下表概述了MLlib支持的損失函數及其梯度和子梯度:
損失函數$$L(w;x,y)$$ | 梯度或子梯度 | |
---|---|---|
hinge loss | $$ \left{ \begin{array}{ll} -yx ~~~ 若~yw^Tx \lt1, \ 0 ~~~~~~~~~otherwise \end{array} \right.$$ | |
logistic loss | ||
squared loss |
下表概述了MLlib支持的損失函數及其梯度和子梯度:
損失函數$$L(w;x,y)$$ | 梯度或子梯度 | |
---|---|---|
hinge loss | $$ \left{ \begin{array}{ll} -yx ~~~ 若~yw^Tx \lt1, \ 0 ~~~~~~~~~otherwise \end{array} \right.$$ | |
logistic loss | ||
squared loss |