Integrating ChatGPT into a Django application allows you to create dynamic and interactive chat interfaces. By following the steps outlined in this article, you can implement ChatGPT in your Django project and provide users with engaging conversational experiences. Experiment with different prompts, improve the user interface and explore further possibilities to enhance the functionality of your chat application.
Steps to Implement ChatGPT in a Django Application using API
Integrating ChatGPT into a Django web application allows you to build interactive chat interfaces and provide dynamic conversational experiences to users. Let’s see the overall steps in summary:
- Setting up the Django Project.
- Define a Django view that handles the chat functionality.
- Configure Django’s URL routing to connect the chat view to a specific URL pattern.
- Designing the User Interface.
- Testing and Running the Application.
Implement ChatGPT in a Django Application
Basic Setup
Install Django using pip, and Create a new Django project. Made changes in setting.py, mention your templates folder in TEMPLATES.
Folder Structure
views.py: Creating the Chat View
The code consists of two functions: query_view and get_completion.
- query_view is a view function that handles the HTTP requests made to the associated endpoint. When a POST request is received, it extracts the prompt from the request data, calls the get_completion function to generate a completion based on the prompt, and returns the completion as a JSON response. For GET requests, it renders the ‘index.html’ template.
- get_completion is a function responsible for generating a completion based on a given prompt using the OpenAI API. It makes a request to the OpenAI API’s Completion endpoint, specifying the prompt, maximum tokens, engine, temperature, and other parameters. It retrieves the generated completion from the API response and returns it as the output of the function.
Python3
from django.shortcuts import render from django.http import JsonResponse import openai openai.api_key = 'YOUR_API_KEY' def get_completion(prompt): print(prompt) query = openai.Completion.create( engine="text-davinci-003", prompt=prompt, max_tokens=1024, n=1, stop=None, temperature=0.5, ) response = query.choices[0].text print(response) return response def query_view(request): if request.method == 'POST': prompt = request.POST.get('prompt') response = get_completion(prompt) return JsonResponse({'response': response}) return render(request, 'index.html') |
urls.py: Setting up URLs
This code sets up two URL patterns: one for the admin site and another for the root URL, and another which is associated with the query_view function in the views module.
Python3
from django.contrib import admin from django.urls import path from . import views urlpatterns = [ path('admin/', admin.site.urls), path('', views.query_view, name='query'), ] |
template/index.py: Creating User Interface
Create an HTML template for the chat interface using HTML, CSS, and Bootstrap, and Set up the necessary JavaScript and AJAX code to handle user interaction.
HTML
<!-- query.html --><html> <head> <title>Query</title> <script src="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0-alpha1/dist/js/bootstrap.bundle.min.js" integrity="sha384-w76AqPfDkMBDXo30jS1Sgez6pr3x5MlQ1ZAGC+nuZB+EYdgRZgiwxhTBTkF7CXvN" crossorigin="anonymous"></script> <link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0-alpha1/dist/css/bootstrap.min.css" rel="stylesheet" integrity="sha384-GLhlTQ8iRABdZLl6O3oVMWSktQOp6b7In1Zl3/Jr59b6EGGoI1aFkw7cmDA6j6gD" crossorigin="anonymous"> <link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/bootstrap-icons@1.10.2/font/bootstrap-icons.css" integrity="sha384-b6lVK+yci+bfDmaY1u0zE8YYJt0TZxLEAFyYSLHId4xoVvsrQu3INevFKo+Xir8e" crossorigin="anonymous"> <script> $(document).ready(function() { // Send the form on enter keypress and avoid if shift is pressed $('#prompt').keypress(function(event) { if (event.keyCode === 13 && !event.shiftKey) { event.preventDefault(); $('form').submit(); } }); $('form').on('submit', function(event) { event.preventDefault(); // get the CSRF token from the cookie var csrftoken = Cookies.get('csrftoken'); // set the CSRF token in the AJAX headers $.ajaxSetup({ headers: { 'X-CSRFToken': csrftoken } }); // Get the prompt var prompt = $('#prompt').val(); var dateTime = new Date(); var time = dateTime.toLocaleTimeString(); // Add the prompt to the response div $('#response').append('<p>('+ time + ') <i class="bi bi-person"></i>: ' + prompt + '</p>'); // Clear the prompt $('#prompt').val(''); $.ajax({ url: '/', type: 'POST', data: {prompt: prompt}, dataType: 'json', success: function(data) { $('#response').append('<p>('+ time + ') <i class="bi bi-robot"></i>: ' + data.response + '</p>'); } }); }); }); </script> </head> <body> <div class="container p-3"> <h3>ChatGPT Clone</h3> <div class="mb-3"> <form method="post"> {% csrf_token %} <label for="prompt" class="form-label"><strong>Prompt: </strong></label> <textarea class="form-control" type="textarea" id="prompt" name="prompt" rows="3"></textarea> <br> <button class="btn btn-primary" type="submit">Submit</button> </form> </div> <br> <div class="mb-3"> <h6>Response:</h6> <div class="container border overflow-auto h-50" id="response"></div> </div> </div> </body> </html> |
Output:
To learn more about Chat GPT, you can refer to:
