xy-boy / ttst Goto Github PK

Thank you very much for your excellent work. But here, I have a question I want to consult, about the inference stage of the matrics calculation I have not seen in the existing code, may I ask which code is used to achieve it? If you are free, please reply, thank you!

sorry

Hi, there：
Can I ask you about how you deal with the AID datasets from changed it from 600px to 512px, and how to make its LR data?

Hi,

Thank you for your code! I have been very interested in one aspect in particular. Many papers' code implementations (including SwinIR, HAT, DAT, and more recently, SPAN) make use of the rgb_mean code that is commented out here.
I have noticed that doing the same with the SPAN code implementation code greatly increases training stability, but breaks compatibility due to its interaction with the network features.

I was wondering if you could explain why you've chosen not to include it in your implementation, and what pros and cons you know of when using it?

Kind regards,
terrainer

Hi,

Thank you so much for your work!

I have just tested your code with an image of 510x339 pixel but I got an error

x = x.view(b, h // window_size, window_size, w // window_size, window_size, c)
RuntimeError: shape '[1, 42, 8, 63, 8, 1]' is invalid for input of size 172890

Therefore, I cropped that image to 128x128 pixel and it works as normal. The reason to choose the size of 128x128 comes from your example image "airport_285.png" in your source code.

Could you please check and fix this problem?

Thank you for your help!

Hi:
this release code can not use MSF module, so how can i add this module to code?

where can I get this folder?