Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

关于训练和测试 #3

Open
yyl-bot opened this issue Feb 15, 2019 · 3 comments
Open

关于训练和测试 #3

yyl-bot opened this issue Feb 15, 2019 · 3 comments

Comments

@yyl-bot
Copy link

yyl-bot commented Feb 15, 2019

在RNN+ATTENTION的model中为什么直接用targets来作为Decoder的Input来训练啊？用每一个时间步（程序中的di）的hidden来作为下一步的input吗？
@airaria
Copy link
Owner

airaria commented Feb 17, 2019

decoder的每一步的input用的是上一步预测位置的真实标签(target)，

@yyl-bot yyl-bot closed this as completed Feb 18, 2019
@yyl-bot
Copy link
Author

yyl-bot commented Feb 18, 2019

训练的时候这样可以理解，但是为什么evaluate的时候第一步也要输入真实标签呢，测试的时候不是没有真实标签的吗？

@yyl-bot yyl-bot reopened this Feb 18, 2019
@airaria
Copy link
Owner

airaria commented Feb 18, 2019

evaluate的时候第一个输入taget[:,0]是一个表示开始的标志位呀，不是真实标签。就是说在数据预处理的时候已经在所有样本的第一位添加了一个特殊标识符，所以测试也是没问题的

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
2 participants
@airaria @yyl-bot