如何从 XGBoost 分类器中的 pandas dataframe 中的 output 叶

Question

I have read in a csv file which contains 8 predictive features ( col_list ) and one target feature (Target variable is called " chd " -> 1= Heart Attack; 0 = No Heart Attack).

df = pd.read_csv(loc+'HeartDisease.csv', index_col=0)

Y = df['chd']
col_list = ['sbp','tobacco','ldl','adiposity','typea','obesity','alcohol','age'] 

I have trained an XGBoost Classifier:

# fit model no training data
model = XGBClassifier(
    base_score=0.1, 
    booster='gbtree', 
    colsample_bylevel=1,
    colsample_bynode=1, 
    colsample_bytree=0.6,
    enable_categorical=False, 
    gamma=0.1, 
    gpu_id=-1,
    importance_type=None, 
    interaction_constraints='',
    learning_rate=0.1, 
    max_delta_step=0,
    max_depth=8,
    min_child_weight=1, 
    monotone_constraints='(1,1,1,1,1,1,1,1)',#,"(1,-1)"
    n_estimators=4, n_jobs=1, 
    nthread=1, 
    num_parallel_tree=1,
    predictor='auto',
    random_state=0, 
    reg_alpha=0, 
    reg_lambda=1,
    scale_pos_weight=1, 
    silent=True, 
    subsample=0.6,
    tree_method='exact',
    validate_parameters=1, 
    verbosity=None)
    

I have then visualized the tree:

fig, ax = plt.subplots(figsize=(30, 30))
plot_tree(model,ax=ax)
plt.show()

How can I create a column called " leaf " in the df dataframe that contains the values of the terminal leaves shown in the picture above?

Answer 1

You can use xgboost.Booster 's method trees_to_dataframe :

df = model.Booster.trees_to_dataframe()