Creating Your Account

To utilize ChatGPT, the first step is to create an account. You'll need to fill out some basic information and agree to certain terms. Since this is a complimentary service, you must consent to some data sharing. However, this is a small trade-off for such a powerful tool.

After completing the registration process and logging in, you will be directed to a new chat session:

Feel free to experiment by asking various questions to familiarize yourself with its functionality. For example, you might request a joke:

Now that we’ve had our fun, the next step is to acquire an API key to integrate ChatGPT into Python.

Obtaining an API Key

Having explored the ChatGPT web interface, it's time to obtain an API key. Navigate to your OpenAI API account management page to generate one.

Click on "Create new secret key" to generate your API key. Make sure to copy and securely save this key, as it will be necessary for your Python projects. Keep it confidential to protect your account.

Setting Up Python

With an account and API key in hand, we can now proceed to use Python with ChatGPT. To begin, you'll need to install a couple of pip packages:

• python-dotenv
• openai

The python-dotenv module allows you to load your API key from a .env file, while the openai module provides the interface to ChatGPT.

If you're not familiar with dotenv, it's a straightforward way to manage environment variables in Python. This enables you to keep sensitive information outside of your source code, securely stored in .env files.

Once you have these modules installed, create a new Python file and include the following code:

from os import getenv

import dotenv

import openai

dotenv.load_dotenv()

openai.api_key = getenv('OPENAI_API_KEY', None)

This code imports the necessary modules, loads the .env file, and assigns the API key for OpenAI. Your .env file should simply contain the following line:

OPENAI_API_KEY='123456abcdefg'

Now you’re prepared to ask ChatGPT some intriguing questions!

Asking Your First Question

Are you ready to engage with one of the most powerful AI models? Add this code snippet to your script right after the previous one:

model_type = 'gpt-3.5-turbo'

question = 'tell me a joke'

system_role = {"role": "system", "content": "You are a funny comedian"}

user_role = {"role": "user", "content": question}

res = openai.ChatCompletion.create(

model=model_type,

messages=[system_role, user_role]

)

print(res)

Let’s break down the components of this code:

• Model Type: This specifies which version of ChatGPT to use. Generally, gpt-3.5-turbo is a suitable choice for most applications.
• Question: This is the inquiry you want to pose. Make it engaging!
• System Role: This provides context about how ChatGPT should respond.
• User Role: This holds the question data.

Finally, the openai.ChatCompletion.create function processes the request with the provided variables and prints the result.

Although this is a simple example of a question-and-answer exchange, you can easily expand it by adding multiple messages to create a more complex dialogue.

Now, let’s modify our print statement to retrieve a cleaner response:

print(res['choices'][0]['message']['content'])

Your output should now resemble the following:

Why did the coffee file a police report?

It got mugged!

Pure comedic brilliance!