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.

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

为什么模型训练一段时间之后 损失值为NAN #1

Closed
wenshinlee opened this issue Aug 10, 2021 · 4 comments
Closed

为什么模型训练一段时间之后 损失值为NAN #1

wenshinlee opened this issue Aug 10, 2021 · 4 comments

Comments

@wenshinlee
Copy link

Iteration:6800, l1_loss:0.0195, time_taken:34.68
Iteration:6850, l1_loss:0.0186, time_taken:34.54
Iteration:6900, l1_loss:0.0211, time_taken:34.56
Iteration:6950, l1_loss:0.0197, time_taken:34.67
Iteration:7000, l1_loss:0.0199, time_taken:34.17
Iteration:7050, l1_loss:0.0182, time_taken:34.50
Iteration:7100, l1_loss:nan, time_taken:34.08
Iteration:7150, l1_loss:nan, time_taken:34.11
Iteration:7200, l1_loss:nan, time_taken:33.50
Iteration:7250, l1_loss:nan, time_taken:33.02
Iteration:7300, l1_loss:nan, time_taken:33.36

@wenshinlee
Copy link
Author

我是在Celeb上进行训练的,掩码是这样的,白色代表缺失
55115
训练的时候,什么也没有更改

@CyrilCsy
Copy link

你好,能问一下,这个问题解决了嘛

@Ellohiye
Copy link

你好,能问一下,这个问题解决了嘛

你好,你复现也遇到这个问题了吗?可以交流一下吗

@Sheeran2000
Copy link

是不是因为mask二进制的值不一样, 作者那个有效值是1,缺失值是0, 你这个是缺失值是1, 有效值是0

我是在Celeb上进行训练的,掩码是这样的,白色代表缺失 55115 训练的时候,什么也没有更改

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

5 participants