GPT Trainer API Python Client

A Python client library for interacting with the GPT Trainer API. This library provides an easy-to-use interface for managing chatbots, agents, sessions, messages, and data sources.

---

Features

• Chatbot Management: Create, update, list, and delete chatbots.
• Agent Management: Manage agents associated with chatbots.
• Session Management: Create and manage chatbot sessions.
• Message Management: Retrieve and manage messages within sessions.
• Data Source Management: Upload, update, and delete chatbot data sources.

---

Installation

Install the library via pip:

pip install gpt-trainer-api

Alternatively, if you're developing locally, clone this repository and install it:

git clone https://github.com/Tahir-Siddique/gpt-trainer-api.git
cd gpt-trainer-api
pip install .

---

Usage

Initialize the API

from gpt_trainer.api import GPTTrainerAPI

# Initialize the GPT Trainer API client
api = GPTTrainerAPI(token="<your-api-token>")

---

Chatbot API

List all chatbots

chatbots = api.chatbot.get_chatbots()
for chatbot in chatbots:
    print(chatbot.name, chatbot.uuid)

Create a chatbot

data = {
    "name": "My New Chatbot",
    "visibility": "private",
    "rate_limit": [20, 240],
    "rate_limit_message": "Too many messages. Try again later.",
    "show_citations": True,
}
new_chatbot = api.chatbot.create_chatbot(data)
print(new_chatbot.uuid, new_chatbot.name)

---

Agent API

List all agents for a chatbot

chatbot_uuid = "<chatbot-uuid>"
agents = api.agent.get_agents(chatbot_uuid)
for agent in agents:
    print(agent.name, agent.type)

Create an agent

data = {
    "name": "Agent 1",
    "description": "Handles FAQs",
    "prompt": "Answer all FAQs concisely.",
    "type": "user-facing",
    "enabled": True,
}
new_agent = api.agent.create_agent(chatbot_uuid, data)
print(new_agent.uuid, new_agent.name)

---

Session API

Create a new session

new_session = api.session.create_session(chatbot_uuid)
print(new_session.uuid, new_session.created_at)

List sessions

sessions = api.session.list_sessions(chatbot_uuid)
for session in sessions:
    print(session.uuid, session.meta)

---

Message API

List messages in a session

session_uuid = "<session-uuid>"
messages = api.message.list_messages(session_uuid)
for message in messages:
    print(message.query, message.response)

---

Data Source API

List all data sources for a chatbot

sources = api.data_source.list_sources(chatbot_uuid)
for source in sources:
    print(source.file_name, source.status)

Upload a file as a data source

file_path = "/path/to/your/file.txt"
new_source = api.data_source.upload_file(chatbot_uuid, file_path)
print(new_source.uuid, new_source.title)

---

Requirements

• Python >= 3.7
• requests library

---

Contributing

1. Fork the repository.
2. Create a feature branch.
3. Commit your changes.
4. Push to your branch.
5. Submit a pull request.

---

License

This project is licensed under the MIT License. See the LICENSE file for details.

---

Support

If you encounter any issues or have questions, please open an issue in the GitHub repository.

---

Authors

• Tahir Siddique - Python Developer - Tahir Siddique

Key Sections

1. Features:
   • Highlights what the library can do.
2. Installation:
   • Instructions for installing the package.
3. Usage:
   • Code examples for common operations.
4. Requirements:
   • Dependencies for the library.
5. Contributing:
   • Instructions for contributing to the project.
6. License:
   • Licensing information.
7. Support:
   • Instructions for getting help or reporting issues.
8. Authors:
   • Credits to the creator(s).

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

[0.2.3] - 2025-01-23

Added

• Added test cases and improved flow.

[0.2.2] - 2025-01-23

Added

• Handled 400 bad request to show the respons on error raise.

[0.2.1] - 2025-01-23

Added

• Added stream_message method to MessageAPI:
  • Supports real-time streaming of chatbot responses from the GPT Trainer API.
  • Returns concatenated streamed response as a single string.
• Added full support for managing session messages, including:
  • Listing messages (list_messages).
  • Deleting individual messages (delete_message).
  • Bulk deleting messages (bulk_delete_messages).

Changed

• Improved error handling and documentation for MessageAPI.

---

[0.2.0] - 2025-01-23

Added

• Implemented full support for Data Sources, including:
  • File upload
  • Adding QA pairs
  • Adding URLs
  • Bulk deletion of sources
  • Re-scraping URLs.
• Unified API client GPTTrainerAPI for easy access to all APIs.

Changed

• Improved source_api.py with additional methods for managing data sources.

---

[0.1.0] - 2025-01-20

Added

• Initial release of gpt_trainer_api:
  • Chatbot management.
  • Agent management.
  • Session creation and message handling.
  • Basic data source management.