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

Autograd-Issue for Training Gaussian WaveNet #213

Open
lambda-delta34 opened this issue Jun 15, 2021 · 0 comments
Open

Autograd-Issue for Training Gaussian WaveNet #213

lambda-delta34 opened this issue Jun 15, 2021 · 0 comments

Comments

@lambda-delta34
Copy link

lambda-delta34 commented Jun 15, 2021

I am using PyTorch 1.9.0.dev20210512+cu111 version to train the Gaussian example. Without any modification but changing demo.json to the gaussian.json (also project name as well), I received the warning Using a non-full backward hook when the forward contains multiple autograd Nodes

Screen Shot 2021-06-15 at 2 27 15 PM

Just wondering does this warning have any impact on the trained model?

Overall, really appreciate for the hard work putting on this project. Even at 5 hrs training audio I can get something reasonably satisfied. Great Thanks!

@lambda-delta34 lambda-delta34 changed the title Autograd-Issue for PyTorch Autograd-Issue for Training Gaussian WaveNet Jun 15, 2021
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

1 participant