app.py

import streamlit as st
import requests
import base64
from PIL import Image
import io

class ChatBot:
    def __init__(self, image_base64, api_key):
        self.image_base64 = image_base64
        self.api_key = api_key
        self.contents = [
            {
                "parts": [
                    {"text": "You are a helpful assistant."},
                    {"inline_data": {"mime_type": "image/jpeg", "data": image_base64}},
                ]
            }
        ]

    def generate_response(self, prompt: str) -> str:
        self.contents.append(
            {
                "parts": [
                    {"text": prompt},
                    {"inline_data": {"mime_type": "image/jpeg", "data": self.image_base64}},
                ]
            }
        )
        headers = {
            "Content-Type": "application/json",
        }
        data = {"contents": self.contents}
        response = requests.post(
            f"https://generativelanguage.googleapis.com/v1beta/models/gemini-pro-vision:generateContent?key={self.api_key}",
            headers=headers,
            json=data,
        )
        response_json = response.json()
        response_text = response_json.get("contents", [{}])[0].get("parts", [{}])[0].get("text", "")
        return response_text

    def get_history(self) -> list:
        return self.contents

# Streamlit app
st.title("Image Chatbot with Google's Gemini API")

st.sidebar.title("Upload Image or Take a Picture")
uploaded_file = st.sidebar.file_uploader("Choose an image...", type=["jpg", "jpeg", "png"])

if uploaded_file is not None:
    image = Image.open(uploaded_file)
    st.sidebar.image(image, caption='Uploaded Image.', use_column_width=True)
    buffered = io.BytesIO()
    image.save(buffered, format="JPEG")
    image_base64 = base64.b64encode(buffered.getvalue()).decode()
    
    api_key = st.sidebar.text_input("Enter your API key", type="password")

    if api_key:
        chatbot = ChatBot(image_base64, api_key)
        
        st.header("Chat with the Bot")
        
        if 'conversation' not in st.session_state:
            st.session_state.conversation = []
        
        user_input = st.text_input("Ask a question about the image:")
        
        if user_input:
            response = chatbot.generate_response(user_input)
            st.session_state.conversation.append({"user": user_input, "bot": response})
        
        if st.session_state.conversation:
            for chat in st.session_state.conversation:
                st.write(f"**You:** {chat['user']}")
                st.write(f"**Bot:** {chat['bot']}")

else:
    st.sidebar.text("Please upload an image to start the conversation.")