Another ifx

pipeline.py

@@ -1021,9 +1021,12 @@ class StableDiffusionControlNetImg2ImgPipeline(DiffusionPipeline, TextualInversi
 
                         down_block_res_samples.append(down_block_res_sample)
                         mid_block_res_samples.append(mid_block_res_sample)
-                
-                    down_block_res_samples = torch.cat(down_block_res_samples, dim=0)
-                    mid_block_res_sample = torch.cat(mid_block_res_samples, dim=0)
+
+                    down_block_res_samples = torch.cat([tensor for tensor in down_block_res_samples], dim=0)
+                    mid_block_res_sample = torch.cat([tensor for tensor in mid_block_res_samples], dim=0)
+
+                    # down_block_res_samples = torch.cat(down_block_res_samples, dim=0)
+                    # mid_block_res_sample = torch.cat(mid_block_res_samples, dim=0)
 
 
                 # predict the noise residual