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

The noise removal command eats up my entire RAM and then gets killed #146

Open
Sandjan opened this issue Jul 26, 2023 · 6 comments
Open

Comments

@Sandjan
Copy link

Sandjan commented Jul 26, 2023

I have 64GB RAM and 30GB swap but when I run the following command:
python main_test_swinir.py --task colour_dn --noise 15 --model_path model_zoo/swinir/005_colorDN_DFWB_s128w8_SwinIR-M_noise15.pth --folder_gt testsets/own

on a few pictures with 12megapixel first my ram is filled up, then the swap and then the program is killed by the system. I haven't tested what happens with smaller images, but I can't imagine it necessarily takes that much RAM.

@OmlineEditor
Copy link

add to the startup command: --tile 400

it will turn out like this: python main_test_swinir.py --task colour_dn --noise 15 --model_path model_zoo/swinir/005_colorDN_DFWB_s128w8_SwinIR-M_noise15.pth --folder_gt testsets/own --tile 400

tell me, did you manage to solve the problem later?

@guoshuyu123
Copy link

请问这个 tile参数是做什么用的 改变他会影响什么???

@OmlineEditor
Copy link

this option processes the image in chunks, small tiles and not all at once, it saves memory and there is no message about low RAM

@guoshuyu123
Copy link

感谢你的告知 我还想问下这个参数400具体指些什么?

@OmlineEditor
Copy link

I do not know exactly what this parameter does inside the program, I know that this parameter helps to process large images without reporting a lack of memory. you need to ask the developer in more detail, but he does not seem to get in touch and is silent.

@babyta
Copy link

babyta commented Jun 20, 2024

这个参数是用来裁剪数据的,不会让显存随着输入图像变大而增加

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

4 participants