Sklearn.metrics

分类指标

1. accuracy_score(y_true, y_pre)
   精度
2. auc(x, y, reorder=False)
   ROC曲线下的面积;较大的AUC代表了较好的performance。
3. average_precision_score(y_true, y_score, average=‘macro’, sample_weight=None)
   根据预测得分计算平均精度
4. brier_score_loss(y_true, y_prob, sample_weight=None, pos_label=None)
   The smaller the Brier score, the better.
5. confusion_matrix(y_true, y_pred, labels=None, sample_weight=None)
   通过计算混淆矩阵来评估分类的准确性 返回混淆矩阵
6. f1_score(y_true, y_pred, labels=None, pos_label=1, average=‘binary’, sample_weight=None)
   F1值
7. log_loss(y_true, y_pred, eps=1e-15, normalize=True, sample_weight=None, labels=None)
   对数损耗，又称逻辑损耗或交叉熵损耗
8. precision_score(y_true, y_pred, labels=None, pos_label=1, average=‘binary’,)
   查准率
9. recall_score(y_true, y_pred, labels=None, pos_label=1, average=‘binary’, sample_weight=None)
   查全率
10. roc_auc_score(y_true, y_score, average=‘macro’, sample_weight=None)
    计算ROC曲线下的面积就是AUC的值，the larger the better
11. roc_curve(y_true, y_score, pos_label=None, sample_weight=None, drop_intermediate=True)
    计算ROC曲线的横纵坐标值，TPR，FPR
    TPR = TP/(TP+FN) = recall(真正例率，敏感度)
    FPR = FP/(FP+TN)(假正例率，1-特异性)
12. classification_report(y_true, y_pred)
    分类结果分析汇总

回归指标

1. explained_variance_score(y_true, y_pred, sample_weight=None, multioutput=‘uniform_average’)
   回归方差(反应自变量与因变量之间的相关程度)
2. mean_absolute_error(y_true, y_pred, sample_weight=None, multioutput=‘uniform_average’)
   平均绝对误差
3. mean_squared_error(y_true, y_pred, sample_weight=None, multioutput=‘uniform_average’)
   均方差
4. median_absolute_error(y_true, y_pred)
   中值绝对误差
5. r2_score(y_true, y_pred, sample_weight=None, multioutput=‘uniform_average’)
   R平方值