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

Training time #9

Open
sstzal opened this issue Sep 23, 2021 · 3 comments
Open

Training time #9

sstzal opened this issue Sep 23, 2021 · 3 comments

Comments

@sstzal
Copy link

sstzal commented Sep 23, 2021

Hi, Alex
Can you show me how much time is needed for training on the ShapeNetv2 and other datasets?

@alextrevithick
Copy link
Owner

I find that about 400000 iterations is sufficient for shapenet datasets, and 250000 for LLFF and Blender.

@sstzal
Copy link
Author

sstzal commented Oct 4, 2021

Hi, Alex
Sorry to ask one more question (lazy). I want to know how much GPU memory (how much G?) is used when training this model?

@alextrevithick
Copy link
Owner

Feel free to ask as many as you need. With memory, I also load all of the shapenet dataset into memory, so you need a good amount of RAM as well, 32gb should be sufficient.

With regard to VRAM, Shapenet usually uses around 12-13gb during training, and I maxed out a 32gb v100 for the other datasets. You can adjust the batch size (N_rand) in the configs to make it work with less RAM, as long as its not too low.

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