Add prints to load_learned_embed_in_clip func

app.py

@@ -95,32 +95,43 @@ pipe = StableDiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4",
 # pipe = StableDiffusionPipeline.from_pretrained("CompVis/stable-diffusion-v1-4", use_auth_token=True, revision="fp16", torch_dtype=torch.float16).to("cuda")
 
 def load_learned_embed_in_clip(learned_embeds_path, text_encoder, tokenizer, token=None):
+  print("1 <****************")
   loaded_learned_embeds = torch.load(learned_embeds_path, map_location="cpu")
   
   # separate token and the embeds
   trained_token = list(loaded_learned_embeds.keys())[0]
+  print("2 <****************")
   embeds = loaded_learned_embeds[trained_token]
+  print("3 <****************")
 
   # cast to dtype of text_encoder
   dtype = text_encoder.get_input_embeddings().weight.dtype
+  print("4 <****************")
   
   # add the token in tokenizer
   token = token if token is not None else trained_token
+  print("5 <****************")
   num_added_tokens = tokenizer.add_tokens(token)
+  print("6 <****************")
   i = 1
   while(num_added_tokens == 0):
     print(f"The tokenizer already contains the token {token}.")
     token = f"{token[:-1]}-{i}>"
     print(f"Attempting to add the token {token}.")
     num_added_tokens = tokenizer.add_tokens(token)
+    print("7 <****************")
     i+=1
+  print("8 <****************")
   
   # resize the token embeddings
   text_encoder.resize_token_embeddings(len(tokenizer))
+  print("9 <****************")
   
   # get the id for the token and assign the embeds
   token_id = tokenizer.convert_tokens_to_ids(token)
+  print("10 <****************")
   text_encoder.get_input_embeddings().weight.data[token_id] = embeds
+  print("11 <****************")
   return token