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 to adjust various parameters #1063

Open
jayjungit opened this issue Nov 14, 2024 · 4 comments
Open

How to adjust various parameters #1063

jayjungit opened this issue Nov 14, 2024 · 4 comments

Comments

@jayjungit
Copy link

When training more than 3,000 images, how should each parameter be adjusted to make the effect better?

@jaco001
Copy link

jaco001 commented Nov 14, 2024

For 50 pic (paper) denoiser start 500 end at 15000, training end at 30000
so...with x60 scaling: start 30 000 end 900 000, training end like 1 800 000
Possible? ;)

@AsherJingkongChen
Copy link

For 50 pic (paper) denoiser start 500 end at 15000, training end at 30000 so...with x60 scaling: start 30 000 end 900 000, training end like 1 800 000 Possible? ;)

Or just increase learning rates

@jayjungit
Copy link
Author

jayjungit commented Nov 15, 2024

对于 50 张图片(论文)去噪器,开始 500 结束于 15000,训练结束于 30000 所以...使用 x60 缩放:开始 30 000 结束 900 000,训练结束如 1 800 000 可能吗?;)

Do other parameters need to be adjusted, except the number of steps? ,"denoiser" is “self.densify_until_iter”?

@jaco001
Copy link

jaco001 commented Nov 15, 2024

More than parameters for 'quality' do pic selection. GS literally put everything that is on the pic. Bad pic - bad effect.

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