Unexpected keyword argument when running Lightgbm on GPU

Question

When running the code below:

import lightgbm as lgb

params = {'num_leaves': 38,              
          'min_data_in_leaf': 50,
          'objective': 'regression',     
          'max_depth': -1,                
          'learning_rate': 0.1,           
          'device': 'gpu'
          }
trn_data = lgb.Dataset(x_train, y_train)
val_data = lgb.Dataset(x_test, y_test)

model = lgb.train(params,
                trn_data,
                20000,
                valid_sets=[trn_data, val_data],
                verbose_eval=300,
                early_stopping_rounds=1000)

I get the follow errors:

train() got an unexpected keyword argument 'verbose_eval'
train() got an unexpected keyword argument 'early_stopping_rounds'

It is important to note that I run this on GPU. When running it on CPU i do not get this error.

Has anyone got an idea how I can incorporate a verbose output and early stopping rounds when running Lightgbm on GPU??

score 1 · Accepted Answer · answered Aug 13 '22 at 17:39

1

For Lightgbm on GPU you can check the official documentation. On documentation there are no configuration option as vebose_ecal and early_stopping_rounds Official Documentation

Also you can check this link Running LightGBM on GPU with python

answered Aug 13 '22 at 17:39

Tunahan Akdogan

246
2
9

1

do you have an idea how i can display the output whilst running lightgbm on gpu using python? – freak11 Aug 14 '22 at 18:40
1

Unfortunately I don't have an idea. You can ask a new question with your intention and example scenarios. – Tunahan Akdogan Aug 14 '22 at 20:02

Unexpected keyword argument when running Lightgbm on GPU

1 Answers1