Shapes incompatible when testing Unet++ on COCO

[sparshgarg23]

I was using your Image segmentation using customized UNET 3+ and hybrid loss and user_guide_models notebook " to train with Unet_plus_2d model on COCO.
but during training,I got the following error
Shapes (2069, 128, 128, 2) and (4, 2069, 128, 128, 3) are incompatible.
Any idea why this is happening,the only changes i made in the seg attention unet backbone notebook was to change the model
Also in the notebook Image segmentation using customized UNET 3+ and hybrid loss
while displaying the images,why is the image so grainy and what can be done to resolve it

[yingkaisha]

Hi, would you mind providing your code and full error message? I will look into it.

When you say grainy, do you mean the images are too coarse? --- it is because all the samples are low-resolution images, it has nothing to do with the model.