add runway inpainting model with diffusers

app.py

@@ -11,23 +11,12 @@ import torch
 from torch import autocast
 import cv2
 from matplotlib import pyplot as plt
-from inpainting import StableDiffusionInpaintingPipeline
 from torchvision import transforms
-
-auth_token = os.environ.get("API_TOKEN") or True
-
-def download_image(url):
-    response = requests.get(url)
-    return PIL.Image.open(BytesIO(response.content)).convert("RGB")
+from diffusers import DiffusionPipeline
 
 device = "cuda" if torch.cuda.is_available() else "cpu"
-pipe = StableDiffusionInpaintingPipeline.from_pretrained(
-    "CompVis/stable-diffusion-v1-4",
-    revision="fp16", 
-    torch_dtype=torch.float16,
-    use_auth_token=auth_token,
-).to(device)
 
+pipe = DiffusionPipeline.from_pretrained("runwayml/stable-diffusion-inpainting", dtype=torch.float16, revision="fp16")
 
 transform = transforms.Compose([
       transforms.ToTensor(),
@@ -39,8 +28,8 @@ def predict(dict, prompt=""):
     with autocast("cuda"):
         init_image = dict["image"].convert("RGB").resize((512, 512))
         mask = dict["mask"].convert("RGB").resize((512, 512))
-        images = pipe(prompt = prompt, init_image=init_image, mask_image=mask, strength=0.8)["sample"]
-    return images[0]
+        output = pipe(prompt = prompt, image=init_image, mask_image=mask, strength=0.8,num_inference_steps=20)
+    return output.images[0]
 
 examples = [[dict(image="init_image.png", mask="mask_image.png"), "A panda sitting on a bench"]]
 

requirements.txt

@@ -1,7 +1,7 @@
 --extra-index-url https://download.pytorch.org/whl/cu113 
 torch
 torchvision
-diffusers
+git+https://github.com/huggingface/diffusers.git
 transformers
 ftfy
 numpy