feat: added gallery

README.md

@@ -1,4 +1,4 @@
-# ![](logo.jpeg "FlaxDiff")
+# ![](images/logo.jpeg "FlaxDiff")
 
 ## A Versatile and Easy-to-Understand Diffusion Library
 
@@ -134,6 +134,49 @@ samples = sampler.generate_images(num_images=64, diffusion_steps=100, start_step
 plotImages(samples, dpi=300)
 ```
 
+## Gallery
+
+### Images generated by Heun Sampler in 10 steps (20 model inferences as Heun takes 2x inference steps)
+
+**Params**: 
+`Dataset: oxford_flowers102`
+`Batch size: 16`
+`Image Size: 64`
+`Training Epochs: 1000`
+`Steps per epoch: 511`
+
+`Training Noise Schedule: EDMNoiseScheduler`
+`Inference Noise Schedule: KarrasEDMPredictor`
+
+`Model: UNet(emb_features=256, 
+            feature_depths=[64, 128, 256, 512],
+            attention_configs=[{"heads":4}, {"heads":4}, {"heads":4}, {"heads":4}, {"heads":4}],
+            num_res_blocks=2,
+            num_middle_res_blocks=1)`
+
+![Heun Sampler results](images/heun.png)
+
+
+### Images generated by DDPM Sampler in 1000 steps
+
+**Params**: 
+`Dataset: oxford_flowers102`
+`Batch size: 16`
+`Image Size: 64`
+`Training Epochs: 1000`
+`Steps per epoch: 511`
+
+`Training Noise Schedule: CosineNoiseSchedule`
+`Inference Noise Schedule: CosineNoiseSchedule`
+
+`Model: UNet(emb_features=256, 
+            feature_depths=[64, 128, 256, 512],
+            attention_configs=[{"heads":4}, {"heads":4}, {"heads":4}, {"heads":4}, {"heads":4}],
+            num_res_blocks=2,
+            num_middle_res_blocks=1)`
+
+![DDPM Sampler results](images/ddpm2.png)
+
 ## Contribution
 
 Feel free to contribute by opening issues or submitting pull requests. Let's make FlaxDiff better together!