hdjsjyl

Label format problem

1

Fist, thanks for your good work. I am a little confused about the lable format. For the annotations of the pedestrian dataset, the annotation has 'ignore', 'pedestrians', 'sitting people', 'others'...

How much is your fid score?

1

Hi, thanks for your great work. What is your fid score? I also trained for 50 epochs, but my fid score is only 15. thank you!

How to reproduce the results on the cityperson dataset?

35

@liuwei16 Thanks for sharing your excellent work. But, when I use two gpu with batch size of 4 images, learning rate with 2e-4 or 4e-4, I can not reproduce the...

Whether running the code has very high requirement for CPU? When I run the code, I found the gpu utility is very low. Thanks

Thanks for sharing your excellent work. Can you share your training log? Thanks

Single image test error " Clipping input data to the valid range for imshow with RGB data ([0..1] for floats or [0..255] for integers). Clipping input data to the valid...

Hi team, Thanks for your great work! I am trying to get object removal functionality from inpainting of SD. How to finetune inpainting model for object removal? What is the...

How to finetune inpainting model for object removal? What is the input prompt for object removal for both training and inference?

1

Hi team, Thanks for your great work! I am trying to get object removal functionality from inpainting of SD. How to finetune inpainting model for object removal? What is the...

How to fit it to inpainting?

2

Hi Author, Thanks for your great work. How to fit the current model to inpainting (condition and uncondition). Thanks

Fid score is different from the paper

4

Hi Author, Thanks for your great work. For the three checkpoints you provided "best_with_denoise", "best_without_denoise" and "checkpoint_300000.pth", I got fid scores around 39 by specifying the argument "--fast_fid". But the...