Dive-into-DL-PyTorch icon indicating copy to clipboard operation
Dive-into-DL-PyTorch copied to clipboard

Published 20 hours ago • verified publisher icon ShusenTang

3.3 节最后的损失函数需要sum后才可以调用backward函数

Open kevindragon opened this issue 4 years ago • 1 comments

3.3.7 训练模型：损失函数需要sum

kevindragon avatar Jan 02 '20 09:01 kevindragon

MSELoss默认会求平均，所以得到的loss已经是个标量。

ShusenTang avatar Jan 02 '20 09:01 ShusenTang