[GPU] LightGBMError: bin size 3142 cannot run on GPU

[crs1910]

Description

Hi, I encountered an error which has been posted by many people while training a LightGBM model using the GPU: LightGBMError: bin size 3142 cannot run on GPU. This error seems to be related to the max_bin parameter setting. I tried reducing the bin size, but it has not resolved the issue.Just wanted to know if there is any fix for this or how I can train lgb model with categorical features having large number of categories.

Reproducible example

params = {
    'boosting_type': 'gbdt',
    'objective': 'binary',
    'metric': 'binary_logloss',
    'device': 'gpu',
    'max_bin': 200, 
    'num_leaves': 31
}

train_data = lgb.Dataset(X_train, label=y_train)
model = lgb.train(params, train_data)

Environment info

lgb version : 4.5.0
cuda version : 11.8
Python 3.7.13

Thanks in advance.

[crs1910]

Hi team, just a gentle reminder.Any update on this ?

[yuhorun]

@crs1910

lgbm version: 4.5.0.99

When using the GPU version of LightGBM, I encountered the error:
LightGBMError: bin size 257 cannot run on GPU.

After checking the issue, I found that a fix for this problem was merged 5 months ago in [#6019]

However, I am still encountering this issue.

Has anyone else encountered this problem? How should I resolve it?

[shiyu1994]

Thanks for reporting the issue. We will verify to see if it still exists in latest commit. We will report back here soon.