mseloss 与 smooth_l1_loss比较

在0-1直接，mseloss的平方是smooth_l1_loss，smooth_l1_loss更小

 

import torch


import torch.nn.functional as F

conf_mask = torch.FloatTensor([3])
conf_data = torch.FloatTensor([1])

loss_fn = torch.nn.MSELoss( reduce=False, size_average=False)

x= loss_fn(conf_mask, conf_data).item()
print('-----',x)


loc_loss = F.smooth_l1_loss(conf_mask, conf_data)
print(loc_loss)

结果：

----- 4.0
tensor(1.5000)

 

 

测试代码：

import torch
import torch.nn.functional as F

conf_mask = torch.FloatTensor([0.0, 10.0, 0.0, 1.0, 1.0])
conf_data = torch.FloatTensor([10.1, 0.9, 0.0, 10.2, 10.2])

loss_fn = torch.nn.MSELoss()# reduce=False, size_average=False)

x= loss_fn(conf_mask, conf_data).item()
print('-----0&1',x)

loc_loss = F.smooth_l1_loss(conf_mask, conf_data)
print(loc_los