Gemini Explorer

Google Gemini, formerly known as Bard, is an artificial intelligence (AI) chatbot platform that uses machine learning and natural language processing (NLP) to mimic human conversations.

Mission Scenario

Create a user-friendly chat interface using Streamlit that connects with Google's state-of-the-art large language model, Gemini. The goal is to provide an accessible platform for exploring and showcasing the capabilities of advanced language models. This project aims to serve as an educational and practical introduction to integrating large language models with intuitive interfaces.

Mission Workflow:

• Task 1: 🌐 Enable Google Cloud
• Task 2: 🧬 Google Cloud Initialization
• Task 3: ☁️ Setting up Google Gemini
• Task 4: 📊 Streamlit Integration
• Task 5: 🗣️ Adding Initial System Messages
• Task 6: 📄 Preparing Submission

Requirments:

• Python version 3.11x or above Python
• Streamlit Streamlit Documentation
• Gcloud account
• Vertexai Vertexai Documentation

Task 1: 🌐 Enabling Google Cloud

• Go to the Google Cloud Platform and Select "Get Started for free".
• Sign in using your Google Account and then provide the necessary details and complete the billing requirments.
• Accept the terms and conditions.
• Complete the payment process to initialize your Google Cloud Account.
• Create a new project (for instance "RadicalX - Gemini Explorer").
• Access the Google Cloud Console.
• Navigation -> Artificial Intelligence -> Vertex AI -> Enable All Recommended APIs

Task 2: 🧬 Google Cloud Initialization

• Install the Google SDK using this Link.
• Run the following command to initialize the SDK:

  gcloud init
  

• Sign in using your Google Account credentials.
• Select an existing project or Create a new project
• Set default compute region and zone (Optional Step)

Task 3: ☁️ Setting up Google Gemini

• Install the streamlit framework

  pip install streamlit
  

• Refer the Streamlit Documentation to see an Implementation using OpenAI's ChatGPT
• In the project, we are using Google's Gemini Pro LLM.
• Use the project ID instead of the project name, like this: project = "project_id". This helps avoid encountering a 403 permission denied error.

Task 4: 📊 Streamlit Integration

• Implement the steps given in the mission.
• Run the python file streamlit run filename.py.

Task 5: 🗣️ Adding Initial System Messages

Task 6: 📄 Preparing Submission

• A GitHub repository for the project containing all the project files.
• Loom Video representing the overall approach. Loom Link

Issues Faced

• ⚠️ Issue 403: Permission Denied Error: Check if you have provided project_id rather than project name in your code or if the service account is activated.

Acknowledgements

Special thanks to Talha Sabri and Mikhail Ocampo for extending to me the opportunity to embark on and complete this AI Mission.