A simple web interface for academic research and experimentation using GPT-3.5.

This is the forked version from the project

If you like this project, please give it a star. If you have come up with more useful academic shortcuts, feel free to open an issue or pull request.

Features

• Automatic paper abstract generation based on a provided LaTeX file
• Automatic code summarization and documentation generation
• C++ project header file analysis
• Python project analysis
• Self-code interpretation and dissection
• Experimental function template

• New interface

• All buttons are dynamically generated by reading functional.py, and custom functions can be freely added to free the clipboard

• Code display is also natural https://www.bilibili.com/video/BV1F24y147PD/

• Supports markdown tables generated by GPT

• If the output contains formulas, it will be displayed in both tex and rendering forms at the same time for easy copying and reading

• Too lazy to look at the project code? Just show off the chatgpt mouth

Usage

Prerequisites

• OpenAI API key (can be obtained from here)
• Python 3.9 or higher

Setup

$pip install academic-chatgpt

Run

1. Set your OpenAI API key and other configurations in chataca.toml or ~/.config/chataca/chataca.toml

The configuration file will locate at current working directory or ~/.config/chataca/. The example of chataca.toml

API_KEY = "sk-zH**********************************************"
API_URL = "https://api.openai.com/v1/chat/completions"
USE_PROXY = false
TIMEOUT_SECONDS = 30
WEB_PORT = 8080
MAX_RETRY = 3
LLM_MODEL = "gpt-3.5-turbo"

If you are in China, you need to set up an overseas agent to use the OpenAI API.

2. Start the server: chataca

Experimental features

C++ project header file analysis

In the project path area, enter the project path and click on "[Experimental] Analyze entire C++ project (input the root directory of the project)

LaTeX project abstract generation

In the project path area, enter the project path and click on "[Experimental] Read LaTeX paper and write abstract (input the root directory of the project)

Python project analysis

In the project path area, enter the project path and click on "[Experimental] Analyze entire Python project (input the root directory of the project)"