Read more

This article will introduce you to how to build a basic GPT chatbot using Python, leveraging Hyperdiv and OpenAI's GPT models. Even beginners can follow the steps below to create an interactive chatbot.

Preparation

Before starting, please ensure that Python 3.9 or higher is installed on your machine. You will also need to install the necessary Python packages, which can be done by running the pip install -r requirements.txt command in the terminal. This command installs all the dependencies listed in the requirements.txt file, which can be found in the project's GitHub repository.

Setting Up the Environment

First, you need an OpenAI API key to interact with GPT models. You can obtain this key by creating an account on the OpenAI platform and generating a new API key from the API section. Once you have the key, you can set it as an environment variable (OPENAI_API_KEY), or directly replace <your OpenAI API key if not set as env var> in the code with your actual API key.

Understanding the Code

Importing Libraries

The script begins by importing necessary Python libraries:

• sys and os for system-level operations.

• openai for interacting with the OpenAI API.

• hyperdiv as hd, for creating UI components for our chatbot.

Setting the API Key

The API key is retrieved from environment variables, allowing the script to use OpenAI's services for authentication.

add_message Function

This function is crucial for maintaining the chat history. It adds a new message to the state, assigning it a role (for example, “user” or “assistant”), content, an ID, and the model used.

request Function

This function sends the collected messages as prompts to the OpenAI API, using the specified GPT model to generate a response. The response is then added to the chat history.

Rendering Messages

The render_user_message function creates a UI component for user messages, including the message content and the GPT model used.

main Function

The main function is where the chatbot's UI is assembled using Hyperdiv components. It includes:

• A loop to render messages.

• An input form for users to input their messages.

• A dropdown menu to select the GPT model.

• A button to start over and reset the chat history.

Detailed Code Introduction

First, we need to import necessary libraries, including sys, os, openai, and hyperdiv. Then, we retrieve the OpenAI API key from the environment variables.

import sys
import os
import openai
import hyperdiv as hd

api_key=os.environ.get("OPENAI_API_KEY","<your OpenAI API key if not set as env var>")

Next, we define several functions, including add_message, request, and render_user_message. These functions are used to add messages to the chat history, send requests to the OpenAI API and obtain replies, and render the user's messages, respectively.

def add_message(role, content, state, gpt_model):
    ...

def request(gpt_model, state):
    ...

def render_user_message(content, gpt_model):
    ...

In the main function, we create a state object to store the chat history, the current reply, the GPT model used, and the message ID. We also create a task object to manage the running state of the chatbot.

def main():
    state = hd.state(messages=(), current_reply="", gpt_model="gpt-4", message_id=0)
    task = hd.task()
    ...

In the user interface, we first render the chat history, then an input form where users can enter their messages and select the GPT model to use. When the form is submitted, we add the user's message to the chat history and rerun the task to get a new reply.

    with hd.box(align="center", gap=1.5):
        ...
        with hd.form(direction="horizontal", width="100%") as form:
            ...
            if form.submitted:
                add_message("user", prompt.value, state, model.value)
                prompt.reset()
                task.rerun(request, model.value, state)

Finally, we use hd.run(main) to start the chatbot.

Running the Chatbot

Simply executing the script runs the chatbot. The Hyperdiv framework will render a user interface where you can interact with the chatbot. Enter your message, select the GPT model you wish to use, and press enter to see the chatbot's response.

This introduction provides a basic understanding of how to create a GPT-based chatbot using Python. By exploring each function and component, you can customize and extend your chatbot according to your needs.

The code: GPTChat