XGBOOST查看特征分数

    #y = bst.predict(sub_trainning_data)

    feature_score = bst.get_fscore()  
    feature_score = sorted(feature_score.items(), key=lambda x:x[1],reverse = True)  
    fs = []
    for (key,value) in feature_score:
         fs.append("{0},{1}\n".format(key,value))
    with open('../sub/submission.csv','w') as f:
         f.writelines("feature,fscore\n")
         f.writelines(fs)

    df = pd.DataFrame(feature_score , columns=['feature', 'fscore'])  
    df['fscore'] = df['fscore'] / df['fscore'].sum()  
    featp = df.plot(kind='barh', x='feature', y='fscore', legend=False, figsize=(6, 10))  
    plt.title('XGBoost Feature Importance')  
    plt.xlabel('relative importance')  

推荐阅读更多精彩内容