Seyyed Hossein Hasanpour

@rwightman out of the last three conv layers, two (2048 and 256) have kernel size=1 and they use padding of 1, that causes the featuremap size to increase from previously...

@rwightman May I ask if your vacation is over and if we can hopefully get this last step worked out?

@rwightman Thanks, I really appreciate it knowing how busy your schedule is. its not really any different than using (zero)-padding on the input. This happened by accident, but after I...

@rwightman That would be great, thanks a lot :) I was actually planning to test them on detection/segmentation as well, to see how that affects the result in different scenarios....

OK, I pushed the changes, hope I didnt do anything wrong.

@rwightman Hi, hope you are doing great, I finally finished training the new weights and I just updated the pr. would you please kindly tell me what you think. Thanks...

@rwightman its been a few months since my last changes, could you kindly please tell me if everything is OK or I'm missing sth here? Id really like to make...

It's not needed as bad as before anyway, especially when the likes of mlir projects (like torch-mlir)are working very well today. Though it's still very unfortunate!

calcPSF is actually creating a kernel which is then shifted and converted into frequency domain (fft), and then multiplied by the image's dft. its called PSF or point spread function...

Thanks, do you plan on releasing one at all? or are they just there to make us dig deeper ourselves and get more involved?