Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[CUDA] Add quantile regression objective for new CUDA version #5605

Merged
merged 60 commits into from
Mar 21, 2023
Merged
Changes from 2 commits
Commits
Show all changes
60 commits
Select commit Hold shift + click to select a range
7eaf3d8
add binary objective for cuda_exp
shiyu1994 Aug 16, 2022
ec84dcc
include <string> and <vector>
shiyu1994 Aug 16, 2022
a00f272
exchange include ordering
shiyu1994 Aug 16, 2022
ce26a61
fix length of score to copy in evaluation
shiyu1994 Aug 17, 2022
3429b68
Merge branch 'master' into cuda/objective-binary
shiyu1994 Aug 29, 2022
e91ec0b
fix EvalOneMetric
shiyu1994 Aug 30, 2022
a7641ca
fix cuda binary objective and prediction when boosting on gpu
shiyu1994 Aug 30, 2022
a6cc1f6
Add white space
shiyu1994 Aug 30, 2022
3b354ea
fix BoostFromScore for CUDABinaryLogloss
shiyu1994 Aug 30, 2022
2ead53a
include <algorithm>
shiyu1994 Aug 30, 2022
a78b146
simplify shared memory buffer
shiyu1994 Aug 30, 2022
6dc035a
add (l2) regression objective for cuda_exp
shiyu1994 Aug 31, 2022
b202478
Merge branch 'master' into cuda/objective-regression
shiyu1994 Aug 31, 2022
ac52e75
fix lint errors
shiyu1994 Aug 31, 2022
1708d42
Merge branch 'cuda/objective-regression' of https://github.com/Micros…
shiyu1994 Aug 31, 2022
bd0010f
add (l1) regression objective for cuda_exp
shiyu1994 Sep 1, 2022
063d066
merge LightGBM/master
shiyu1994 Sep 1, 2022
8a581ed
remove RenewTreeOutputCUDA from CUDARegressionL2loss
shiyu1994 Sep 1, 2022
95898fd
remove mutable and use CUDAVector
shiyu1994 Sep 1, 2022
3a1af9c
remove white spaces
shiyu1994 Sep 1, 2022
1107e32
remove TODO and document in (#5459)
shiyu1994 Sep 1, 2022
0b5a63a
Merge remote-tracking branch 'origin/master' into cuda/objective-regr…
shiyu1994 Sep 2, 2022
67820d8
add huber regression for cuda_exp
shiyu1994 Sep 2, 2022
8e4fb0f
renew tree output on GPU
shiyu1994 Sep 2, 2022
57f65a9
remove useless changes
shiyu1994 Sep 2, 2022
4bf0d15
add white space
shiyu1994 Sep 2, 2022
2b7299a
Merge branch 'cuda/objective-regression' of https://github.com/Micros…
shiyu1994 Sep 2, 2022
713a391
change percentile in CUDARegressionL1loss::BoostFromScore to 0.5
shiyu1994 Sep 6, 2022
502267a
add fair regression objective
shiyu1994 Sep 6, 2022
2e93075
remove useless changes in test_engine.py
shiyu1994 Sep 6, 2022
e918b35
Merge remote-tracking branch 'origin/master' into cuda/objective-regr…
shiyu1994 Sep 15, 2022
b0b89ca
add poisson regression objective for cuda_exp
shiyu1994 Sep 15, 2022
0fbdfa8
enable Poisson regression for cuda_exp
shiyu1994 Sep 15, 2022
2e31f9e
refactor cuda objective functions
shiyu1994 Sep 26, 2022
0ee495d
remove useless changes
shiyu1994 Sep 26, 2022
be4ad32
fix linter errors
shiyu1994 Sep 26, 2022
7d6c7c0
Merge branch 'cuda/objective-regression' of https://github.com/Micros…
shiyu1994 Sep 26, 2022
4d2ad0e
remove redundant buffer in cuda poisson regression objective
shiyu1994 Sep 27, 2022
e4e3202
fix log of cuda_exp binary objective
shiyu1994 Sep 27, 2022
1ccb55d
fix threshold of poisson objective result
shiyu1994 Sep 27, 2022
545ffef
remove useless changes
shiyu1994 Sep 27, 2022
29d5735
Merge branch 'master' into cuda/objective-regression
jameslamb Nov 4, 2022
dcd932d
Merge branch 'master' into cuda/objective-regression
shiyu1994 Nov 9, 2022
c7f7d8b
fix compilation errors
shiyu1994 Nov 11, 2022
1cf097d
Merge branch 'master' into cuda/objective-regression
jameslamb Nov 21, 2022
549df0d
add cuda quantile regression objective
shiyu1994 Nov 25, 2022
73cb6ba
Merge branch 'cuda/objective-regression' of https://github.com/Micros…
shiyu1994 Nov 25, 2022
2e986d5
merge master
shiyu1994 Dec 7, 2022
0b80042
remove white space
shiyu1994 Dec 13, 2022
3df130e
merge master
shiyu1994 Dec 28, 2022
d0a7610
merge with origin/cuda/objective-regression-quantile
shiyu1994 Dec 28, 2022
c29e688
resolve merge conflicts
shiyu1994 Dec 28, 2022
73aaa79
remove useless changes
shiyu1994 Dec 28, 2022
ab465ea
remove useless changes
shiyu1994 Dec 28, 2022
8f489aa
Merge remote-tracking branch 'origin/master' into cuda/objective-regr…
shiyu1994 Mar 15, 2023
616b600
enable cuda quantile regression objective
shiyu1994 Mar 15, 2023
c9443f7
add a test case for quantile regression objective
shiyu1994 Mar 15, 2023
4ca25d0
remove useless changes
shiyu1994 Mar 15, 2023
58bfd9b
remove useless changes
shiyu1994 Mar 15, 2023
e5c7af6
reduce DP_SHARED_HIST_SIZE to 5176 for CUDA 10
shiyu1994 Mar 17, 2023
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions tests/python_package_test/test_engine.py
Original file line number Diff line number Diff line change
@@ -128,7 +128,7 @@ def test_regression(objective):
lgb_train,
num_boost_round=50,
valid_sets=lgb_eval,
callbacks=[lgb.record_evaluation(evals_result), lgb.log_evaluation(1)]
callbacks=[lgb.record_evaluation(evals_result)]
)
ret = mean_squared_error(y_test, gbm.predict(X_test))
if objective == 'huber':
@@ -1196,7 +1196,7 @@ def test_feature_name_with_non_ascii():
X_train = np.random.normal(size=(100, 4))
y_train = np.random.random(100)
# This has non-ascii strings.
feature_names = [u'F_�?', u'F_一', u'F_�?', u'F_�?']
feature_names = [u'F_零', u'F_一', u'F_二', u'F_三']
params = {'verbose': -1}
lgb_train = lgb.Dataset(X_train, y_train)