RuntimeError: The size of tensor a (60) must match the size of tensor b (59) at non-singleton dimension 3

[daemonkillerr]

I have a custom dataset of 59x59 images and I face this error. Kindly help.

File openstl\openstl\models\simvp_model.py:95, in Decoder.forward(self, hid, enc1)
93 for i in range(0, len(self.dec)-1):
94 hid = self.deci
---> 95 Y = self.dec[-1](hid + enc1)
96 Y = self.readout(Y)
97 return Y

RuntimeError: The size of tensor a (60) must match the size of tensor b (59) at non-singleton dimension 3

Thanks

[chengtan9907]

In SimVP, the image is downsampled and then upsampled back to its original size. However, 59x59 images cannot be upsampled back to the same dimensions. To address this, I suggest padding your image to 60x60 at the beginning of the forward function and removing the padding at the end of the forward function.

[chengtan9907]

I will close this issue if there are no more questions. Please feel free to open a new issue when you have new questions.