Create app.py

app.py

@@ -0,0 +1,57 @@
+import os
+from PIL import Image, ImageDraw
+import cv2
+import numpy as np
+from IPython.display import HTML
+from base64 import b64encode
+
+import torch
+from torch import autocast
+from torch.nn import functional as F
+from diffusers import StableDiffusionPipeline, AutoencoderKL
+from diffusers import UNet2DConditionModel, PNDMScheduler, LMSDiscreteScheduler
+from diffusers.schedulers.scheduling_ddim import DDIMScheduler
+from transformers import CLIPTextModel, CLIPTokenizer
+from tqdm.auto import tqdm
+import gradio as gr
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+
+from diffusers import StableDiffusionInpaintPipeline
+pipe = StableDiffusionInpaintPipeline.from_pretrained(
+    "ShreeKanade07/Real-Image-pipeline", torch_dtype=torch.float16
+)
+#pipe = pipe.to("cuda")
+
+
+
+
+# Define the predict function
+def predict(image,mask,prompt):
+
+    prompt = prompt
+    image = Image.fromarray(image)
+    image=image.convert("RGB").resize((512, 512))
+    mask_image = Image.fromarray(mask)
+    mask_image=mask_image.convert("RGB").resize((512, 512))
+    strength=0.9
+    generator = torch.manual_seed(32)
+    negative_prompt="zoomed in, blurry, oversaturated, warped,artifacts,flickers"
+    images = pipe(prompt=prompt, image=image, mask_image=mask_image, strength=strength, negative_prompt=negative_prompt, generator=generator).images
+    return images[0]
+
+
+
+# Create the Gradio interface
+gr.Interface(
+    predict,
+    title='Stable Diffusion In-Painting',
+    inputs=[
+        gr.Image(label='Image'),
+        gr.Image(label='Mask'),
+        gr.Textbox(label='Prompt')
+    ],
+    outputs=[
+        gr.Image(label='Output Image')
+    ]
+).launch(debug=True, share=True)