get the ldm pipeline working

models/vision/latent_diffusion/modeling_latent_diffusion.py

@@ -773,7 +773,6 @@ class AutoencoderKL(ModelMixin, ConfigMixin):
         in_channels,
         resolution,
         z_channels,
-        n_embed,
         embed_dim,
         remap=None,
         sane_index_shape=False,  # tell vector quantizer to return indices as bhw
@@ -794,7 +793,6 @@ class AutoencoderKL(ModelMixin, ConfigMixin):
             in_channels=in_channels,
             resolution=resolution,
             z_channels=z_channels,
-            n_embed=n_embed,
             embed_dim=embed_dim,
             remap=remap,
             sane_index_shape=sane_index_shape,
@@ -877,17 +875,16 @@ class LatentDiffusion(DiffusionPipeline):
         
         # get text embedding
         text_input = self.tokenizer(prompt, padding="max_length", max_length=77, return_tensors='pt').to(torch_device)
-        text_embedding = self.bert(**text_input)[0]
+        text_embedding = self.bert(text_input.input_ids)[0]
         
         num_trained_timesteps = self.noise_scheduler.num_timesteps
         inference_step_times = range(0, num_trained_timesteps, num_trained_timesteps // num_inference_steps)
 
         image = self.noise_scheduler.sample_noise(
-            (batch_size, self.unet.in_channels, self.unet.resolution // 8, self.unet.resolution // 8),
+            (batch_size, self.unet.in_channels, self.unet.image_size, self.unet.image_size),
             device=torch_device,
             generator=generator,
         )
-
         for t in tqdm.tqdm(reversed(range(num_inference_steps)), total=num_inference_steps):
             # get actual t and t-1
             train_step = inference_step_times[t]
@@ -928,8 +925,8 @@ class LatentDiffusion(DiffusionPipeline):
             else:
                 image = pred_mean
         
-        image = 1 / image
-        image = self.vqvae(image)
+        image = 1 /  0.18215 * image
+        image = self.vqvae.decode(image)
         image = torch.clamp((image+1.0)/2.0, min=0.0, max=1.0)
 
         return image

src/diffusers/models/unet_ldm.py

@@ -1026,7 +1026,7 @@ class UNetLDMModel(ModelMixin, ConfigMixin):
         hs = []
         if not torch.is_tensor(timesteps):
             timesteps = torch.tensor([timesteps], dtype=torch.long, device=x.device)
-        t_emb = timestep_embedding(timesteps, self.model_channels, repeat_only=False)
+        t_emb = timestep_embedding(timesteps, self.model_channels)
         emb = self.time_embed(t_emb)
 
         if self.num_classes is not None: