import gradio as gr
import numpy as np
from PIL import Image
from transformers import AutoProcessor, BlipForConditionalGeneration
import os
# Load the pretrained processor and model
processor = AutoProcessor.from_pretrained("Salesforce/blip-image-captioning-large")
model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-large")

def caption_image(input_image: np.ndarray):
    # Convert numpy array to PIL Image and convert to RGB
    raw_image = Image.fromarray(input_image).convert('RGB')

    # Process the image
    inputs = processor(raw_image, return_tensors="pt")


    # Generate a caption for the image
    out = model.generate(**inputs,max_length=50)

    # Decode the generated tokens to text
    caption = processor.decode(out[0], skip_special_tokens=True)

    return caption

# Save the data to the Hugging Face dataset
HF_TOKEN = os.getenv("HF_TOKEN")
hf_writer = gr.HuggingFaceDatasetSaver(HF_TOKEN, "crowdsourced-images-data")


# Define examples
examples = [
    ["1.jpg"], 
    ["2.jpg"], 
    ["3.jpg"],
    ["4.jpg"],
]



# Create a Gradio interface

iface = gr.Interface(
    fn=caption_image, 
    inputs=gr.Image(), 
    outputs=gr.Textbox(label="Generated Caption", lines=2),
    title="🔍 Image Caption Generator 🖼️",
    description = "Generate stunning captions for your images with our AI-powered model! 🌟\n\n🚫📚 Note: Please avoid entering any sensitive or personal information, as inputs may be reviewed or used for training purposes.",
    allow_flagging="auto",
    flagging_callback=hf_writer,
    examples=examples,
    
)

iface.launch()