Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/nemmusu/gpt-api
The GPT class in this Python script is designed to utilize OpenAI's GPT language model to generate responses based on the provided questions.
https://github.com/nemmusu/gpt-api
api command-line gpt interface openai pyqt5 python python3
Last synced: about 18 hours ago
JSON representation
The GPT class in this Python script is designed to utilize OpenAI's GPT language model to generate responses based on the provided questions.
- Host: GitHub
- URL: https://github.com/nemmusu/gpt-api
- Owner: nemmusu
- Created: 2024-01-01T01:19:15.000Z (11 months ago)
- Default Branch: main
- Last Pushed: 2024-06-08T19:38:21.000Z (5 months ago)
- Last Synced: 2024-06-09T20:06:08.666Z (5 months ago)
- Topics: api, command-line, gpt, interface, openai, pyqt5, python, python3
- Language: Python
- Homepage:
- Size: 50 MB
- Stars: 2
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Readme
[README: Italiano](./README_IT.md)
[README: English](./README.md)
# GPT Chat Bot
This project provides a Python script to interface with the OpenAI GPT language model and a GUI to use it.
## Table of Contents
1. [Readme](#readme)
- [README: Italiano](./README_IT.md)
- [README: English](./README.md)
2. [Installation](#installation)
3. [Usage](#usage)
- [Configuration File](#configuration-file)
- [Using the GPT Class](#using-the-gpt-class)
- [gpt.py Script](#gptpy-script)
4. [Interface](#interface)
5. [Creating the Executable](#creating-the-executable)
- [PyInstaller](#pyinstaller)
- [cx_Freeze](#cxfreeze)
6. [Screenshots](#screenshots)
## Installation
To install the project, follow these steps:
1. Download the project from the GitHub repository. Open the terminal and navigate to the folder where you want to download the project. You can do this using the `cd` command followed by the folder path.
```shell
cd /path/to/folder
```
2. Clone the repository using the `git clone` command followed by the repository URL:
```shell
git clone https://github.com/nemmusu/gpt-api.git
```
3. Access the project folder:
```shell
cd gpt-api
```
4. Install the required `openai` library for this project:
```shell
pip install openai
```
### Configuration File
The `config.ini` file is used to configure various aspects of the chat bot. Here's an example of how it could be configured:
```ini
[OPENAI]
api_key = your_api_key
question = What do you recommend doing today?
temperature = 0.7
model = gpt-3.5-turbo
system_role_message = You are an expert assistant and will answer my questions in a specific and detailed manner exclusively in markdown format.
```
### Using the GPT Class
To use the GPT class, instantiate a GPT object and call the `get_chat_response()` method. All parameters are optional and if not specified, will be retrieved from the configuration file.
Here's an example usage without parameters:
```python
from gpt import GPT
chatbot = GPT() # Uses default values from the configuration file
response = chatbot.get_chat_response()
print(response)
```
And an example usage with parameters:
```python
from gpt import GPT
chatbot = GPT(config_file='/etc/gpt-api/config.ini', api_key='your_api_key', question='What is the meaning of life?',
temperature=0.5, model='gpt-4',
system_role_message='You are an expert assistant and will answer my questions in a specific and detailed manner exclusively in markdown format.')
response = chatbot.get_chat_response()
print(response)
```
The available parameters are:
- `config_file`: The path to the configuration file (default: `config.ini`)
- `api_key`: The OpenAI API key
- `question`: The question to pass to the GPT model
- `temperature`: The temperature for the GPT model
- `model`: The GPT model to use
- `system_role_message`: The system role message for the GPT model
### gpt.py Script
The `gpt.py` script can be run standalone or imported as a module. All parameters are optional and if not specified, will be retrieved from the configuration file.
You can run it using the following command:
```shell
python gpt.py --question "What do you recommend doing today?" --temperature 0.7 --model gpt-3.5-turbo
```
The available parameters are:
- `--config`: The path to the configuration file (default: `config.ini`).
- `--api_key`: The API key for OpenAI.
- `--question`: The question to pass to the GPT model.
- `--temperature`: The temperature for the GPT model.
- `--model`: The GPT model to use.
- `--system_role_message`: The system role message for the GPT model.
To view the command's help, run:
```shell
python gpt.py --help
```
## Interface
The interface implemented in `gui.pyw` uses PyQt5 and other modules. To install, run the following command:
```shell
pip install pyqt5 markdown bleach pygments openai
```
To run the GUI interface, execute the following command:
```shell
python gui.pyw
```
### PyInstaller
To create an executable using PyInstaller, you need to install PyInstaller on your system. You can do this using the following command:
```
pip install pyinstaller
```
Once PyInstaller is installed, you can create the executable using the following command:
```
pyinstaller --onefile --name "GUI_ChatGPT" gui.pyw --hidden-import=gpt --hidden-import=argparse --hidden-import=configparser --hidden-import=openai --hidden-import=PyQt5.QtCore --hidden-import=PyQt5.QtWidgets
```
This command creates a single executable file named "GUI_ChatGPT" using "gui.pyw" as the entry point. Hidden dependencies such as "gpt", "argparse", "configparser", "openai", "PyQt5.QtCore", and "PyQt5.QtWidgets" are also specified to be included in the executable.
### cx_Freeze
To create an executable using cx_Freeze, you need to install cx_Freeze on your system. You can do this using the following command:
```
pip install cx_Freeze
```
Once cx_Freeze is installed, you can create the executable using the following command:
```
python compile_cx_freeze.py build
```
This command will create a structured package with folders containing the executable and the necessary libraries.
