update composable diffusion for an updated diffuser library

Author: nanlliu

examples/community/README.md

@@ -355,43 +355,45 @@ out = pipe(
 import torch as th
 import numpy as np
 import torchvision.utils as tvu
+
 from diffusers import DiffusionPipeline
 
+import argparse
+
+parser = argparse.ArgumentParser()
+parser.add_argument("--prompt", type=str, default="mystical trees | A magical pond | dark",
+                    help="use '|' as the delimiter to compose separate sentences.")
+parser.add_argument("--steps", type=int, default=50)
+parser.add_argument("--scale", type=float, default=7.5)
+parser.add_argument("--weights", type=str, default="7.5 | 7.5 | -7.5")
+parser.add_argument("--seed", type=int, default=2)
+parser.add_argument("--model_path", type=str, default="CompVis/stable-diffusion-v1-4")
+parser.add_argument("--num_images", type=int, default=1)
+args = parser.parse_args()
+
 has_cuda = th.cuda.is_available()
 device = th.device('cpu' if not has_cuda else 'cuda')
 
+prompt = args.prompt
+scale = args.scale
+steps = args.steps
+
 pipe = DiffusionPipeline.from_pretrained(
-    "CompVis/stable-diffusion-v1-4",
-    use_auth_token=True,
+    args.model_path,
     custom_pipeline="composable_stable_diffusion",
 ).to(device)
 
-
-def dummy(images, **kwargs):
-    return images, False
-
-pipe.safety_checker = dummy
-
-images = []
-generator = torch.Generator("cuda").manual_seed(0)
-
-seed = 0
-prompt = "a forest | a camel"
-weights = " 1 | 1"  # Equal weight to each prompt. Can be negative
+pipe.safety_checker = None
 
 images = []
-for i in range(4):
-    res = pipe(
-        prompt,
-        guidance_scale=7.5,
-        num_inference_steps=50,
-        weights=weights,
-        generator=generator)
-    image = res.images[0]
-    images.append(image)
+generator = th.Generator("cuda").manual_seed(args.seed)
+for i in range(args.num_images):
+    image = pipe(prompt, guidance_scale=scale, num_inference_steps=steps,
+                 weights=args.weights, generator=generator).images[0]
+    images.append(th.from_numpy(np.array(image)).permute(2, 0, 1) / 255.)
+grid = tvu.make_grid(th.stack(images, dim=0), nrow=4, padding=0)
+tvu.save_image(grid, f'{prompt}_{args.weights}' + '.png')
 
-for i, img in enumerate(images):
-    img.save(f"./composable_diffusion/image_{i}.png")
 ```
 
 ### Imagic Stable Diffusion