Update README.md

README.md

@@ -1,12 +1,18 @@
 ---
 {}
 ---
-```# first git clone this repo
+```
 import os
 import torch
 if not os.path.exists("seed2"):
     os.system("git clone https://huggingface.co/ontocord/seed2")
+from diffusers import StableUnCLIPImg2ImgPipeline
 from seed2.seed2_tokenizer import Seed2Tokenizer
+tokenizer = None 
+pipe = None
+torch.cuda.empty_cache()
+pipe = StableUnCLIPImg2ImgPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-unclip', torch_dtype=torch.float16)
+pipe = pipe.to('cuda')
 tokenizer = Seed2Tokenizer.from_pretrained("seed2", torch_dtype=torch.float16).to('cuda')
 if not os.path.exists("cat.jpg"):
     os.system("wget https://images.unsplash.com/photo-1574158622682-e40e69881006?w=300 -O cat.jpg")
@@ -14,8 +20,9 @@ tokens = tokenizer.encode_image("cat.jpg")
 print (tokens)
 image_embeds = tokenizer.model.get_codebook_entry(tokens)
 print (image_embeds) 
-from diffusers import StableUnCLIPImg2ImgPipeline
-pipe = StableUnCLIPImg2ImgPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-unclip', torch_dtype=torch.float16)
-pipe = pipe.to('cuda')
-tokenizer.decode(pipe, tokens)[0]
+import time
+t = time.time()
+image = tokenizer.decode(pipe, tokens)[0] # this is using a fixed latent initalized by the model, so a call with the same tokens will produce the same image
+print (time.time()-t)
+image
 ```