Unterschiedliche Parameter der Konzerndimension in der Up-and-Down-Stichprobenphase von U-förmigen Netzwerken im Diffusi ⇐ Python

[Anonymous]

Der Fehler, den ich bekomme, ist < /p>

Code: Select all

  File "/media/xd/hdd/wxy/spectraldiff_diffusion-master/codes/unet3d.py", line 113, in forward
x = torch.cat((x, residual_x), dim=1)
RuntimeError: Sizes of tensors must match except in dimension 1. Expected size 24 but got size 25 for tensor number 1 in the list.

Die Eingabe des Trainingsdatensatzes zum Modell ist Paviau.

Code: Select all

down = torch.Size([4, 64, 25, 64, 64])
up = torch.Size([4, 64, 24, 64, 64])
< /code>
Der gesamte Code wie folgt: < /p>

def forward(self, x, timestep, feature=False):
# Embedd time
t = self.time_mlp(timestep)
# Initial conv
x = self.conv0(x)
# Unet
residual_inputs = []
for down in self.downs:
x = down(x, t)
residual_inputs.append(x)
for up in self.ups:
residual_x = residual_inputs.pop()
# print("down=",residual_x.shape, "up=", x.shape)
# Add residual x as additional channels
x = torch.cat((x, residual_x), dim=1)*
if feature:
self.features.append(x.detach().cpu().numpy())
x = up(x, t)
return self.output(x)