Installed and loaded XGBoost on Weka, but can't run on data as classifier

SigDelta / weka-xgboost

A Weka wrapper package for the XGBoost 4J

MIT License

9 stars 7 forks source link

Installed and loaded XGBoost on Weka, but can't run on data as classifier #4

Open huhrichard opened 4 years ago

huhrichard commented 4 years ago

Hi,

I am a Mac user using Weka 3.9.4. XGBoost is successfully loaded on Weka by package manager.

I can choose XGBoost as a classifier under 'trees'. However, once I start, nothing is shown, even with debug mode on.

The XGBoost on my computer is installed by pip, does it matter? How can I run XGBoost on Weka successfully?

Thanks.

jmSfernandes commented 2 years ago

Same happened to me. Weka 3.9.6... it just prints the building model information but nothing happens (see image)... @huhrichard I don't think you need to install anything through pip since this should be a wrapper for the XGBoost4J model. However, you can install the wekaPython package and then you can use the python implementation of XGBoost... And that seems to work.

Ushie-Sylvester commented 2 years ago

Hello @jmSfernandes i am having same issue as you, but i dont know how to use the Python implemetation package for XGboost, i have succesfully installed the Wekapython package but i dont know what to do from here. could you please guide me