GBDT分类模型的主要参数，训练用时，数据切分，数据洗牌，交叉验证，节点分裂评价准则，学习其数量，无放回采样比例，划分时考虑的最大特征比例，英文版

The main parameters of a GBDT classification model include:

- Number of trees: This refers to the number of decision trees that will be generated by the model.
- Learning rate: This controls the step size at each iteration while moving toward a minimum of a loss function.
- Max depth: This sets the maximum depth of each decision tree.
- Max features: This is the maximum number of features each tree is allowed to use when splitting a node.
- Min samples split: This is the minimum number of samples required to split an internal node.
- Min samples leaf: This is the minimum number of samples required to be at a leaf node.
- Subsample: This is the fraction of samples to be used for fitting the individual base learners.
- Loss function: This is the evaluation metric used to measure the performance of the model.

Other important considerations for training a GBDT model include:

- Data splitting: This involves dividing the dataset into training and testing sets.
- Data shuffling: This involves randomly shuffling the order of the training data to prevent the model from learning any sequence patterns.
- Cross-validation: This involves splitting the data into k-folds and training the model on k-1 folds while using the remaining fold for testing.
- Evaluation metric: This is used to measure the performance of the model during training and testing. Common examples include accuracy, precision, recall, and F1 score.

英文版：

GBDT分类模型的主要参数包括：

- 树的数量：指模型生成的决策树的数量。
- 学习率：控制每次迭代时向损失函数最小值移动的步长。
- 最大深度：设置每个决策树的最大深度。
- 最大特征数：每个树在分裂节点时允许使用的最大特征数。
- 最小样本分裂数：分裂内部节点所需的最小样本数。
- 最小样本叶节点数：叶节点所需的最小样本数。
- 子采样：用于拟合单个基学习器的样本比例。
- 损失函数：用于衡量模型性能的评估指标。

其他训练GBDT模型的重要考虑因素包括：

- 数据切分：将数据集分为训练集和测试集。
- 数据洗牌：随机打乱训练数据的顺序，以防止模型学习任何序列模式。
- 交叉验证：将数据集分成k个折，使用k-1个折训练模型，使用剩余的折进行测试。
- 评估指标：用于在训练和测试过程中衡量模型性能的指标。常见的例子包括准确率、精度、召回率和F1分数。