app.py

import gradio as gr
from transformers import BlipProcessor, BlipForConditionalGeneration
from PIL import Image

# Load the BLIP image captioning model and processor
processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning-base")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")

# Function to generate description for an image using BLIP
def describe_image(image: Image.Image):
    try:
        inputs = processor(images=image, return_tensors="pt")
        out = model.generate(**inputs)
        description = processor.decode(out[0], skip_special_tokens=True)
        return description
    except Exception as e:
        return f"Error describing the image: {e}"

# Chatbot logic
def chat(user_input, chat_history, image):
    try:
        response = f"AI Response: {user_input}"
        if image is not None:
            image_description = describe_image(image)
            response += f"\n\n[Image Description]: {image_description}"
        chat_history.append(("User", user_input))
        chat_history.append(("AI", response))
        formatted_history = "\n".join([f"{role}: {msg}" for role, msg in chat_history])
        return formatted_history, chat_history
    except Exception as e:
        return f"Error: {e}", chat_history

# Build Gradio interface
with gr.Blocks(css="""
    body {
        background-color: #f7f7f7;
        color: #333;
        font-family: 'Roboto', sans-serif;
    }
    .gradio-container {
        width: 100%;
        max-width: 700px;
        margin: auto;
        background-color: #ffffff;
        padding: 30px;
        border-radius: 15px;
        box-shadow: 0px 10px 30px rgba(0, 0, 0, 0.1);
    }
    .textbox, .button {
        margin-bottom: 15px;
    }
    #chatbox {
        height: 300px;
        overflow-y: auto;
        border: 1px solid #dcdcdc;
        padding: 20px;
        border-radius: 10px;
        background-color: #f9f9f9;
        margin-bottom: 20px;
        font-size: 14px;
        line-height: 1.6;
    }
""") as demo:
    
    gr.Markdown("## 🤖 **AI Chatbot with Image Captioning (BLIP + Gradio)**")
    gr.Markdown("Upload an image to generate a caption, then chat with the bot.")

    with gr.Column():
        user_input = gr.Textbox(label="Your Message", placeholder="Type your message here...", lines=2)
        submit_button = gr.Button("Send Message")
        clear_button = gr.Button("Clear Chat")
        chatbot_output = gr.Textbox(label="Chat History", lines=12, interactive=False, elem_id="chatbox")

        image_input = gr.Image(label="Upload Image", type="pil", elem_id="image-upload")
        upload_button = gr.Button("Describe Image")
        image_caption = gr.Textbox(label="Image Description", interactive=False)

    chat_history = gr.State([])

    submit_button.click(fn=chat, inputs=[user_input, chat_history, image_input], outputs=[chatbot_output, chat_history])
    clear_button.click(fn=lambda: ("", []), inputs=[], outputs=[chatbot_output, chat_history])
    upload_button.click(fn=describe_image, inputs=[image_input], outputs=[image_caption])

# Launch the app
if __name__ == "__main__":
    demo.launch()