search

jeya-maria-jose / Medical-Transformer

Official Pytorch Code for "Medical Transformer: Gated Axial-Attention for Medical Image Segmentation" - MICCAI 2021
MIT License
791 stars 176 forks source link

请教，einsum()函数报错 #77

Open orderer0001 opened 1 year ago

orderer0001 commented 1 year ago

RuntimeError: einsum(): operands do not broadcast with remapped shapes [original->remapped]: [277, 8, 1, 234]->[277, 8, 234, 1, 1] [1, 64, 64]->[1, 1, 64, 64, 1]

Drwr97 commented 1 year ago

Hello, has this problem been solved?

LenisLin commented 1 year ago

I also met the same problem. You could review the codes and noticed the input size of your data

djy123d commented 1 year ago

我也遇到这种情况了需要怎么解决呢

nightrain-vampire commented 1 year ago

I also encounter this problem：RuntimeError: size of dimension does not match previous size, operand 1, dim 1. I have resized the images to 512 × 512. My q is [512, 8, 1, 256] and q_embedding is [1, 64, 64]. How can I solve it?