some more images

run.py

@@ -19,14 +19,14 @@ unet.to(torch_device)
 vqvae.to(torch_device)
 
 # generate gaussian noise to be decoded
-generator = torch.manual_seed(23)
+generator = torch.manual_seed(1)
 noise = torch.randn(
     (1, unet.in_channels, unet.image_size, unet.image_size),
     generator=generator,
 ).to(torch_device)
 
 # set inference steps for DDIM
-scheduler.set_timesteps(num_inference_steps=50)
+scheduler.set_timesteps(num_inference_steps=200)
 
 image = noise
 for t in tqdm.tqdm(scheduler.timesteps):
@@ -64,8 +64,8 @@ import tqdm
 pipeline = LatentDiffusionUncondPipeline.from_pretrained("./")
 
 # generatae image by calling the pipeline
-generator = torch.manual_seed(23)
-image = pipeline(generator=generator, num_inference_steps=50)["sample"]
+generator = torch.manual_seed(1)
+image = pipeline(generator=generator, num_inference_steps=200)["sample"]
 
 # process image
 image_processed = image.cpu().permute(0, 2, 3, 1)