mirror of
https://github.com/clearml/clearml-serving
synced 2025-01-31 02:46:54 +00:00
23 lines
597 B
Python
23 lines
597 B
Python
|
import lightgbm as lgb
|
||
|
|
from sklearn.datasets import load_iris
|
||
|
|
from sklearn.model_selection import train_test_split
|
||
|
|
|
||
|
|
from clearml import Task
|
||
|
|
|
||
|
|
task = Task.init(project_name="serving examples", task_name="train lightgbm model", output_uri=True)
|
||
|
|
|
||
|
|
iris = load_iris()
|
||
|
|
y = iris['target']
|
||
|
|
X = iris['data']
|
||
|
|
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1)
|
||
|
|
dtrain = lgb.Dataset(X_train, label=y_train)
|
||
|
|
|
||
|
|
params = {
|
||
|
|
'objective': 'multiclass',
|
||
|
|
'metric': 'softmax',
|
||
|
|
'num_class': 3
|
||
|
|
}
|
||
|
|
lgb_model = lgb.train(params=params, train_set=dtrain)
|
||
|
|
|
||
|
|
lgb_model.save_model("lgbm_model")
|