do not noise for the last step in ddim

dalle2_pytorch/dalle2_pytorch.py

@@ -1059,10 +1059,10 @@ class DiffusionPrior(nn.Module):
 
             c1 = eta * ((1 - alpha / alpha_next) * (1 - alpha_next) / (1 - alpha)).sqrt()
             c2 = ((1 - alpha_next) - torch.square(c1)).sqrt()
-            new_noise = torch.randn_like(image_embed)
+            noise = torch.randn_like(image_embed) if time_next > 0 else 0.
 
             img = x_start * alpha_next.sqrt() + \
-                  c1 * new_noise + \
+                  c1 * noise + \
                   c2 * pred_noise
 
         return image_embed
@@ -1537,10 +1537,12 @@ class Unet(nn.Module):
         # text encoding conditioning (optional)
 
         self.text_to_cond = None
+        self.text_embed_dim = None
 
         if cond_on_text_encodings:
             assert exists(text_embed_dim), 'text_embed_dim must be given to the unet if cond_on_text_encodings is True'
             self.text_to_cond = nn.Linear(text_embed_dim, cond_dim)
+            self.text_embed_dim = text_embed_dim
 
         # finer control over whether to condition on image embeddings and text encodings
         # so one can have the latter unets in the cascading DDPMs only focus on super-resoluting
@@ -1769,6 +1771,8 @@ class Unet(nn.Module):
         text_tokens = None
 
         if exists(text_encodings) and self.cond_on_text_encodings:
+            assert self.text_embed_dim == text_encodings.shape[-1], f'the text encodings you are passing in have a dimension of {text_encodings.shape[-1]}, but the unet was created with text_embed_dim of {self.text_embed_dim}.'
+
             text_tokens = self.text_to_cond(text_encodings)
             text_tokens = text_tokens[:, :self.max_text_len]
 
@@ -2043,7 +2047,7 @@ class Decoder(nn.Module):
         self.noise_schedulers = nn.ModuleList([])
 
         for ind, (unet_beta_schedule, unet_p2_loss_weight_gamma, sample_timesteps) in enumerate(zip(beta_schedule, p2_loss_weight_gamma, self.sample_timesteps)):
-            assert sample_timesteps <= timesteps, f'sampling timesteps {sample_timesteps} must be less than or equal to the number of training timesteps {timesteps} for unet {ind + 1}'
+            assert not exists(sample_timesteps) or sample_timesteps <= timesteps, f'sampling timesteps {sample_timesteps} must be less than or equal to the number of training timesteps {timesteps} for unet {ind + 1}'
 
             noise_scheduler = NoiseScheduler(
                 beta_schedule = unet_beta_schedule,
@@ -2271,9 +2275,10 @@ class Decoder(nn.Module):
 
             c1 = eta * ((1 - alpha / alpha_next) * (1 - alpha_next) / (1 - alpha)).sqrt()
             c2 = ((1 - alpha_next) - torch.square(c1)).sqrt()
+            noise = torch.randn_like(img) if time_next > 0 else 0.
 
             img = x_start * alpha_next.sqrt() + \
-                  c1 * torch.randn_like(img) + \
+                  c1 * noise + \
                   c2 * pred_noise
 
         img = self.unnormalize_img(img)

dalle2_pytorch/version.py

@@ -1 +1 @@
-__version__ = '0.19.1'
+__version__ = '0.19.4'