How to forward some fixed arguments to LGBMClassifier?

[leonleeann]

I'm using verstack.LGBMTuner to optimize params of a lightgbm.LGBMClassifier, that should resolve a multiclass classification problem.

My trouble is that the dataset is heavily imbalanced, so I need pass an argument class_weight='balanced' to LGBMClassifier's constructor.

I read the docs of LGBMTuner, but I didn't find any hint about how to forwarding parameters to a potential estimator.

Is there a way I can do it?

[DanilZherebtsov]

Hi @leonleeann

Thank you for your comment. I've had this feature request quite a few times by now and just shipped the update.

pip install --upgrade verstack

You should get a verstack.version == 3.8.0 and LGBMTuner.version == 1.1.0

Now you can pass any supported LGBM parameters at init as follows:

from verstack import LGBMTuner

my_custom_params = {'class_weight' : 'balanced'}

tuner = LGBMTuner(metric = 'accuracy', custom_lgbm_params = my_custom_params)

It is all reflected in the documentation.

[leonleeann]

Thank you @DanilZherebtsov !
Sorry, maybe I refered an outdated doc.

[DanilZherebtsov]

Thank you @DanilZherebtsov !
Sorry, maybe I refered an outdated doc.

No, you were right. I just shipped the update today. Let me know if it works.