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

about the dataset #2

Open
HalbertCH opened this issue Oct 3, 2019 · 2 comments
Open

about the dataset #2

HalbertCH opened this issue Oct 3, 2019 · 2 comments

Comments

@HalbertCH
Copy link

Hello, I saw you used the "imagenet64.mat" file as your dataset, and I think the file is much more convenient to use than those original imagenet images. Now I want to train my model one the 128*128 imagenet dataset, do you know if there exists an "imagenet128.mat" file? If exists, do you know where to download the file? Thank you very much!

@MingtaoGuo
Copy link
Owner

If you want, i will help you to generate imagenet128.mat, but it will cost more memory than imagenet64.mat.
My E-mail: [email protected]

@HalbertCH
Copy link
Author

I don't need it anymore, but thank you all the same!Besides, I have another question, that is, I saw you used the learning rate of 1e-4 in D and 4e-4 in G in your code, while in the papers of BigGAN and SAGAN the learning rate of D is larger, which is opposite to your setting. May I ask why you set the learning rate in such a differnt way?

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