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

Could you provide more information about different training settings, like how many days are needed for training when 4 gpus are used, when 2 gpus are used, when 1 gpus is used?, #5

Open
hdjsjyl opened this issue Oct 23, 2020 · 2 comments

Comments

@hdjsjyl
Copy link

hdjsjyl commented Oct 23, 2020

Thanks for your excellent work! As the question, could you provide the training time for different settings of gpus, input resolutions, the number of training images? And advice would be appreciated, I am looking forward to your reply. Thank you!

@LiUzHiAn
Copy link

LiUzHiAn commented Nov 30, 2021

Also want to know the GPU Memory usage under different batch_size settings. Thanks.

@icaoyu
Copy link

icaoyu commented Jul 18, 2022

I trained the encoder on ffhq256 for more than one week with a single GPU, the snapshot is present poor, is it normal? below is the current information.
[2022-07-18 11:14:46][INFO] Iter 357650/875000, data time: 0.000s, iter time: 0.995s, run time: 7d07h, lr (discriminator): 1.255e-06, lr (encoder): 1.255e-06, recon_loss: 0.365, adv_loss: 0.107, loss_fake: 0.342, loss_real: 0.357, real_grad_penalty: 0.036 (memory: 17.6G) (ETA: 5d22h)

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

3 participants