Google Cloud Vertex AI Chatbot

This repository contains code and instructions on creating a simple chatbot using Langchain w/ Google Cloud's Vertex AI platform and the Gemini 1.5 Pro model.

Prerequisites

• A Google Cloud Platform (GCP) account with billing enabled.
• Python 3.9 or later installed on your local machine.
• A text editor or IDE of your choice.

Setup

1. Create a Google Cloud Project:

• If you don't have an existing project, create one in the Google Cloud Console.

2. Enable the Vertex AI API:

• Navigate to the Vertex AI section in the console and enable the API if it's not already enabled.

3. Create a Service Account:

• Go to the IAM & Admin > Service Accounts section.
• Create a new service account.
• Grant the "Vertex AI User" role to the service account.
• Download the JSON key file for the service account.

4. Set Environment Variables:

• Create a .env file in the root directory of this repository.
• Add the following line, replacing path/to/your/keyfile.json with the actual path to your downloaded key file:

GOOGLE_APPLICATION_CREDENTIALS="path/to/your/keyfile.json"

Installation

1. Clone the Repository:

git clone https://github.com/brainiakk/langchain-gemini-test.git
cd langchain-gemini-test

2. Install Dependencies:

pip install -r requirements.txt

Running the Chatbot

1. Run the Python Script:

python -m main

2. Interact with the Chatbot:

• The script will start a loop and prompt you for input.
• Type your query and press Enter.
• The chatbot will respond using the Gemini Pro model.

Code Explanation

• load_dotenv(): This line loads environment variables from the .env file, including the path to your service account key file.
• llm = ChatVertexAI(...): This line initializes the ChatVertexAI object from the vertexai.preview.language_models library. It specifies the Gemini Pro model and a temperature value for controlling the randomness of the responses.
• response = llm.invoke(text): This line sends the user's input (text) to the Vertex AI API for processing by the Gemini Pro model.
• print(response.content): This line prints the model's response to the console.

Notes

• The code provided is a basic example and can be extended with additional features.
• Refer to the Vertex AI documentation for more advanced usage and customization options.
• Remember to replace path/to/your/keyfile.json with the actual path to your service account key file.

Disclaimer

This code is provided for educational purposes only.