欢迎您访问程序员文章站本站旨在为大家提供分享程序员计算机编程知识!
您现在的位置是: 首页  >  IT编程

【机器学习要点记录】实用代码

程序员文章站 2022-06-15 19:13:23
总是想不起来怎么查找,这里记录一下sklearn.metrics.SCORERS.keys()dict_keys(['explained_variance', 'r2', 'max_error', 'neg_median_absolute_error', 'neg_mean_absolute_error', 'neg_mean_squared_error', 'neg_mean_squared_log_error', 'neg_root_mean_squared_error', 'neg_mea...

1、cross_val_score 能用的分数指标

sklearn.metrics.SCORERS.keys()
dict_keys(['explained_variance', 'r2', 'max_error', 'neg_median_absolute_error', 'neg_mean_absolute_error', 'neg_mean_squared_error', 'neg_mean_squared_log_error', 'neg_root_mean_squared_error', 'neg_mean_poisson_deviance', 'neg_mean_gamma_deviance', 'accuracy', 'roc_auc', 'roc_auc_ovr', 'roc_auc_ovo', 'roc_auc_ovr_weighted', 'roc_auc_ovo_weighted', 'balanced_accuracy', 'average_precision', 'neg_log_loss', 'neg_brier_score', 'adjusted_rand_score', 'homogeneity_score', 'completeness_score', 'v_measure_score', 'mutual_info_score', 'adjusted_mutual_info_score', 'normalized_mutual_info_score', 'fowlkes_mallows_score', 'precision', 'precision_macro', 'precision_micro', 'precision_samples', 'precision_weighted', 'recall', 'recall_macro', 'recall_micro', 'recall_samples', 'recall_weighted', 'f1', 'f1_macro', 'f1_micro', 'f1_samples', 'f1_weighted', 'jaccard', 'jaccard_macro', 'jaccard_micro', 'jaccard_samples', 'jaccard_weighted'])

2、目录下文件

root = os.getcwd()
files_list = []
for root,dirs,files in os.walk(root):
    for file in files:
        if '.csv' in file:
            files_list.append(file)
files_list

【机器学习要点记录】实用代码

3、上采样

from imblearn.over_sampling import SMOTE
sm = SMOTE(random_state=42)
X_train_bl, y_train_bl = sm.fit_sample(X_train, y_train)

4、分类报告

from sklearn.metrics import classification_report
print("\n分类报告:\n{0}".format(classification_report(y_test, y_predict_gbd, target_names=["0", "1"])))

【机器学习要点记录】实用代码

5、方差过滤

X = X[(X.var() > 0).index]

6、互信息法

from sklearn.feature_selection import mutual_info_classif as MIC

X_mic = X.loc[:, MIC(X_var, y, random_state=42) > 0]

ps:一定要固定随机种子,否则每次运行结果都会不一样

本文地址:https://blog.csdn.net/lvhuike/article/details/109045995

相关标签: 机器学习 python