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

How many data was need to train a realistic generator? #34

Open
zengjingming opened this issue Mar 14, 2023 · 2 comments
Open

How many data was need to train a realistic generator? #34

zengjingming opened this issue Mar 14, 2023 · 2 comments

Comments

@zengjingming
Copy link

zengjingming commented Mar 14, 2023

Hi,

I'm trying to train a GAN by myself. Could you please tell me how many images are needed to get a realistic generator?

@zengjingming
Copy link
Author

Also, I found that all of the links about VGGFace2 I can search from google are unavailable. Could you please where to find the VGGFace2 dataset?

@zengjingming zengjingming changed the title Docker train model How many data was need to train a realistic generator? Mar 16, 2023
@felixrosberg
Copy link
Owner

I suggest you look into academic torrents if the dataset has been removed from the main page. The dataset contains around 3.3 million facial image. I randomly sample target/source pairs during training by shuffling the file_path order along with a shuffle buffer. I deemed FaceDancer finished training after 300k iterations (using batch size 10).

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