Update app.py

app.py

@@ -1,14 +1,103 @@
 import gradio as gr
 import spaces
 import torch
-
-zero = torch.Tensor([0]).cuda()
-print(zero.device) # <-- 'cpu' 🤔
+import os
 
 @spaces.GPU
 def greet(n):
     print(zero.device) # <-- 'cuda:0' 🤗
     return f"Hello {zero + n} Tensor"
 
-demo = gr.Interface(fn=greet, inputs=gr.Number(), outputs=gr.Text())
-demo.launch()
+#demo = gr.Interface(fn=greet, inputs=gr.Number(), outputs=gr.Text())
+#demo.launch()
+
+
+import gradio as gr
+
+IMAGE_FOLDER = "./images"  # Change this to your folder path
+
+def get_image_list(folder):
+    if not os.path.exists(folder):
+        return []
+    valid_extensions = (".png", ".jpg", ".jpeg", ".webp", ".gif")
+    return [f for f in os.listdir(folder) if f.lower().endswith(valid_extensions)]
+
+image_files = get_image_list(IMAGE_FOLDER)
+
+def load_image(selected_file):
+    if not selected_file:
+        return None
+    return os.path.join(IMAGE_FOLDER, selected_file)
+
+# Generation pipeline function using both the source image and text prompt
+def process_and_generate(selected_file, prompt):
+    if not selected_file:
+        return None, "Please select an input image first."
+    if not prompt:
+        return None, "Please enter a text prompt."
+    
+    input_path = os.path.join(IMAGE_FOLDER, selected_file)
+    
+    # -------------------------------------------------------------
+    # PLACEHOLDER: Insert your Multi-ControlNet / SD pipeline here
+    # e.g., output = pipeline(image=input_path, prompt=prompt)
+    # -------------------------------------------------------------
+    
+    # For demonstration, returning the input image path
+    generated_output = input_path 
+    
+    return generated_output, f"Generated matching prompt: '{prompt}'"
+
+with gr.Blocks() as demo:
+    gr.Markdown("# Guided Image Generation Pipeline")
+    
+    with gr.Row():
+        # Column 1: Source Image Selection
+        with gr.Column():
+            gr.Markdown("### 1. Source Image")
+            image_dropdown = gr.Dropdown(
+                choices=image_files, 
+                label="Select Input Image from Folder",
+                value=image_files if image_files else None
+            )
+            input_preview = gr.Image(type="filepath", label="Input Preview")
+            
+        # Column 2: Prompt Engineering & Execution
+        with gr.Column():
+            gr.Markdown("### 2. Generation Settings")
+            prompt_input = gr.Textbox(
+                label="Text Prompt", 
+                placeholder="Describe modifications, styles, or Canny/Depth map directions...",
+                lines=3
+            )
+            generate_btn = gr.Button("Generate Image", variant="primary")
+            
+        # Column 3: Output Visualization
+        with gr.Column():
+            gr.Markdown("### 3. Pipeline Output")
+            output_image = gr.Image(type="filepath", label="Generated Output")
+            status_text = gr.Textbox(label="Status", interactive=False)
+            
+    # Core event binds
+    image_dropdown.change(
+        fn=load_image, 
+        inputs=image_dropdown, 
+        outputs=input_preview
+    )
+    
+    generate_btn.click(
+        fn=process_and_generate,
+        inputs=[image_dropdown, prompt_input],
+        outputs=[output_image, status_text]
+    )
+    
+    # Initial state setup
+    if image_files:
+        demo.load(
+            fn=load_image, 
+            inputs=image_dropdown, 
+            outputs=input_preview
+        )
+
+if __name__ == "__main__":
+    demo.launch()