Kaixuan Wei

Well it looks like a computing arch issue that is tricky to be solved. Please open a new issue in [https://github.com/matteo-ronchetti/torch-radon](url) and ask the author to see whether he can...

have you ever tried manually tweak the hyperparameters of the Plug-and-play solver to see its performance instead of using the RL policy?

well, it might be a case, PnP algorithm with a generic denoiser trained on natural datasets doesn't work well in your CT dataset. In this situation, probably a training-based approach...

the [link](https://icvl.cs.bgu.ac.il/hyperspectral/) is accessible for me

yes you only need to center crop

yes you are right

hi I checked the OneDrive link. It perfectly works. Perhaps you'd better to have a VPN to access to OneDrive

Hi, the issue described above is due to the IQA model (in this repo) can only assess quality degradations arising from the distortion types (i.e. super-resolved artifacts) that it has...

emmm, the issue might be caused by the official version of PyBM3D. But since I cannot install your custom version of PyBM3D, I'm not sure whether there is a problem....