vibe or bust

.gitignore

@@ -0,0 +1,62 @@
+# Python virtual environments
+venv/
+env/
+.virtualenv/
+.venv/
+
+# Python bytecode and caches
+__pycache__/
+*.py[cod]
+*$py.class
+*.so
+
+# Virtual environment activation files
+*.egg-info/
+dist/
+build/
+*.egg
+
+# Dependency lock files (optional, include if you don’t want them tracked)
+# requirements.lock
+# Pipfile.lock
+
+# IDE and editor files
+.idea/
+.vscode/
+*.sublime-project
+*.sublime-workspace
+
+# OS-generated files
+.DS_Store
+Thumbs.db
+
+# Logs and temporary files
+*.log
+*.tmp
+temp/
+
+# Model weights and caches (if downloaded locally)
+*.ckpt
+*.safetensors
+*.bin
+*.pth
+model_cache/
+cache/
+diffusers_cache/
+
+# Output files from generation
+output/
+*.png
+*.jpg
+*.jpeg
+*.gif
+
+# Hugging Face Spaces-specific ignores (optional)
+# Uncomment if you don’t want these tracked
+# .hf_cache/
+# .gradio/
+
+# Miscellaneous
+*.bak
+*.swp
+*~

app.py

@@ -0,0 +1,55 @@
+import torch
+from diffusers import StableDiffusionControlNetPipeline, ControlNetModel, UniPCMultistepScheduler
+import numpy as np
+import cv2
+from PIL import Image
+import gradio as gr
+
+# Load ControlNet and Stable Diffusion models
+controlnet = ControlNetModel.from_pretrained(
+    "lllyasviel/sd-controlnet-canny",
+    torch_dtype=torch.float16
+)
+pipe = StableDiffusionControlNetPipeline.from_pretrained(
+    "runwayml/stable-diffusion-v1-5",
+    controlnet=controlnet,
+    torch_dtype=torch.float16
+)
+pipe.scheduler = UniPCMultistepScheduler.from_config(pipe.scheduler.config)
+pipe.to("cuda")  # Use GPU in Spaces
+pipe.enable_model_cpu_offload()  # Optimize memory usage
+
+# Function to process image with ControlNet
+def generate_image(input_image, prompt):
+    # Convert input image to numpy array and extract Canny edges
+    image = np.array(input_image)
+    low_threshold = 100
+    high_threshold = 200
+    canny_image = cv2.Canny(image, low_threshold, high_threshold)
+    canny_image = canny_image[:, :, None]
+    canny_image = np.concatenate([canny_image, canny_image, canny_image], axis=2)
+    canny_image = Image.fromarray(canny_image)
+
+    # Generate image with ControlNet
+    output = pipe(
+        prompt,
+        image=canny_image,
+        num_inference_steps=20,
+        controlnet_conditioning_scale=0.8
+    ).images[0]
+    return output
+
+# Gradio interface
+interface = gr.Interface(
+    fn=generate_image,
+    inputs=[
+        gr.Image(type="pil", label="Upload an Image"),
+        gr.Textbox(label="Prompt", placeholder="e.g., 'a futuristic city'")
+    ],
+    outputs=gr.Image(type="pil", label="Generated Image"),
+    title="Stable Diffusion with ControlNet (Canny)",
+    description="Upload an image and enter a prompt to generate a new image guided by Canny edges."
+)
+
+# Launch the app
+interface.launch()

requirements.txt

@@ -0,0 +1,7 @@
+torch>=2.0.0
+diffusers>=0.27.0
+transformers>=4.30.0
+accelerate>=0.20.0
+gradio>=4.0.0
+numpy>=1.23.0
+opencv-python>=4.8.0