Press "Enter" to skip to content

使用feature Importance进行特征选择

在前一篇机器学习之特征选择的文章中讲到了树模型中GBDT也可用来作为基模型进行特征选择。今天在此基础上进行拓展,介绍除决策树外用的比较多的XGBoost、LightGBM。

 

DecisionTree

 

决策树的feature_importances_属性,返回的重要性是按照决策树种被用来分割后带来的增益(gain)总和进行返回。

 

The importance of a feature is computed as the (normalized) total reduction of the criterion brought by that feature. It is also known as the Gini importance.

 

关于信息增益(Gain)相关介绍可以决策树简介。

 

参考链接: https://scikit-learn.org/stable/modules/generated/sklearn.tree.DecisionTreeClassifier.html#sklearn.tree.DecisionTreeClassifier.feature_importances_

 

GradientBoosting和ExtraTrees与DecisionTree类似。

 

XGBoost

 

get_score(fmap='', importance_type='weight')
Get feature importance of each feature. Importance type can be defined as:
‘weight’: the number of times a feature is used to split the data across all trees.
‘gain’: the average gain across all splits the feature is used in.
‘cover’: the average coverage across all splits the feature is used in.
‘total_gain’: the total gain across all splits the feature is used in.
‘total_cover’: the total coverage across all splits the feature is used in.

 

其中:

weight:该特征被选为分裂特征的次数。
gain:该特征的带来平均增益(有多棵树)。在tree中用到时的gain之和/在tree中用到的次数计数。gain = total_gain / weight
cover:该特征对每棵树的覆盖率。
total_gain:在所有树中,某特征在每次分裂节点时带来的总增益
total_cover:在所有树中,某特征在每次分裂节点时处理(覆盖)的所有样例的数量。

参考链接: https://xgboost.readthedocs.io/en/latest/python/python_api.html#xgboost.Booster.get_score

 

LightGBM

 

feature_importance(importance_type='split', iteration=None)
Get feature importances.
 importance_type (string, optional (default="split")) – How the importance is calculated. If “split”, result contains numbers of times the feature is used in a model. If “gain”, result contains total gains of splits which use the feature.
 iteration (int or None, optional (default=None)) – Limit number of iterations in the feature importance calculation. If None, if the best iteration exists, it is used; otherwise, all trees are used. If <= 0, all trees are used (no limits).

 

其中:

split就是特征在所有决策树中被用来分割的总次数。
gain就是特征在所有决策树种被用来分割后带来的增益(gain)总和

参考链接: https://lightgbm.readthedocs.io/en/latest/pythonapi/lightgbm.Booster.html#lightgbm.Booster.feature_importance

Be First to Comment

发表回复

您的电子邮箱地址不会被公开。 必填项已用*标注