DEV Community

Cover image for Creating a Generative AI Chatbot with Python
Juan Brendon Luna Juarez
Juan Brendon Luna Juarez

Posted on

Creating a Generative AI Chatbot with Python

AI models have become incredibly popular in the development of chatbots. With Python, it's easy to create such a chatbot using deep learning techniques and libraries like transformers, TensorFlow, and PyTorch. In this article, weโ€™ll guide you step by step to build a simple generative AI chatbot using Python.

Why we use a AI Chatbot with Python?
Using an AI chatbot with Python offers several advantages, particularly in the areas of simplicity, flexibility, and access to powerful machine learning frameworks.

  • First step to start with ai chatbot development Install the necessary dependencies 
pip install transformers torch
Enter fullscreen mode Exit fullscreen mode

Install virtualenv in console if you donโ€™t have it already:

pip install virtualenv
Enter fullscreen mode Exit fullscreen mode
  • Create a virtual environment 
virtualenv chatbot-env
Enter fullscreen mode Exit fullscreen mode
  • We need Activate the environment. I use Windows 
chatbot-env\Scripts\activate
Enter fullscreen mode Exit fullscreen mode
  • Importing Required Libraries 
from transformers import GPT2LMHeadModel, GPT2Tokenizer
import torch
Enter fullscreen mode Exit fullscreen mode
  • Loading Pre-trained GPT model 
model_name = "gpt2"
model = GPT2LMHeadModel.from_pretrained(model_name)
tokenizer = GPT2Tokenizer.from_pretrained(model_name)

model.eval()
Enter fullscreen mode Exit fullscreen mode
  • Creating the Chatbotโ€™s Response Generator 
def generate_response(user_input):

    inputs = tokenizer.encode(user_input + tokenizer.eos_token, return_tensors="pt")

    outputs = model.generate(inputs, max_length=150, num_return_sequences=1, no_repeat_ngram_size=2, temperature=0.7)

    response = tokenizer.decode(outputs[0], skip_special_tokens=True)

    return response
Enter fullscreen mode Exit fullscreen mode
  • Building the Chatbot Interface.

Finally, weโ€™ll create a simple loop where the user can input text, and the chatbot will generate a response.

print("Hello! I am your chatbot. Type 'quit' to exit.")

while True:
    user_input = input("You: ")

    if user_input.lower() == "quit":
        break

    response = generate_response(user_input)
    print("Chatbot: ", response)
Enter fullscreen mode Exit fullscreen mode
  • Finally Testing the Chatbot!! 
python chatbot.py
Enter fullscreen mode Exit fullscreen mode

Conclusions
We've shown you how to create a simple generative AI chatbot using Python. The chatbot uses a pre-trained model to generate responses based on user input. With some fine-tuning and improvements, you can create a more sophisticated and context-aware chatbot. Chatbots like this one can be used in a variety of applications, such as customer service, education, and entertainment.

Top comments (0)