损失函数-交叉熵的原理及实现

咕咕咕~不催不更

什么是损失函数，损失函数的作用是？

有哪些常用的损失函数？数学表达式是什么？

实现

交叉熵的原理

实现

pytorch自带的交叉熵：

class CrossEntropyLoss(_WeightedLoss):
    __constants__ = ['ignore_index', 'reduction']
    ignore_index: int
 
    def __init__(self, weight: Optional[Tensor] = None, size_average=None, ignore_index: int = -100,
                 reduce=None, reduction: str = 'mean') -> None:
        super(CrossEntropyLoss, self).__init__(weight, size_average, reduce, reduction)
        self.ignore_index = ignore_index
 
    def forward(self, input: Tensor, target: Tensor) -> Tensor:
        assert self.weight is None or isinstance(self.weight, Tensor)
        return F.cross_entropy(input, target, weight=self.weight,
                               ignore_index=self.ignore_index, reduction=self.re