be4
07 Dec, 2016 1 commit
- simplify cmake · 70c31c07
  Guolin Ke authored Dec 07, 2016
  
  70c31c07
06 Dec, 2016 1 commit

add feature importance in python (#109) · 2cd024e9

wxchan authored Dec 06, 2016

* add feature importances in python; add pandas support

* solve best_iteration issue

2cd024e9

05 Dec, 2016 4 commits
- fix bug in save model · 6f7669df
  Guolin Ke authored Dec 06, 2016
  
  6f7669df
- update omp_set_num_threads in c_api · 1be76baf
  Guolin Ke authored Dec 05, 2016
  
  1be76baf
- use ".empty()" to check container · 866a2f91
  Guolin Ke authored Dec 05, 2016
  
  866a2f91
- Categorical feature support (#108) · 1466f907
  Guolin Ke authored Dec 05, 2016
```
Categorical feature support (#108)
```
  1466f907
02 Dec, 2016 1 commit

Squash into one commit: · eba6d200

wxchan authored Dec 02, 2016

1. merge python-package
2. add dump model to json
3. fix bugs
4. clean code with pylint
5. update python examples

eba6d200