重庆在线观看,长沙网站seo按天计费,网站负责人不是法人,企业网站的重要性1.基础
loss function损失函数#xff1a;预测输出与实际输出 差距 越小越好
- 计算实际输出和目标之间的差距 - 为我们更新输出提供依据#xff08;反向传播#xff09;
1. L1 torch.nn.L1Loss(size_averageNone, reduceNone, reduction‘mean’) 2. 平方差#xff08;…1.基础
loss function损失函数预测输出与实际输出 差距 越小越好
- 计算实际输出和目标之间的差距 - 为我们更新输出提供依据反向传播
1. L1 torch.nn.L1Loss(size_averageNone, reduceNone, reduction‘mean’) 2. 平方差L2 torch.nn.MSELoss(size_averageNone, reduceNone, reduction‘mean’) 3. 交叉熵 torch.nn.CrossEntropyLoss(weightNone, size_averageNone, ignore_index-100, reduceNone, reduction‘mean’, label_smoothing0.0)
2. 例子
代码
import torch
from torch import nninput torch.tensor([1,2,3],dtypetorch.float32)
input torch.reshape(input,[1,1,1,3])
target torch.tensor([1,2,5],dtypetorch.float32)
target torch.reshape(target,[1,1,1,3])
# L1
l1 nn.L1Loss(reductionsum)
result1 l1(input,target)
print(result1)
# L2
l2 nn.MSELoss()
result2 l2(input,target)
print(result2)# 交叉熵损失
x torch.tensor([0.1,0.2,0.3])
y torch.tensor([1])
x torch.reshape(x,[1,3])
loss_cross nn.CrossEntropyLoss()
result loss_cross(x,y)
print(result)输出
