在cross_validate或cross_val_score中，参数scoring，与分类、聚类和回归算法的评价指标有关。
3.4.3. The scoring parameter: defining model evaluation rules
For the most common use cases, you can designate a scorer object with the scoring parameter via a string name; the table below shows all possible values. All scorer objects follow the convention that higher return values are better than lower return values. Thus metrics which measure the distance between the model and the data, like metrics.mean_squared_error, are available as 'neg_mean_squared_error' which return the negated value of the metric
对于最常见的用例，您可以通过字符串名称使用 scoring 参数指定一个评分对象；下表显示了所有可能的值。所有评分对象都遵循这样的约定：返回值越高越好。因此，像 metrics.mean_squared_error 这样衡量模型与数据之间距离的指标，会以 'neg_mean_squared_error' 的形式提供，返回该指标的负值。
1、分类
| 字符串 | 函数 | 公式 |
|---|---|---|
accuracy |
metrics.accuracy_score | a c c u r a c y ( y , y ^ ) = 1 n ∑ i = 0 n − 1 1 ( y ^ i = y i ) accuracy(y,\hat{y}) = \frac{1}{n}\sum_{i=0}^{n-1}1(\hat{y}_i=y_i) accuracy(y,y^)=n1∑i=0n−11(y^i=yi) |
balanced_accuracy |
metrics.balanced_accuracy_score | b a l a n c e d − a c c u r a c y = 1 2 ( T P T P + F N + T N T N + F P ) balanced-accuracy=\frac{1}{2}(\frac{TP}{TP+FN}+\frac{TN}{TN+FP}) balanced−accuracy=21(TP+FNTP+TN+FPTN) |
top_k_accuracy |
metrics.top_k_accuracy_score | t o p − k a c c u r a c y ( y , y ^ ) = 1 n ∑ i = 0 n − 1 ∑ j = 1 k 1 ( f ^ i , j = y i ) top-k\ \ accuracy(y,\hat{y}) = \frac{1}{n}\sum_{i=0}^{n-1}\sum_{j=1}^{k}1(\hat{f}_{i,j}=y_i) top−k accuracy(y,y^)=n1∑i=0n−1∑j=1k1(f^i,j=yi) |
average_precision |
metrics.average_precision_score | A P = ∑ n ( R n − R n − 1 ) P n AP = \sum_{n}(R_n-R_{n-1})P_n AP=∑n(Rn−Rn−1)Pn |
neg_brier_score |
metrics.brier_score_loss | B S = 1 n ∑ i = 0 n − 1 ( y i − p i ) 2 = 1 n ∑ i = 0 n − 1 ( y i − p r e d i c t _ p r o b a ( y = 1 ) ) 2 BS= \frac{1}{n}\sum_{i=0}^{n-1}(y_i-p_i)^2=\frac{1}{n}\sum_{i=0}^{n-1}(y_i-predict\_{proba}(y=1))^2 BS=n1∑i=0n−1(yi−pi)2=n1∑i=0n−1(yi−predict_proba(y=1))2 |
f1 |
metrics.f1_score | F 1 = 2 ∗ T P 2 ∗ T P + F P + F N F1=\frac{2*TP}{2*TP+FP+FN} F1=2∗TP+FP+FN2∗TP (average parameter) |
neg_log_loss |
metrics.log_loss | L l o g ( y , p ) = − l o g P r ( y ∣ p ) = − ( y l o g ( p ) + ( 1 − y ) l o g ( 1 − p ) ) L_{log}(y,p)=-logPr(y|p)=-(ylog(p)+(1-y)log(1-p)) Llog(y,p)=−logPr(y∣p)=−(ylog(p)+(1−y)log(1−p)) L l o g ( Y , P ) = − l o g P r ( Y ∣ P ) = − 1 N ∑ i = 0 N − 1 ∑ k = 0 K − 1 y i , k l o g p i , k L_{log}(Y,P)=-logPr(Y|P)=-\frac{1}{N}\sum_{i=0}^{N-1}\sum_{k=0}^{K-1}y_{i,k}logp_{i,k} Llog(Y,P)=−logPr(Y∣P)=−N1∑i=0N−1∑k=0K−1yi,klogpi,k |
precision |
metrics.precision_score | P = T P T P + F P P=\frac{TP}{TP+FP} P=TP+FPTP |
recall |
metrics.recall_score | R = T P T P + F N R=\frac{TP}{TP+FN} R=TP+FNTP |
jaccard |
metrics.jaccard_score | J ( y , y ^ ) = y ⋂ y ^ y ⋃ y ^ J(y,\hat{y})=\frac{y\bigcap\hat{y}}{y\bigcup\hat{y}} J(y,y^)=y⋃y^y⋂y^ |
roc_auc |
metrics.roc_auc_score | Compute Area Under the Receiver Operating Characteristic Curve (ROC AUC) from prediction scores |