python - How to manually assign hyperparameter to LGBM

Question

model = lightgbm.LGBMClassifier()  
hyperparameter_dictionary = {'boosting_type': 'goss',   'num_leaves': 25, 'n_estimators': 184, ...}  

How do I assign the model's hyperparameters by the dictionary, so that I can try out the model's behavior with such hyperparameters?

Thanks!

标签： pythonmachine-learningscikit-learnhyperparameterslightgbm

Answer 1

将超参数字典传递给模型构造函数，添加**到 dict 以传递每个 dict 项，如 lgbm 期望的 kwarg 参数https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.LGBMClassifier.html#lightgbm .LGBM分类器：

hyperparameter_dictionary = {'boosting_type': 'goss', 'num_leaves': 25, 'n_estimators': 184}
model = lightgbm.LGBMClassifier(**hyperparameter_dictionary)

测试：

print(model)

LGBMClassifier(boosting_type='goss', ... n_estimators=184, n_jobs=-1, num_leaves=25,...)