训练模型后检索 lightGBM 超参数-解网

问：

我使用 optuna 为数据集创建了数百个模型。很少有有用的模型，我不小心没有为好的模型保存超参数。

我确实将模型保存为 .txt。有没有办法从已经训练好的模型中提取超参数？

我问chatGPT，它提到

import lightgbm as lgb

# Load the model
gbm = lgb.Booster(model_file='path_to_your_model.txt')  # Replace with the actual path to your model file

# Get hyperparameters
hyperparameters = gbm.get_params()

# Print hyperparameters
for key, value in hyperparameters.items():
    print(f"{key}: {value}")

但是我收到此错误：

AttributeError: 'Booster' object has no attribute 'get_params'

Python 机器学习 LightGBM

import lightgbm as lgb
from sklearn.datasets import make_regression

# train a model
X, y = make_regression(n_samples=10_000)
dtrain = lgb.Dataset(X, label=y)
model = lgb.train(
    params={
        "learning_rate": 0.0708,
        "num_leaves": 11,
        "objective": "regression"
    },
    train_set=dtrain,
    num_boost_round=6
)

# save model to disk
model.save_model("model.txt")

# create a Booster
model_from_file = lgb.Booster(model_file="model.txt")
print(model_file.params)

您将在那里看到所有参数：

{
   "objective": "regression",
   ...
   "learning_rate": 0.0708,
   "num_leaves": 11,
   ...
}

上一个：在 R 中保存并加载 lightGBM 模型

下一个：LightGBM 中的early_stopping_rounds

训练模型后检索 lightGBM 超参数

Retrieving lightGBM hyper parameters after a model has been trained

评论