MeteorsHub / GLIA-Net

A segmentation network for intracranial aneurysm on CTA images using pytorch
39 stars 10 forks source link

RuntimeError: Given input size: (64x1x1x1). Calculated output size: (64x0x0x0). Output size is too small #5

Closed zhang5957 closed 3 years ago

zhang5957 commented 3 years ago

I used the default parameters and 512 * 512 image training,and still got the error

zhang5957 commented 3 years ago

I printed the size of the input data before entering the model,it shows torch.Size([3, 4, 4, 4]). And no matter how I modify the patch size, it doesn't change,can you help me?

MeteorsHub commented 3 years ago

You should make sure your data format is correct. If not, you can debug in the dataloader file