OpenAI Python

OpenAI GPT integration project using Python.

Features

This project provide the below features:

• OpenAI Chat
  • "Given a list of messages comprising a conversation, the model will return a response."
• OpenAI Exception Handler
  • Consider specific exception scenarios to handle it.
• Token Counter
  • Counts how many tokens exist in the input, estimate the output tokens, and then select a proper model to use it.
• Chatbot UI
  • Frontend application in HTML, CSS and Javascript to interaction with the solution.
• OpenAI Assistants
  • "The Assistants API allows you to build AI assistants within your own applications. An Assistant has instructions and can leverage models, tools, and knowledge to respond to user queries."
  • Reference: https://platform.openai.com/docs/assistants/overview
• Thread History
  • Assistant will consider or not the previous conversation in the future responses.
• Knowledge Retrieval
  • "Retrieval augments the Assistant with knowledge from outside its model, such as proprietary product information or documents provided by your users. Once a file is uploaded and passed to the Assistant, OpenAI will automatically chunk your documents, index and store the embeddings, and implement vector search to retrieve relevant content to answer user queries."
  • Reference: https://platform.openai.com/docs/assistants/tools/knowledge-retrieval
• Function Calling
  • "Similar to the Chat Completions API, the Assistants API supports function calling. Function calling allows you to describe functions to the Assistants and have it intelligently return the functions that need to be called along with their arguments. The Assistants API will pause execution during a Run when it invokes functions, and you can supply the results of the function call back to continue the Run execution."
  • Reference: https://platform.openai.com/docs/assistants/tools/function-calling

Installation

Download the project using the below GIT command:

$ git clone https://github.com/diogoaltoe/openai-python.git

Activate the virtual environment (venv)

Using the Terminal, execute the command below.

On Windows:

.venv\Scripts\activate

On Unix or MacOS:

source .venv/bin/activate

Install project dependencies

Use pip to install the project dependencies listed in the requirements.txt file:

pip install -r requirements.txt

Publish static assets

To publish the static assets, execute the command below:

python manage.py collectstatic 

Usage

Environment Variables

Create inside root folder a .env file with the below variables:

SECRET_KEY=<your_django_key>
OPENAI_API_KEY=<your_openai_api_key>
OPENAI_API_ASSISTANT_ID=<your_openai_api_assistant_id>

OpenAI API

To use OpenAI API, you need to:

• Go to OpenAI Platform Playground (https://platform.openai.com);
• In the left menu, click on API keys option;
• Click on + Create new secret key button;
• Fill the Name field and click on Create secret key button;
• Copy the key and set the OPENAI_API_KEY environment variable inside the created .env file (see Environment Variables).

OpenAI Assistants

To use Assistants feature, you need to:

• Go to OpenAI Platform Playground (https://platform.openai.com/playground);
• Select Assistants option;
• Create a new assistant;
• In Name field, give a proper name for it;
• In Instructions field, add these instructions:

You are an e-commerce customer service chatbot.
You should not answer questions that are not related to e-commerce!

• Copy the Assistant ID that is below to the Name field and set the OPENAI_API_ASSISTANT_ID environment variable inside the created .env file (see Environment Variables).

Knowledge Retrieval

To work with Retrieval feature, you need to:

• Go to OpenAI Platform Playground (https://platform.openai.com/playground);
• Select Assistants option;
• In TOOLS option:
  • Enable Retrieval;
  • Upload the 2 files existing inside chatbot/data folder:
    • info.md
    • policy.md

Functions Calling

To work with Functions feature, you need to:

• Go to OpenAI Platform Playground (https://platform.openai.com/playground);
• Select Assistants option;
• In TOOLS option:
  • In Functions, click in +Function button;
  • Open the validate_promotional_code.json file inside chatbot/data, copy the content and paste it in the function field;
  • Then, click on Save button;
  • It should save the new function and show below Functions the validate_promotional_code function name.

Run Project

Run the project using the command below:

python manage.py runserver

Then access the application, clicking in the link: http://127.0.0.1:8000/

Using EcoMart Chat

In the top bar of EcoMart Chat, in the center, you have the option of using the bot in Chat or Assistant mode.

Still at the top right, you have the option to clear the conversation history (Trash icon). If you are using the bot in Assistant mode, it will lose the past reference of the conversation.

In the bottom bar, you can enter your message and click the Send button to send the message.

Example using Chat Mode

Example using Assistant Mode with Knowledge Retrieval

Example using Assistant Mode with Knowledge Retrieval and Thread History

Example using Assistant Mode with Function Calling

Libraries

• Django
• python-dotenv
• OpenAI Python API library
• tiktoken
• Font Awesome
• Bootstrap

For more all the libraries, check requirements.txt file.

License

This project is licensed under the MIT License.