简述网站主要流程站长统计工具

Django AI 聊天机器人项目：基于 ChatGPT 的 Django REST API

本文档将介绍如何使用 Django 和 Django REST Framework 构建一个 AI 聊天机器人项目，并结合 OpenAI 的 GPT 模型提供代码解释服务。步骤包括创建 Django 项目、配置 API、与 OpenAI 集成，并最终提供一个可通过 REST API 调用的服务。

---

项目结构

drf_chatgpt/         # Django 项目目录├── api/             # API 应用目录├── src/             # Django 配置文件目录├── manage.py        # Django 管理工具├── .env             # 环境变量文件├── requirements.txt # 项目依赖

---

步骤 1：创建虚拟环境

首先，创建一个虚拟环境来隔离项目的依赖：

python -m venv venv

激活虚拟环境：

• Windows:

  venv\Scripts\activate
  

• Linux / MacOS:

  source venv/bin/activate
  

步骤 2：安装依赖

安装 Django、Django REST Framework 和 OpenAI SDK：

pip install django djangorestframework openai

生成 requirements.txt 以便日后使用：

pip freeze > requirements.txt

---

步骤 3：创建 Django 项目

使用 django-admin 命令创建一个名为 src 的 Django 项目：

django-admin startproject src .

修改 src/settings.py

在 INSTALLED_APPS 中添加所需的应用：

INSTALLED_APPS = [# external apps'rest_framework','rest_framework.authtoken',# internal apps'api',# default apps'django.contrib.admin','django.contrib.auth','django.contrib.contenttypes','django.contrib.sessions','django.contrib.messages','django.contrib.staticfiles',
]

---

步骤 4：创建数据库并运行开发服务器

1. 创建数据库迁移并应用：

   python manage.py migrate
   

2. 创建超级用户以便访问 Django 管理后台：

   python manage.py createsuperuser
   

3. 启动开发服务器：

   python manage.py runserver
   

---

步骤 5：获取 OpenAI API Key

使用环境变量来存储 OpenAI API 密钥。首先，创建一个 .env 文件并将 API 密钥添加到其中：

# .env 文件
OPENAI_APIKEY="sk-Wxxxxxxxx"

在 settings.py 中加载该变量：

import os
from dotenv import load_dotenv
load_dotenv()APIKEY = os.getenv("OPENAI_APIKEY")

---

步骤 6：创建 API 应用

创建一个新的 Django 应用 api：

python manage.py startapp api

设置 API 请求到 OpenAI 的工具函数

在 api/utils.py 中创建与 OpenAI API 的集成函数：

import openai
from django.conf import settingsopenai.api_key = settings.APIKEYdef send_code_to_api(code):try:res = openai.ChatCompletion.create(model="gpt-3.5-turbo",messages=[{"role": "user", "content": f"Tell me what language is this code written? {code}"},{"role": "system", "content": "You are a helpful assistant that provides code explanations."},{"role": "assistant", "content": "Sure! Please provide the code you want me to explain."}],)return res["choices"][0]["message"]["content"]except openai.error.APIError as e:raise ValueError(f"OpenAI API returned an API Error: {e}")except openai.error.APIConnectionError as e:raise ValueError(f"Failed to connect to OpenAI API: {e}")except openai.error.RateLimitError as e:raise ValueError(f"OpenAI API request exceeded rate limit: {e}")

---

步骤 7：创建模型

在 api/models.py 中定义一个用于存储代码及其解释的模型：

from django.db import modelsclass CodeExplainer(models.Model):_input = models.TextField()_output = models.TextField()class Meta:db_table = "t_code_explainer"

创建并迁移模型：

python manage.py makemigrations api
python manage.py migrate api

---

步骤 8：配置 URL 和视图

1. 在项目的 urls.py 中包含 api.urls：

# src/urls.py
from django.contrib import admin
from django.urls import path, includeurlpatterns = [path('admin/', admin.site.urls),path('api/v1/', include('api.urls'))
]

2. 在 api/urls.py 中定义 API 路由：

# api/urls.py
from django.urls import path
from api.views import UserView, TokenView, CodeExplainViewurlpatterns = [path('users/', UserView.as_view(), name='users'),path('tokens/', TokenView.as_view(), name='tokens'),path('code-explain/', CodeExplainView.as_view(), name='code-explain')
]

3. 在 api/views.py 中实现视图逻辑：

# api/views.py
from rest_framework import views, status
from rest_framework.response import Response
from rest_framework.authentication import TokenAuthentication
from rest_framework.permissions import AllowAny
from api.serializers import CodeExplainSerializer, UserSerializer, TokenSerializer
from api.models import CodeExplainerclass CodeExplainView(views.APIView):serializer_class = CodeExplainSerializerauthentication_classes = [TokenAuthentication]def get(self, request, format=None):qs = CodeExplainer.objects.all()serializer = self.serializer_class(qs, many=True)return Response(serializer.data)def post(self, request, format=None):serializer = self.serializer_class(data=request.data)if serializer.is_valid():serializer.save()return Response(serializer.data, status=status.HTTP_201_CREATED)return Response(serializer.errors, status=status.HTTP_400_BAD_REQUEST)# 用户和 token 视图 (稍后定义)
class UserView(views.APIView):serializer_class = UserSerializerpermission_classes = [AllowAny]class TokenView(ObtainAuthToken):serializer_class = TokenSerializer

---

步骤 9：实现序列化器

在 api/serializers.py 中编写模型序列化器：

# api/serializers.py
from rest_framework import serializers
from django.contrib.auth.models import User
from rest_framework.authtoken.models import Token
from api.models import CodeExplainer
from api.utils import send_code_to_apiclass CodeExplainSerializer(serializers.ModelSerializer):class Meta:model = CodeExplainerfields = ("id", "_input", "_output")extra_kwargs = {"_output": {"read_only": True}}def create(self, validated_data):code_explainer = CodeExplainer(**validated_data)_output = send_code_to_api(validated_data["_input"])code_explainer._output = _outputcode_explainer.save()return code_explainerclass UserSerializer(serializers.ModelSerializer):class Meta:model = Userfields = ("id", "username", "email", "password")extra_kwargs = {"password": {"write_only": True}}def create(self, validated_data):password = validated_data.pop("password")user = User.objects.create(**validated_data)user.set_password(password)user.save()Token.objects.create(user=user)return userclass TokenSerializer(serializers.Serializer):username = serializers.CharField()password = serializers.CharField(style={"input_type": "password"}, trim_whitespace=False)def validate(self, attrs):username = attrs.get("username")password = attrs.get("password")user = authenticate(request=self.context.get("request"), username=username, password=password)if not user:msg = "Credentials are not provided correctly..."raise serializers.ValidationError(msg, code="authentication")attrs["user"] = userreturn attrs