Delete app.py

app.py

@@ -1,161 +0,0 @@
-import gradio as gr
-import torch
-import time
-from PIL import Image, ImageDraw, ImageFont
-from transformers import (
-    AutoProcessor, 
-    Owlv2ForObjectDetection,
-    Qwen2VLForConditionalGeneration
-)
-
-# Initialize models
-obj_processor = AutoProcessor.from_pretrained("google/owlv2-base-patch16-ensemble")
-obj_model = Owlv2ForObjectDetection.from_pretrained("google/owlv2-base-patch16-ensemble")
-cbt_model = Qwen2VLForConditionalGeneration.from_pretrained(
-    "Qwen/Qwen2-VL-2B-Instruct",
-    torch_dtype="auto",
-    device_map="auto",
-)
-cbt_processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-2B-Instruct")
-
-# Color palette for bounding boxes
-colors = [
-    (255, 0, 0), (0, 255, 0), (0, 0, 255), (255, 165, 0), (75, 0, 130),
-    (255, 255, 0), (0, 255, 255), (255, 105, 180), (138, 43, 226), (0, 128, 0)
-]
-
-# Conversation history
-history = [
-    {
-        "role": "system",
-        "content": [
-            {"type": "text", "text": "You are a helpful AI assistant."}
-        ]
-    }
-]
-
-# Helper Functions
-def detect_single_image(image, objects):
-    """Process one image for object detection."""
-    inputs = obj_processor(text=[objects], images=image, return_tensors="pt")
-    with torch.no_grad():
-        outputs = obj_model(**inputs)
-    target_sizes = torch.Tensor([image.size[::-1]])
-    results = obj_processor.post_process_object_detection(
-        outputs=outputs, threshold=0.2, target_sizes=target_sizes
-    )[0]
-    return results["boxes"], results["scores"], results["labels"]
-
-def annotate_image(image, boxes, scores, labels, objects):
-    """Draw bounding boxes on image."""
-    draw = ImageDraw.Draw(image)
-    font = ImageFont.load_default()
-    object_list = [obj.strip() for obj in objects.split(",")]
-    
-    for box, score, label in zip(boxes, scores, labels):
-        box = [round(coord, 2) for coord in box.tolist()]
-        color = colors[label % len(colors)]
-        draw.rectangle(box, outline=color, width=3)
-        draw.text((box[0], box[1]), f"{object_list[label]}: {score:.2f}", fill=color)
-    return image
-
-# Core Functions
-def detect_objects(images, objects):
-    """Process multiple images for object detection."""
-    if not images:
-        return []
-    
-    annotated_images = []
-    for image in images:
-        boxes, scores, labels = detect_single_image(image, objects)
-        annotated_images.append(annotate_image(image, boxes, scores, labels, objects))
-    
-    history.append({
-        'role': 'system',
-        'content': [{'type': 'text', 'text': f'Processed {len(images)} images'}]
-    })
-    return annotated_images
-
-def user(message, chat_history):
-    """Handle user message submission."""
-    return "", chat_history + [[message, ""]]
-
-def chat_function(images, chat_history):
-    """Generate chatbot response based on images and chat history."""
-    if not chat_history:
-        return chat_history
-    
-    message = chat_history[-1][0]
-    history.append({
-        "role": "user",
-        "content": [
-            {"type": "text", "text": message},
-            *[{"type": "image"} for _ in images]
-        ]
-    })
-
-    inputs = cbt_processor(
-        text=[cbt_processor.apply_chat_template(history, add_generation_prompt=True)],
-        images=images,
-        padding=True,
-        return_tensors="pt"
-    ).to("cuda" if torch.cuda.is_available() else "cpu")
-
-    output_ids = cbt_model.generate(**inputs, max_new_tokens=1024)
-    bot_output = cbt_processor.batch_decode(output_ids, skip_special_tokens=True)[0]
-    
-    history.append({"role": "assistant", "content": [{"type": "text", "text": bot_output}]})
-    chat_history[-1][1] = bot_output
-    return chat_history
-
-# Gradio Interface
-with gr.Blocks() as demo:
-    with gr.Row():
-        with gr.Column(scale=1):
-            gr.Markdown("## Image Upload")
-            image_input = gr.Gallery(
-                label="Upload Images",
-                type="pil",
-                columns=2,
-                height="auto"
-            )
-            objects_input = gr.Textbox(
-                label="Objects to Detect",
-                placeholder="cat, dog, car...",
-                interactive=True
-            )
-            detect_button = gr.Button("Detect Objects")
-            image_output = gr.Gallery(
-                label="Detection Results",
-                columns=2,
-                height="auto"
-            )
-
-        with gr.Column(scale=2):
-            chatbot = gr.Chatbot(height=500)
-            msg = gr.Textbox(
-                label="Chat with the AI",
-                placeholder="Ask about the images..."
-            )
-            clear = gr.ClearButton([msg, chatbot, image_input, image_output])
-
-    # Event handlers
-    detect_button.click(
-        fn=detect_objects,
-        inputs=[image_input, objects_input],
-        outputs=image_output
-    )
-    
-    msg.submit(
-        fn=user,
-        inputs=[msg, chatbot],
-        outputs=[msg, chatbot],
-        queue=False
-    ).then(
-        fn=chat_function,
-        inputs=[image_input, chatbot],
-        outputs=[chatbot]
-    )
-
-if __name__ == "__main__":
-    demo.launch()