Update model and sample

README.md

@@ -11,15 +11,20 @@ The models included was trained on "multi-resolution" images of "Lego Sets"
 
 ## Using the model
 
-* common subject prompt tokens: `<wathever> lego set`
+* common subject prompt tokens: `lego set <wathever>`
 
 ## Example prompts
 
-`mcdonald restaurant AND lego set`:
+`mcdonald restaurant lego set`:
 
 <img src="https://huggingface.co/cyburn/lego_set/resolve/main/1.jpg" alt="Picture." width="500"/>
 
-`a digital drawing of crow, skull, animal, symmetry, flower AND lego set`:
+`lego set crow, skull`:
 
 <img src="https://huggingface.co/cyburn/lego_set/resolve/main/2.jpg" alt="Picture." width="500"/>
 
+## img2img example
+
+`lego set ottawa parliament building sharp focus`:
+
+<img src="https://huggingface.co/cyburn/lego_set/resolve/main/3.jpg" alt="Picture." width="500"/>

lego_set.ckpt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f3dfb8d1ed4e4c6f2b4ddc9ae197f8d7ae0114d2f47ae5b7c6be45d384d32513
+oid sha256:5317a379be6961b1ab54b5daa1ea8dc686534ac8b975fab6509631b38d31320a
 size 2132856686