balabala

Question about the DownScaleLoss

1

There is no DownScaleLoss in the paper, could we set the param 'lambda_bicubic' to 0?

Thanks for providing the codes! I have successfully converted my model to tflite with fixed input shapes, but I have no idea how to convert it with dynamic input shapes....

How to evaluate the pretrained model on the jumpcp dataset.

1

Hi, thanks for your great work! Now I am trying to test the trained model (with weight `cpjump_cellpaint_bf_channelvit_small_p8_with_hcs_supervised`) on jumpcp with my custom dataloader (since I don't want so many...

I didn't find it in the paper. Does anyone know where it came from?

Hi! I found that you trained different models for different image sizes, e.g., latent size 16x16 and 24x24. This setting makes sense, but I wonder have you tried to use...