Update README.md

README.md

@@ -11,11 +11,20 @@ Run the model as follows:
 from diffusers import UNetModel, GaussianDiffusion
 import torch
 
-model = UNetModel.from_pretrained("fusing/ddpm_dummy")
+# 1. Load model
+unet = UNetModel.from_pretrained("fusing/ddpm_dummy")
 
+# 2. Do one denoising step with model
 batch_size, num_channels, height, width = 1, 3, 32, 32
 dummy_noise = torch.ones((batch_size, num_channels, height, width))
 time_step = torch.tensor([10])
+image = unet(dummy_noise, time_step)
 
-image = model(dummy_noise, time_step)
+# 3. Load sampler
+sampler = GaussianDiffusion.from_config("fusing/ddpm_dummy")
+
+# 4. Sample image from sampler passing the model
+image = sampler.sample(model, batch_size=1)
+
+print(image)
 ```