loss函数之MarginRankingLoss

2021-07-17 本文已影响0人 ltochange

MarginRankingLoss

排序损失函数

对于包含 $N$ 个样本的batch数据 $D(x1,x2,y)$ , $x1$ , $x2$ 是给定的待排序的两个输入， $y$ 代表真实的标签，属于 $\{1,-1\}$ 。当 $y=1$ 是， $x1$ 应该排在 $x2$ 之前， $y=-1$ 是， $x1$ 应该排在 $x2$ 之后。第 $n$ 个样本对应的 $loss$ 计算如下:

$l_{n}=\max (0,-y *(x 1-x 2)+\operatorname{margin})$

若 $x1$ , $x2$ 排序正确且 $-y *(x 1-x 2)>\operatorname{margin}$ , 则loss为0；其他情况下是loss为 $-y *(x 1-x 2)+\operatorname{margin}$

class MarginRankingLoss(_Loss):
    __constants__ = ['margin', 'reduction']
    def __init__(self, margin=0., size_average=None, reduce=None, reduction='mean'):
        super(MarginRankingLoss, self).__init__(size_average, reduce, reduction)
        self.margin = margin
    def forward(self, input1, input2, target):
        return F.margin_ranking_loss(input1, input2, target, margin=self.margin, reduction=self.reduction)

pytorch中通过torch.nn.MarginRankingLoss类实现，也可以直接调用F.margin_ranking_loss 函数，代码中的size_average与reduce已经弃用。reduction有三种取值mean, sum, none，对应不同的返回 $\ell(x, y)$ 。默认为mean，对应于上述 $loss$ 的计算

$L=\left\{l_{1}, \ldots, l_{N}\right\}$

$\ell(x, y)=\left\{\begin{array}{ll}\operatorname L, & \text { if reduction }=\text { 'none' } \\ \frac1{N}\sum_{i=1}^{N} l_{i}, & \text { if reduction }=\text { 'mean' } \\ \sum_{i=1}^{N} l_{i} & \text { if reduction }=\text { 'sum' }\end{array} \right.$

$margin$ 默认取值0

例子：

import torch
import torch.nn.functional as F
import torch.nn as nn
import math


def validate_MarginRankingLoss(input1, input2, target, margin):
    val = 0
    for x1, x2, y in zip(input1, input2, target):
        loss_val = max(0, -y * (x1 - x2) + margin)
        val += loss_val
    return val / input1.nelement()


torch.manual_seed(10)
margin = 0
loss = nn.MarginRankingLoss()
input1 = torch.randn([3], requires_grad=True)
input2 = torch.randn([3], requires_grad=True)
target = torch.tensor([1, -1, -1])
print(target)
output = loss(input1, input2, target)
print(output.item())

output = validate_MarginRankingLoss(input1, input2, target, margin)
print(output.item())

loss = nn.MarginRankingLoss(reduction="none")
output = loss(input1, input2, target)
print(output)

输出：

tensor([ 1, -1, -1])
0.015400052070617676
0.015400052070617676
tensor([0.0000, 0.0000, 0.0462], grad_fn=<ClampMinBackward>)

loss函数之MarginRankingLoss

MarginRankingLoss

猜你喜欢

热点阅读