Update model.py

model.py

@@ -461,20 +461,18 @@ class StreamMultiDiffusion(nn.Module):
                 do_classifier_free_guidance=(self.guidance_scale > 1.0),
                 negative_prompt=[negative_prompt],
             )  # ((1, 77, 768): cond, (1, 77, 768): uncond)
-            print('111111')
 
             self.state['background'].image = image
             self.state['background'].latent = (
                 self.encode_imgs(T.ToTensor()(image_)[None].to(self.device, self.dtype))
             )  # (1, 3, H, W)
+            print(self.state['background'].latent.dtype, self.state['background'].latent.device, self.vae.dtype, self.vae.device)
             self.state['background'].prompt = prompt
             self.state['background'].negative_prompt = negative_prompt
             self.state['background'].embed = embed
-            print('222222')
 
             if self.bootstrap_steps[0] > 0:
                 if self.white is None:
-                    print('333333333')
                     self.white = self.encode_imgs(torch.ones(1, 3, self.height, self.width, dtype=self.dtype, device=self.device))
                 mix_ratio = self.bootstrap_mix_ratios[:, None, None, None]
                 self.bootstrap_latent = mix_ratio * self.white + (1.0 - mix_ratio) * self.state['background'].latent