app.py

import base64
import requests
from io import BytesIO
from PIL import Image
import gradio as gr
from pymongo import MongoClient
import time
import uuid

# MongoDB setup
mongo_client = MongoClient("mongodb+srv://atharva2021:[email protected]/")
db = mongo_client['bajaj']
collection = db['client']

# Function to encode the image to base64
def encode_image(img):
    buffered = BytesIO()
    img.save(buffered, format="PNG")
    encoded_string = base64.b64encode(buffered.getvalue()).decode("utf-8")
    return encoded_string

# Chat function with Pixtral and MongoDB saving
def chat_with_pixtral(uploaded_file, mrn_number, user_question):
    if uploaded_file is not None and mrn_number.strip() != "":
        base64_img = encode_image(uploaded_file)

        api = "https://api.hyperbolic.xyz/v1/chat/completions"
        api_key = "eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.eyJzdWIiOiJyZzMyNzAyNEBnbWFpbC5jb20ifQ._frFve-BYZdb0Qo6FIj6xcDcxpY-6QlC2O-ToQxBjkc"  # Add your API key here

        headers = {
            "Content-Type": "application/json",
            "Authorization": f"Bearer {api_key}",
        }

        payload = {
            "messages": [
                {
                    "role": "user",
                    "content": [
                        {"type": "text", "text": user_question},  
                        {
                            "type": "image_url",
                            "image_url": {"url": f"data:image/jpeg;base64,{base64_img}"},
                        },
                    ],
                }
            ],
            "model": "mistralai/Pixtral-12B-2409",
            "max_tokens": 2048,
            "temperature": 0.7,
            "top_p": 0.9,
        }

        response = requests.post(api, headers=headers, json=payload)

        # Process API response
        if response.status_code == 200:
            response_data = response.json()
            if 'choices' in response_data:
                assistant_response = response_data['choices'][0]['message']['content']
            else:
                assistant_response = "Response format is incorrect"
        else:
            assistant_response = f"API request failed: {response.status_code} - {response.text}"

        # Generate a unique ID for the request
        unique_id = str(uuid.uuid4())

        # Save the result to MongoDB with the specified format
        document = {
            'mrn_number': mrn_number,
            'ocr_result': assistant_response,  # This will be the OCR/API result
            'unique_id': unique_id,
            'got_mode': "plain texts OCR",
            'timestamp': time.time()
        }

        collection.insert_one(document)  # Insert the document into MongoDB

        return assistant_response
    return "Upload your image, enter MRN number, and enter your question."

# Gradio GUI
iface = gr.Interface(
    fn=chat_with_pixtral,
    inputs=[
        gr.Image(type="pil", label="Upload Your Image"),
        gr.Textbox(label="Enter MRN Number"),
        gr.Textbox(label="Please enter your question")
    ],
    outputs="text",
    title="Pixtral Image Chat",
    description="Upload your Image, enter MRN number, and get insights out of the Image"
)

iface.launch(share=True)