one more residual, after seeing good results on unconditional generation locally

dalle2_pytorch/dalle2_pytorch.py

@@ -1488,7 +1488,7 @@ class Unet(nn.Module):
             ]))
 
         self.final_conv = nn.Sequential(
-            ResnetBlock(dim, dim, groups = resnet_groups[0]),
+            ResnetBlock(dim * 2, dim, groups = resnet_groups[0]),
             nn.Conv2d(dim, self.channels_out, 1)
         )
 
@@ -1560,6 +1560,7 @@ class Unet(nn.Module):
         # initial convolution
 
         x = self.init_conv(x)
+        r = x.clone() # final residual
 
         # time conditioning
 
@@ -1689,9 +1690,7 @@ class Unet(nn.Module):
 
             x = upsample(x)
 
-        if len(hiddens) > 0:
-            x = torch.cat((x, hiddens.pop()), dim = 1)
-
+        x = torch.cat((x, r), dim = 1)
         return self.final_conv(x)
 
 class LowresConditioner(nn.Module):

dalle2_pytorch/version.py

@@ -1 +1 @@
-__version__ = '0.9.1'
+__version__ = '0.9.2'