LGB 的 .feature_importance() 函数

对于机器学习训练来说，每一种特征对于机器学习训练的模型的重要性可能是不一样的，此时，这个函数就起到了作用。

下面是文档

---

feature_importance(importance_type='split', iteration=None)

• Get feature importances.
  Parameters:

  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).

下面是一个py文件中的代码样例

---

for i, (train_ind, val_ind) in enumerate(kf.split(train)):  # i 返回的是第几折， train_ind 是训练集， val_ind 是测试集
    
    print(f'Beginning fold {i}')
    
    train_data = lgb.Dataset(X.iloc[train_ind],
                           label=y.iloc[train_ind])

    val_data = lgb.Dataset(X.iloc[val_ind],
                           label=y.iloc[val_ind])


    model = lgb.train(params,
                      train_data,
                      num_boost_round=10000,
                      valid_sets = [train_data, val_data],
                      verbose_eval=100,
                      early_stopping_rounds = 100)
    
    #Save both split and gain importances, averaged over 5 folds
     # 统计某种特征在该机器学习算法模型中使用过的次数(这里求的是5次训练的平均值)
    feats['importance_split'] += model.feature_importance()/5
    # 统计某种特征在整个.py文件中使用的次数(这里求的是5次训练的平均值)
    feats['importance_gain'] += model.feature_importance(importance_type='gain')/5