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

seems the loss of both generator and discriminator would collapse? #8

Open
sszzsupersupersupersuper opened this issue Oct 23, 2019 · 1 comment

Comments

@sszzsupersupersupersuper
Copy link

sszzsupersupersupersuper commented Oct 23, 2019

Hello, I am running your code with SWDA dataset but the loss going like this:

train_loss_AE:2.7721 train_loss_G:298.7391 train_loss_D:-300.4698
DialogWAE_GMP-basic|SWDA@gpu0 epo:[84/100] iter:[1200/1279] step_time:56s elapsed:0:11:33<0:0:46

train_loss_AE:3.0337 train_loss_G:203.4530 train_loss_D:-84.7853

Valid begins with 41 batches with 28 left over samples
Validation valid_loss_AE:3.0522 valid_loss_G:353.3996 valid_loss_D:-353.3996
Valid begins with 5165 batches with 0 left over samples

I did not change any configuration, but I haven't checked the generated data yet. The loss
is just supposed to be like this? or it has already collapsed?

Thank you for the help!

@guxd
Copy link
Owner

guxd commented Oct 23, 2019

The loss should be normal since it could contain negative values

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

2 participants