Add examples

.gitattributes

@@ -34,3 +34,12 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 2-base/unet/diffusion_pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
 2-1/unet/diffusion_pytorch_model.bin filter=lfs diff=lfs merge=lfs -text
+imgs filter=lfs diff=lfs merge=lfs -text
+imgs/21f-576-1024.png filter=lfs diff=lfs merge=lfs -text
+imgs/2b-1024-576.png filter=lfs diff=lfs merge=lfs -text
+imgs/2b-576-1024.png filter=lfs diff=lfs merge=lfs -text
+imgs/2bf-1024-576.png filter=lfs diff=lfs merge=lfs -text
+imgs/2bf-576-1024.png filter=lfs diff=lfs merge=lfs -text
+imgs/21-1024-576.png filter=lfs diff=lfs merge=lfs -text
+imgs/21-576-1024.png filter=lfs diff=lfs merge=lfs -text
+imgs/21f-1024-576.png filter=lfs diff=lfs merge=lfs -text

README.md

@@ -21,11 +21,16 @@ For usage, see - [How to Get Started with the Model](#how-to-get-started-with-th
 
 ### It aims to solve the following issues:
 1. Generated images looks like they are cropped from a larger image.
-Examples:
 
 2. Generating non-square images creates weird results, due to the model being trained on square images.
 Examples:
 
+| resolution      | model   |   stable diffusion           |   flex diffusion              |
+|:---------------:|:-------:|:----------------------------:|:-----------------------------:|
+| 576x1024 (9:16) | v2-1    | ![img](imgs/21-576-1024.png) | ![img](imgs/21f-576-1024.png) |
+| 576x1024 (9:16) | v2-base | ![img](imgs/2b-576-1024.png) | ![img](imgs/2bf-576-1024.png) |
+| 1024x576 (16:9) | v2-1    | ![img](imgs/21-1024-576.png) | ![img](imgs/21f-1024-576.png) |
+| 1024x576 (16:9) | v2-base | ![img](imgs/2b-1024-576.png) | ![img](imgs/2bf-1024-576.png) |
 
 ### Limitations:
 1. It's trained on a small dataset, so it's improvements may be limited.