app.py

import gradio
import torch
import PIL
from torchvision import transforms

from diffusers import StableDiffusionInpaintPipeline

pipeline = StableDiffusionInpaintPipeline.from_pretrained("runwayml/stable-diffusion-inpainting",
                                                          revision="fp16",
                                                          torch_dtype=torch.float16,
                                                          safety_checker=lambda images, **kwargs: (images, False))

pipeline.to("cuda")
#generator = torch.Generator(device).manual_seed(seed)

def diffuse(prompt, negativePrompt, inputImage, mask, guidanceScale, numInferenceSteps):
    return pipeline(prompt=prompt,
                    negative_prompt=negativePrompt,
                    image=inputImage,
                    mask_image=mask,
                    guidance_scale=guidanceScale,
                    num_inference_steps=numInferenceSteps).images[0]

prompt = gradio.Textbox(label="Prompt", placeholder="A person in a room", lines=3)
negativePrompt = gradio.Textbox(label="Negative Prompt", placeholder="Text", lines=3)

inputImage = gradio.Image(label="Input Image", type="pil")
#inputFeed = gradio.Image(label="Input Feed", source="webcam", streaming=True)
mask = gradio.Image(label="Mask", type="pil")
outputImage = gradio.Image(label="Extrapolated Field of View")
guidanceScale = gradio.Slider(label="Guidance Scale", maximum=1, value = 0.75)
numInferenceSteps = gradio.Slider(label="Number of Inference Steps", maximum=100, value = 0)

ux = gradio.Interface(fn=diffuse, title="View Diffusion", inputs=[prompt, negativePrompt, inputImage, mask, guidanceScale, numInferenceSteps], outputs=outputImage, live=True)
ux.launch()