Building a Simple Q&A App with Gemini Pro and Python

Step 1

Before everything, you need to have API key fto use Gemini models. If you dont know how to create one, reffer this blog post.

Everything About Gemini API and How to Get Your Key

Google’s Gemini is a powerful new AI model family. Let’s dive into what makes it special and how you can start using it. What is Gemini? Gemini is a versatile model family from Google. Each Gemini model is built for specific use cases, making it adaptable for various tasks from

FutureWiseTechChanaka Prasanna

Step 2

Create a virtual environment Using a virtual environment helps keep your project dependencies separate. Here are two ways to do it:

1. With conda:

conda create --name gemini-env python=3.9 # Create environment
conda activate gemini-env # Activate the environment

2. With Python:

python -m venv gemini-env  # Create environment
source gemini-env/bin/activate  # Activate the environment

# On Windows, use: gemini-env\Scripts\activate

Step 3

Create a file named 'requirements.txt' with these packages

streamlit
google-generativeai
python-dotenv

Why use these packages?

1. Streamlit - Creates web apps easily with Python. It's great for quick prototypes and data apps.
2. google-generativeai - Official library for using Google's AI models, including Gemini Pro.
3. python-dotenv - Loads environment variables from a file. This keeps sensitive info (like API keys) separate from your code.

Step 4

Run this command to install the required packages

pip install -r requirements.txt

Step 5

Make a file named .env and add your API key

GOOGLE_API_KEY="YOUR_API_KEY_HERE"

This keeps your API key secure and separate from your code.

Step 6

Create a new Python file (e.g., 'app.py') and add these imports

from dotenv import load_dotenv
load_dotenv()
import streamlit as st
import os
import google.generativeai as genai

Here's what each import does:

• dotenv - Loads environment variables from .env file
• streamlit - Creates the web interface
• os - Accesses environment variables
• google.generativeai - Interacts with the Gemini models

Step 7

Choose your model.To see available models:

for x in genai.list_models():
    print(x.name)

We'll use gemini-pro for this project.

Step 8

Configure API key Set up the API key for use

genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))

This line gets the API key from the environment variable and configures the genai library.

Step 9

Load the model Initialize the Gemini Pro model:

model = genai.GenerativeModel('gemini-pro')

This creates an instance of the Gemini Pro model for generating responses.

Step 10

Create a function to get responses Define a function to interact with the model:

def get_gemini_response(question):
    response = model.generate_content(question)
    return response.text

This function takes a question, sends it to the model, and returns the generated response.

Step 11

Set up the Streamlit app Create the user interface

st.set_page_config(page_title='MYGPT')
st.header("Simple Gemini Application")
input = st.text_input("Write me something: ", key='input')
submit = st.button("Get Answer")

This code:

• Sets the page title
• Adds a header
• Creates a text input field
• Adds a submit button

Step 12

Handle response and display Process the user's input and show the result:

if submit:
    with st.spinner('Generating response...'):
        response = get_gemini_response(input)
    st.write(response)

This code

• Checks if the submit button is clicked
• Shows a loading spinner while generating the response
• Displays the model's response

That's it! You now have a simple Q&A application using the Gemini Pro model. To run it, use the command.

streamlit run app.py

This will start a local web server and open the app in your browser. You can type questions or prompts, and the Gemini Pro model will generate responses.

Watch this video for a demo

Remember to keep your API key secret and never share it publicly.

Happy coding!