Skip to content

Issues: microsoft/LightGBM

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.

Sign up for GitHub

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

382 Open 3,114 Closed
382 Open 3,114 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

ValueError: not enough values to unpack (expected 2, got 0) in lgb.plot_importance()
#6865 opened Mar 14, 2025 by Dixhom
I cannot replicate LightGBM's refit behaviours when using a custom loss function question
#6864 opened Mar 12, 2025 by mlondschien
4
[python-package] [R-package] every test catching an error should check for a specific error message good first issue maintenance r-package
#6860 opened Mar 11, 2025 by jameslamb
30 tasks
[python-package] column in 'weight_column' is still included in features bug
#6856 opened Mar 8, 2025 by Fearycxy
3
[python-package] Load a lightgbm.LGBMRegressor from file awaiting response question
#6854 opened Mar 7, 2025 by EnnFour
3
Shrinkage different on first tree between default and custom objective question
#6853 opened Mar 4, 2025 by Shanxce
2
[python-package] Model parameters are lost when loading Booster from model serialized string bug duplicate
#6851 opened Mar 2, 2025 by Samsagax
1
4
[python-package] Fitting on Polars Dataframe fails due to missing setter for fetures_names_in_ bug
#6849 opened Feb 28, 2025 by FelicitasTengenQC
@borchero
1
[R-package] Printing a booster trained with lgb.train() errors unexpectedly bug r-package
#6848 opened Feb 28, 2025 by walkerjameschris
1
2
[CUDA] Potential missed inversion of discounts within CUDA implementation of LambdarankNDCG question
#6847 opened Feb 28, 2025 by DJ-Hayden
@shiyu1994
2
Training crashes when "monotone_constraints_method" is set to "advanced" bug
#6841 opened Feb 26, 2025 by sv1990
2
[CUDA] Defaulting to malloc in CHAllocator question
#6840 opened Feb 24, 2025 by pr1nce-harry
3
arrow.h brings lots of C++ to the c_api.h header bug
#6839 opened Feb 20, 2025 by Ten0
@jameslamb
3
What does booster.refit actually do? question
#6838 opened Feb 20, 2025 by mlondschien
1
CUDA backend uses significantly more memory than OpenCL question
#6835 opened Feb 19, 2025 by adamreeve
4
[python-package] lgb.train() init_model does not allow a model that has fewer features than the training data question
#6831 opened Feb 16, 2025 by vnherdeiro
2
[c++] forcedsplits_filename pointing at a non-existent file is silently ignored bug good first issue
#6830 opened Feb 15, 2025 by jameslamb
1
1
[ci] [python-package] [R-package] support MUSL-based Linux distributions (like Alpine) feature request maintenance r-package
#6829 opened Feb 15, 2025 by jameslamb
2
[RFC] [CUDA] distribute Python wheels with CUDA support feature request
#6828 opened Feb 15, 2025 by jameslamb
2
NuGet package is missing README doc good first issue
#6827 opened Feb 15, 2025 by jameslamb
How is NDCG computed for queries where there is only one document? question
#6823 opened Feb 11, 2025 by cjsombric
2
[python-package] [dask] AttributeError: 'Future' object has no attribute 'get_params()' in DaskLGBMClassifier.fit when using Dask Gateway dask
#6822 opened Feb 11, 2025 by rewaishi
3
What happens to missing data during training when use_missing=False? question
#6819 opened Feb 7, 2025 by hsorsky
1
Rank Evaluation Metric NDCG inconsistent between LGBMRanker.fit vs. LGBMRanker.predict() question
#6814 opened Feb 6, 2025 by cjsombric
5
[GPU] GOSS boosting error on GPU H100 question
#6811 opened Feb 2, 2025 by SergeevVladislav
4
ProTip! Follow long discussions with comments:>50.